|
|
Linux Advocates Fear Solaris 10.
That is why they post so much FUD on sites such as slashdot.org. The reason
Linux became popular was that they were not as conservative as BSD, and
would add features constantly. That meant if you wanted some enterprise
functionality for free, you'd have to go with linux (SMP, etc.).
The downside of that linux philosophy was that backwards compatiblitiy went
down the drain. glibc interfaces and kernel calls would change
significantly between versions. Sometimes the "enterprise" feature would
be botched, like the scheduler in the 2.4 kernel.
Now that Solaris 10 is going to be open sourced, Linux advocates rightly
fear that Linux will be abandoned in favor of Solaris 10. Similar to how
BSD was left in droves for Linux. That's right, SUN's brilliant strategy of
making Solaris open sourced is using the linux advocates' techniques
against them.
The reason people will move to Solaris in droves is the same that caused
BSD's user flight. It supports more enterprise features. But the
difference this time is that Solaris is cheaper than Redhat, is completely
backwards compatible, and scales much better on many more CPUs.
Now if only SUN would release its compiler under a BSD license....
|
|
0
|
|
|
|
Reply
|
 Mike
|
12/13/2004 5:50:54 AM |
|
Mike Cox wrote:
> Now that Solaris 10 is going to be open sourced, Linux advocates rightly
> fear that Linux will be abandoned in favor of Solaris 10. Similar to how
> BSD was left in droves for Linux. That's right, SUN's brilliant strategy
> of making Solaris open sourced is using the linux advocates' techniques
> against them.
Having used Solaris.
I doubt it.
|
|
|
0
|
|
|
|
Reply
|
John
|
12/13/2004 6:58:59 AM
|
|
begin Mike Cox wrote:
< snip typical M Cox idiocy >
Well, you know, it is called "Slowaris" for a reason
And for that same reason you wish it would replace linux. Then Win would not
stand out in sharp contrast as a buggy, slow POS
--
Microsoft software doesn't get released - it escapes, leaving
a trail of destruction behind it.
|
|
|
0
|
|
|
|
Reply
|
Peter
|
12/13/2004 7:33:27 AM
|
|
Error log for Sun, 12 Dec 2004 21:50:54 -0800: Mike Cox caused a Page
Fault at address <324sb4F3h3ijiU1@individual.net>, details...
> Now that Solaris 10 is going to be open sourced, Linux advocates rightly
> fear that Linux will be abandoned in favor of Solaris 10. Similar to
> how BSD was left in droves for Linux. That's right, SUN's brilliant
> strategy of making Solaris open sourced is using the linux advocates'
> techniques against them.
>
> The reason people will move to Solaris in droves is the same that caused
> BSD's user flight. It supports more enterprise features. But the
> difference this time is that Solaris is cheaper than Redhat, is
> completely backwards compatible, and scales much better on many more
> CPUs.
Though I haven't researched it much, I truly doubt that Sun will release
Solaris 100% Open Source. Perhaps portions of it, but this is still a far
cry from where Linux is today. Sun would have to willing to totally let
go of the reigns, and I don't think they are prepared, or even able, to do
that.
If it does take off and some enterprising fellows decide to release Open
Solaris, I really don't think that it'll catch up to Linux anytime soon,
if it catches on at all.
--
rapskat - 02:47:04 up 21 days, 9:06, 6 users, load average: 0.02, 0.34, 0.37
You single-handedly fought your way into this hopeless mess.
|
|
|
0
|
|
|
|
Reply
|
 rapskat (1101)
|
12/13/2004 7:55:58 AM
|
|
Peter,
Peter K�hlmann wrote:
> begin Mike Cox wrote:
>
> < snip typical M Cox idiocy >
>
> Well, you know, it is called "Slowaris" for a reason
The OP is a troll, don't feed him. Don't belittle yourself by stooping
to his level.
>
> And for that same reason you wish it would replace linux. Then Win would not
> stand out in sharp contrast as a buggy, slow POS
This sentence makes no sense.
MfG,
Nathan Dietsch
|
|
|
0
|
|
|
|
Reply
|
Nathan
|
12/13/2004 8:50:51 AM
|
|
In article <MJcvd.205$Fb3.4757@nnrp1.ozemail.com.au>,
Nathan Dietsch <usenet@ndietsch.com> wrote:
>Peter,
>
>Peter K�hlmann wrote:
>> begin Mike Cox wrote:
>>
>> < snip typical M Cox idiocy >
>>
>> Well, you know, it is called "Slowaris" for a reason
>
>The OP is a troll, don't feed him. Don't belittle yourself by stooping
>to his level.
Peter K�hlmann also seems to be a troll (as he uses the word
"Slowlaris").
___________________
/| /| | |
||__|| | Please do |
/ O O\__ NOT |
/ \ feed the |
/ \ \ trolls |
/ _ \ \ ______________|
/ |\____\ \ ||
/ | | | |\____/ ||
/ \|_|_|/ \ __||
/ / \ |____| ||
/ | | /| | --|
| | |// |____ --|
* _ | |_|_|_| | \-/
*-- _--\ _ \ // |
/ _ \\ _ // | /
* / \_ /- | - | |
* ___ c_c_c_C/ \C_c_c_c____________
--
EMail:joerg@schily.isdn.cs.tu-berlin.de (home) J�rg Schilling D-13353 Berlin
js@cs.tu-berlin.de (uni) If you don't have iso-8859-1
schilling@fokus.fraunhofer.de (work) chars I am J"org Schilling
URL: http://www.fokus.fraunhofer.de/usr/schilling ftp://ftp.berlios.de/pub/schily
|
|
|
0
|
|
|
|
Reply
|
js
|
12/13/2004 10:38:41 AM
|
|
On Sun, 12 Dec 2004 21:50:54 -0800, Mike Cox wrote:
> That is why they post so much FUD on sites such as slashdot.org.
No one is afraid of Solaris
> The reason
> Linux became popular was that they were not as conservative as BSD, and
> would add features constantly. That meant if you wanted some enterprise
> functionality for free, you'd have to go with linux (SMP, etc.).
>
The reason it became popular was it was unencumbered by proprietary before
BSD was.
--
Rick
|
|
|
0
|
|
|
|
Reply
|
 none10 (3395)
|
12/13/2004 10:38:41 AM
|
|
In article <MJcvd.205$Fb3.4757@nnrp1.ozemail.com.au>,
Nathan Dietsch <usenet@ndietsch.com> wrote:
>Peter K�hlmann wrote:
>> begin Mike Cox wrote:
>>
>> < snip typical M Cox idiocy >
>>
>> Well, you know, it is called "Slowaris" for a reason
>
>The OP is a troll, don't feed him. Don't belittle yourself by stooping
>to his level.
Peter K�hlmann uses the word "Slowaris" so he is a troll too :-(
___________________
/| /| | |
||__|| | Please do |
/ O O\__ NOT |
/ \ feed the |
/ \ \ trolls |
/ _ \ \ ______________|
/ |\____\ \ ||
/ | | | |\____/ ||
/ \|_|_|/ \ __||
/ / \ |____| ||
/ | | /| | --|
| | |// |____ --|
* _ | |_|_|_| | \-/
*-- _--\ _ \ // |
/ _ \\ _ // | /
* / \_ /- | - | |
* ___ c_c_c_C/ \C_c_c_c____________
--
EMail:joerg@schily.isdn.cs.tu-berlin.de (home) J�rg Schilling D-13353 Berlin
js@cs.tu-berlin.de (uni) If you don't have iso-8859-1
schilling@fokus.fraunhofer.de (work) chars I am J"org Schilling
URL: http://www.fokus.fraunhofer.de/usr/schilling ftp://ftp.berlios.de/pub/schily
|
|
|
0
|
|
|
|
Reply
|
js
|
12/13/2004 10:40:48 AM
|
|
Mike Cox wrote:
> That is why they post so much FUD on sites such as slashdot.org. The reason
> Linux became popular was that they were not as conservative as BSD, and
> would add features constantly. That meant if you wanted some enterprise
> functionality for free, you'd have to go with linux (SMP, etc.).
>
> The downside of that linux philosophy was that backwards compatiblitiy went
> down the drain. glibc interfaces and kernel calls would change
> significantly between versions. Sometimes the "enterprise" feature would
> be botched, like the scheduler in the 2.4 kernel.
>
> Now that Solaris 10 is going to be open sourced, Linux advocates rightly
> fear that Linux will be abandoned in favor of Solaris 10.
On the contrary. My guess is that Linux will adopt the best features of
Solaris and become an even better OS.
> Similar to how
> BSD was left in droves for Linux. That's right, SUN's brilliant strategy of
> making Solaris open sourced is using the linux advocates' techniques
> against them.
>
> The reason people will move to Solaris in droves is the same that caused
> BSD's user flight. It supports more enterprise features. But the
> difference this time is that Solaris is cheaper than Redhat, is completely
> backwards compatible, and scales much better on many more CPUs.
>
> Now if only SUN would release its compiler under a BSD license....
|
|
|
0
|
|
|
|
Reply
|
Eltee
|
12/13/2004 2:39:16 PM
|
|
Mike Cox wrote:
> That is why they post so much FUD on sites such as slashdot.org. The
reason
> Linux became popular was that they were not as conservative as BSD,
and
> would add features constantly. That meant if you wanted some
enterprise
> functionality for free, you'd have to go with linux (SMP, etc.).
>
> The downside of that linux philosophy was that backwards
compatiblitiy went
> down the drain. glibc interfaces and kernel calls would change
> significantly between versions. Sometimes the "enterprise" feature
would
> be botched, like the scheduler in the 2.4 kernel.
>
> Now that Solaris 10 is going to be open sourced, Linux advocates
rightly
> fear that Linux will be abandoned in favor of Solaris 10. Similar to
how
> BSD was left in droves for Linux. That's right, SUN's brilliant
strategy of
> making Solaris open sourced is using the linux advocates' techniques
> against them.
>
> The reason people will move to Solaris in droves is the same that
caused
> BSD's user flight. It supports more enterprise features. But the
> difference this time is that Solaris is cheaper than Redhat, is
completely
> backwards compatible, and scales much better on many more CPUs.
>
> Now if only SUN would release its compiler under a BSD license....
If Solaris is better than Linux, and it really does become open source,
why would current Linux users have any problem switching to it?
What Eltee wrote is more likely, that Linux will adopt all the best
parts of Solaris, and become an even better OS.
|
|
|
0
|
|
|
|
Reply
|
Edwin
|
12/13/2004 4:03:33 PM
|
|
On Sun, 12 Dec 2004, Mike Cox wrote:
> The reason people will move to Solaris in droves is the same that caused
> BSD's user flight. It supports more enterprise features. But the
> difference this time is that Solaris is cheaper than Redhat, is completely
> backwards compatible, and scales much better on many more CPUs.
I know that MIke Cox is a troll, but I think this is the first sensible
thing he's ever written!
--
Rich Teer, SCNA, SCSA, author of "Solaris Systems Programming"
President,
Rite Online Inc.
Voice: +1 (250) 979-1638
URL: http://www.rite-group.com/rich
|
|
|
0
|
|
|
|
Reply
|
Rich
|
12/13/2004 5:31:01 PM
|
|
On Mon, 13 Dec 2004, Peter [UTF-8] Köhlmann wrote:
> Well, you know, it is called "Slowaris" for a reason
Only by people who are as out of touch as you are. As documented
on this page: http://www.sun.com/2004-1012/feature/ Apache 1.3.29
on Solaris 10 can handle substantially more connections per second
than RHEL 3.0.
--
Rich Teer, SCNA, SCSA, author of "Solaris Systems Programming"
President,
Rite Online Inc.
Voice: +1 (250) 979-1638
URL: http://www.rite-group.com/rich
|
|
|
0
|
|
|
|
Reply
|
Rich
|
12/13/2004 5:36:07 PM
|
|
In article <1102953813.259715.143030@z14g2000cwz.googlegroups.com>,
"Edwin" <thorne25@juno.com> writes:
[...]
> If Solaris is better than Linux, and it really does become open source,
> why would current Linux users have any problem switching to it?
Because they believe in the following illogic:
M$ is evil
M$ is a corporation
corporations are evil (give or take a blind eye to Red Scat)
> What Eltee wrote is more likely, that Linux will adopt all the best
> parts of Solaris, and become an even better OS.
Yes, and I like my pork chops on the wing, too...
--
mailto:rlhamil@smart.net http://www.smart.net/~rlhamil
Lasik/PRK theme music:
"In the Hall of the Mountain King", from "Peer Gynt"
|
|
|
0
|
|
|
|
Reply
|
Richard
|
12/13/2004 5:38:28 PM
|
|
begin Rich Teer wrote:
> On Mon, 13 Dec 2004, Peter [UTF-8] Köhlmann wrote:
>
>> Well, you know, it is called "Slowaris" for a reason
>
> Only by people who are as out of touch as you are. As documented
> on this page: http://www.sun.com/2004-1012/feature/ Apache 1.3.29
> on Solaris 10 can handle substantially more connections per second
> than RHEL 3.0.
>
Oh yes. I see. An independent study from certainly very independent parties
Just like the very independent TCO studies from MS
It certainly serves to remind us that we never should trust statistics we
haven't fabricated ourselves out of whole cloth
--
We may not return the affection of those who like us,
but we always respect their good judgement.
|
|
|
0
|
|
|
|
Reply
|
Peter
|
12/13/2004 5:44:23 PM
|
|
On Mon, 13 Dec 2004, Eltee wrote:
> On the contrary. My guess is that Linux will adopt the best features of
> Solaris and become an even better OS.
Linux will doubtless "borrow" some features from Solaris, but some
of the best features will be left out because Linus won't want them.
I mean, if he won't accept the ability to produce a core dump on
kernel panic, what chance has something like DTrace got?
--
Rich Teer, SCNA, SCSA, author of "Solaris Systems Programming"
President,
Rite Online Inc.
Voice: +1 (250) 979-1638
URL: http://www.rite-group.com/rich
|
|
|
0
|
|
|
|
Reply
|
Rich
|
12/13/2004 5:45:21 PM
|
|
On Mon, 13 Dec 2004, Edwin wrote:
> If Solaris is better than Linux, and it really does become open source,
> why would current Linux users have any problem switching to it?
For many users, the choice of OS is a religeous issue. I mean,
look how many Windoze advocates there are, and I think we can all
agree that their OS of choice is underneath the bottom of the pile
when it comes to OS technology!
> What Eltee wrote is more likely, that Linux will adopt all the best
> parts of Solaris, and become an even better OS.
See my response to Eltee...
--
Rich Teer, SCNA, SCSA, author of "Solaris Systems Programming"
President,
Rite Online Inc.
Voice: +1 (250) 979-1638
URL: http://www.rite-group.com/rich
|
|
|
0
|
|
|
|
Reply
|
Rich
|
12/13/2004 5:46:47 PM
|
|
In article <Pine.SOL.4.58.0412130929150.8903@zaphod>,
Rich Teer <rich.teer@rite-group.com> wrote:
>I know that MIke Cox is a troll, but I think this is the first sensible
>thing he's ever written!
As this is the first senseful reply on his posting, I need to
send you a ;-)
--
EMail:joerg@schily.isdn.cs.tu-berlin.de (home) J�rg Schilling D-13353 Berlin
js@cs.tu-berlin.de (uni) If you don't have iso-8859-1
schilling@fokus.fraunhofer.de (work) chars I am J"org Schilling
URL: http://www.fokus.fraunhofer.de/usr/schilling ftp://ftp.berlios.de/pub/schily
|
|
|
0
|
|
|
|
Reply
|
js
|
12/13/2004 6:25:27 PM
|
|
In article <Pine.SOL.4.58.0412130930180.8903@zaphod>,
Rich Teer <rich.teer@rite-group.com> wrote:
>On Mon, 13 Dec 2004, Peter [UTF-8] Köhlmann wrote:
>
>> Well, you know, it is called "Slowaris" for a reason
>
>Only by people who are as out of touch as you are. As documented
>on this page: http://www.sun.com/2004-1012/feature/ Apache 1.3.29
>on Solaris 10 can handle substantially more connections per second
>than RHEL 3.0.
This is something that is also proven by reallity.
After I did upgrade www.berlios.de from Linus to Solaris, I did
feel that there is a second CPU in the box.
--
EMail:joerg@schily.isdn.cs.tu-berlin.de (home) J�rg Schilling D-13353 Berlin
js@cs.tu-berlin.de (uni) If you don't have iso-8859-1
schilling@fokus.fraunhofer.de (work) chars I am J"org Schilling
URL: http://www.fokus.fraunhofer.de/usr/schilling ftp://ftp.berlios.de/pub/schily
|
|
|
0
|
|
|
|
Reply
|
js
|
12/13/2004 6:27:06 PM
|
|
begin Joerg Schilling wrote:
> In article <Pine.SOL.4.58.0412130930180.8903@zaphod>,
> Rich Teer <rich.teer@rite-group.com> wrote:
>>On Mon, 13 Dec 2004, Peter [UTF-8] Köhlmann wrote:
>>
>>> Well, you know, it is called "Slowaris" for a reason
>>
>>Only by people who are as out of touch as you are. As documented
>>on this page: http://www.sun.com/2004-1012/feature/ Apache 1.3.29
>>on Solaris 10 can handle substantially more connections per second
>>than RHEL 3.0.
>
> This is something that is also proven by reallity.
> After I did upgrade www.berlios.de from Linus to Solaris, I did
> feel that there is a second CPU in the box.
>
I am sure you did
Just as zhose windows guys in here who never ever received a virus
--
"I don't want to belong to any club that would have me as a member."
-- Groucho Marx
|
|
|
0
|
|
|
|
Reply
|
Peter
|
12/13/2004 6:42:37 PM
|
|
In article <cpknp9$653$03$2@news.t-online.com>,
Peter =?UTF-8?B?S8O2aGxtYW5u?= <Peter.Koehlmann@t-online.de> wrote:
>> This is something that is also proven by reallity.
>> After I did upgrade www.berlios.de from Linus to Solaris, I did
>> feel that there is a second CPU in the box.
>>
>
>I am sure you did
>Just as zhose windows guys in here who never ever received a virus
Looks like it makes no sense to treat you like an honest person :-(
The speed difference of administrative scripts that run about one our
cannot be just sweeped away by a religion like the one you seem to be
stuck to.
--
EMail:joerg@schily.isdn.cs.tu-berlin.de (home) J�rg Schilling D-13353 Berlin
js@cs.tu-berlin.de (uni) If you don't have iso-8859-1
schilling@fokus.fraunhofer.de (work) chars I am J"org Schilling
URL: http://www.fokus.fraunhofer.de/usr/schilling ftp://ftp.berlios.de/pub/schily
|
|
|
0
|
|
|
|
Reply
|
js
|
12/13/2004 7:05:06 PM
|
|
["Followup-To:" header set to comp.os.linux.advocacy.]
On Mon, 13 Dec 2004 19:42:37 +0100, Peter K�hlmann wrote:
> begin Joerg Schilling wrote:
>
>> In article <Pine.SOL.4.58.0412130930180.8903@zaphod>,
>> Rich Teer <rich.teer@rite-group.com> wrote:
>>>On Mon, 13 Dec 2004, Peter [UTF-8] Köhlmann wrote:
>>>
>>>> Well, you know, it is called "Slowaris" for a reason
>>>
>>>Only by people who are as out of touch as you are. As documented
>>>on this page: http://www.sun.com/2004-1012/feature/ Apache 1.3.29
>>>on Solaris 10 can handle substantially more connections per second
>>>than RHEL 3.0.
>>
>> This is something that is also proven by reallity.
>> After I did upgrade www.berlios.de from Linus to Solaris, I did
>> feel that there is a second CPU in the box.
>>
>
> I am sure you did
> Just as zhose windows guys in here who never ever received a virus
I've *received* plenty, I'm just not dumb enough to actually be infected.
--
FreeBSD 4.8-RELEASE i386
1:10PM up 20 days, 18:32, 0 users, load averages: 0.00, 0.00, 0.00
|
|
|
0
|
|
|
|
Reply
|
General
|
12/13/2004 7:15:22 PM
|
|
Sun Solaris 10 will win back some converts in places like the database
arena and such. Linux users will win more machines in places were
commodity matters, like server farms and the such.
Solaris 10 has a lot of great points like dtrace and ZFS that make it
the first version of Solaris in many years worth looking at.
Cheers,
Tom
|
|
|
0
|
|
|
|
Reply
|
Thomas
|
12/13/2004 8:57:09 PM
|
|
In article <cpkmtq$2kp$1@news.cs.tu-berlin.de>, Joerg Schilling wrote:
>This is something that is also proven by reallity.
Say what you like, but there's only one 'l' in my version of "reality".
Got the guitar CD ?
--
Elvis Notargiacomo master AT barefaced DOT cheek
http://www.notatla.org.uk/goen/
7.031: OnACPower returned value( 0x1 ) which is Equal To 0x1
|
|
|
0
|
|
|
|
Reply
|
elvis
|
12/13/2004 9:01:01 PM
|
|
General Protection Fault wrote:
<snip>
>
> I've received plenty, I'm just not dumb enough to actually be infected.
Observation here:
Assuming that you consider yourself above "average" and so, don't get
infected, that you recieved plenty of viruses means that there are waaayyy
too many of your "average and below" fellow users getting infected. Your
point?
Responding to the thread:
<conspiracy mode>
Of course, we are now getting a rash of "sunvocates" (Is this TMed already?)
supplementing the winvocates /and/ spouting the official SUN line in COLA,
whereas we hardly heard a peep from the Sun faction previously.
</conspiracy mode>
Gentlemen, please be reminded that Linux != RHEL, so any arguments that the
general business-Linux population would move en-masse to Solaris because of
the price of RHEL support is at best highly speculative.
But the main selling point is still the FREEDOM to do what you want with
Linux anyway. This is still the best insurance, even if RH went belly-up
tomorrow.
(and AFAIK, RH isn't known to be the speediest distro around OOTB)
Cheers,
WS
--
Change to leews to mail.
Linux user #61399
The beginning of the
end
|
|
|
0
|
|
|
|
Reply
|
 see.sig5 (410)
|
12/14/2004 3:03:19 AM
|
|
On Tue, 14 Dec 2004 11:03:19 +0800, Lee Wei Shun wrote:
> General Protection Fault wrote:
>
><snip>
>
>>
>> I've received plenty, I'm just not dumb enough to actually be infected.
>
> Observation here:
>
> Assuming that you consider yourself above "average" and so, don't get
> infected, that you recieved plenty of viruses means that there are waaayyy
> too many of your "average and below" fellow users getting infected. Your
> point?
Kohlmann claims any Windows user who never got a virus is lying. I'm
claiming otherwise.
--
FreeBSD 4.8-RELEASE i386
8:10AM up 21 days, 13:32, 1 user, load averages: 0.18, 0.06, 0.01
|
|
|
0
|
|
|
|
Reply
|
 generalpf (2660)
|
12/14/2004 2:17:01 PM
|
|
begin General Protection Fault wrote:
> On Tue, 14 Dec 2004 11:03:19 +0800, Lee Wei Shun wrote:
>> General Protection Fault wrote:
>>
>><snip>
>>
>>>
>>> I've received plenty, I'm just not dumb enough to actually be infected.
>>
>> Observation here:
>>
>> Assuming that you consider yourself above "average" and so, don't get
>> infected, that you recieved plenty of viruses means that there are
>> waaayyy too many of your "average and below" fellow users getting
>> infected. Your point?
>
> Kohlmann claims any Windows user who never got a virus is lying. I'm
> claiming otherwise.
>
I am claiming that any windows user connected to the internet got a virus
And I see that you are still not even able to write my name correctly. What
other deficiancies do you have, apart from blatant lying?
--
Micro$oft. What's broken today?
|
|
|
0
|
|
|
|
Reply
|
 Peter.Koehlmann (13202)
|
12/14/2004 3:23:21 PM
|
|
On Tue, 14 Dec 2004 16:23:21 +0100, Peter K�hlmann wrote:
> begin General Protection Fault wrote:
>
>> On Tue, 14 Dec 2004 11:03:19 +0800, Lee Wei Shun wrote:
>>> General Protection Fault wrote:
>>>
>>><snip>
>>>
>>>>
>>>> I've received plenty, I'm just not dumb enough to actually be infected.
>>>
>>> Observation here:
>>>
>>> Assuming that you consider yourself above "average" and so, don't get
>>> infected, that you recieved plenty of viruses means that there are
>>> waaayyy too many of your "average and below" fellow users getting
>>> infected. Your point?
>>
>> Kohlmann claims any Windows user who never got a virus is lying. I'm
>> claiming otherwise.
>>
>
> I am claiming that any windows user connected to the internet got a virus
> And I see that you are still not even able to write my name correctly. What
> other deficiancies do you have, apart from blatant lying?
I have 3 Windows machines connected to the internet and non have ever
gotten a virus. Also, I don't respect you enough to spell your name
correctly, Kohlmann.
--
FreeBSD 4.8-RELEASE i386
10:00AM up 21 days, 15:22, 0 users, load averages: 0.00, 0.00, 0.00
|
|
|
0
|
|
|
|
Reply
|
generalpf (2660)
|
12/14/2004 4:05:35 PM
|
|
In comp.os.linux.advocacy Rich Teer <rich.teer@rite-group.com> wrote:
> On Sun, 12 Dec 2004, Mike Cox wrote:
>
>> The reason people will move to Solaris in droves is the same that caused
>> BSD's user flight. It supports more enterprise features. But the
>> difference this time is that Solaris is cheaper than Redhat, is completely
>> backwards compatible, and scales much better on many more CPUs.
>
> I know that MIke Cox is a troll, but I think this is the first sensible
> thing he's ever written!
His point is valid, but I see it more as the reason Sun will stay
relevant on the high end rather than the reason Linux users will
switch. The fact is, most servers are four processors or less,
and Linux scales very nicely at that level. If Sun tries to 'take
back' the commodity server market with open source Solaris, they
have a hard struggle ahead of them. They might have gotten some
traction five years ago... but the market has changed a lot
since then.
Sun's best strategy is to differentiate themselves on hardware,
fold Solaris's high end features into Linux, and save themselves
a bucketload of development cost in the process. I just don't
expect open source Solaris to gain the developer following
that Linux already has, so Sun will likely continue to carry
most of the Solaris development burden. If they go Linux, donate
some Sun big iron to a few universities and prominent kernel
developers, they begin to reap immediate benefits.
--
Thad Phetteplace
http://www.AliensForBush.com - Geeky Political Humor
|
|
|
0
|
|
|
|
Reply
|
thad01
|
12/14/2004 5:53:33 PM
|
|
thad01@dexter.glaci.com wrote:
> In comp.os.linux.advocacy Rich Teer <rich.teer@rite-group.com> wrote:
,,,
>>I know that MIke Cox is a troll, but I think this is the first sensible
>>thing he's ever written!
>
>
> His point is valid, but I see it more as the reason Sun will stay
> relevant on the high end rather than the reason Linux users will
> switch. The fact is, most servers are four processors or less,
> and Linux scales very nicely at that level. If Sun tries to 'take
> back' the commodity server market with open source Solaris, they
> have a hard struggle ahead of them. They might have gotten some
> traction five years ago... but the market has changed a lot
> since then.
>
> Sun's best strategy is to differentiate themselves on hardware,
> fold Solaris's high end features into Linux, and save themselves
> a bucketload of development cost in the process. I just don't
> expect open source Solaris to gain the developer following
> that Linux already has, so Sun will likely continue to carry
> most of the Solaris development burden. If they go Linux, donate
> some Sun big iron to a few universities and prominent kernel
> developers, they begin to reap immediate benefits.
>
I'm not Sun's greatest advocate, but I do use Sun equipment.
I think Sun has a very agressive strategy. I like what I've seen
in their Opteron designs... I think Sun may well become the leader
in that space. I believe Sun has set an agressive goal to get
20% (?) or so of the x86 server market. To me the biggest
transition is moving from high-margin SPARC (with lower volumes)
to low-margin Opteron (with extremely high volumes). It is
a significant change for Sun... but IMHO, they have about the
best chance of pulling it off vs. any other company. Sun has a lot
of guts and teeth where it counts. Let's see what they can do.
With regards to differentiation, look for Sun to have the first
128-way Opteron (I used to say 64... but that's done now).
Sun is already flooding the market with VERY inexpensive dual
Opteron platforms (I own one, w2100z... easily had for <$2000 from them
now... WOW!). Shoot, just last year I bought a SunBlade 150
and the sucker cost us almost $3000. That $2000 Opteron runs circles
around even a SunBlade 2000 that costs $20000!
Sun obviously wants people to run Solaris... HOWEVER, they WILL
NOT miss a sale if you want to run Linux... or EVEN WINDOWS.
All of the new Opteron platforms are WHQL (sp?) certified to
run Windows. Sun wants to be THE supplier of your hardware
whether you choose Solaris, Linux or Windows.
At the price levels they are offering, I think they have the
best Opteron based platform out there right now.
Hope nobody is too in love with SPARC (please remember that
Sun is Sun.... NOT Sun is SPARC). I wish Sun the very
best on this new direction.
No comment on OpenSolaris.... in my opinion, it's actually
not necessary in Sun's overall new plan. Which actually
makes OpenSolaris more valuable in many respects.
|
|
|
0
|
|
|
|
Reply
|
Chris
|
12/14/2004 7:02:41 PM
|
|
On Tue, 14 Dec 2004 thad01@dexter.glaci.com wrote:
> switch. The fact is, most servers are four processors or less,
> and Linux scales very nicely at that level. If Sun tries to 'take
Perhaps, but Solaris has other enterprise features that are important
even on low end, 1-4 way systems.
> back' the commodity server market with open source Solaris, they
> have a hard struggle ahead of them. They might have gotten some
I agree that Sun has a challenge ahead of them. But I think their
future is rosy.
> Sun's best strategy is to differentiate themselves on hardware,
> fold Solaris's high end features into Linux, and save themselves
Completely disagree. Why would Sun waste R&D $$$ on Linux? They
already have an enterprise class (soon to be) open source OS, that
is free. If you want Solaris features, why not migrate to Solaris?
> a bucketload of development cost in the process. I just don't
> expect open source Solaris to gain the developer following
> that Linux already has, so Sun will likely continue to carry
Maybe, maybe not. You seem to forget that it's only relatively
recently that Linux was the prime open source platform. It wasn't
that long ago when stuff was first developed on Solaris/SunOS.
It could well be that those that aren't rabid Linux advocates
might see Open Solaris as a new and interesting intellectual
challenge, and tag along for the ride for the sheer hack value.
> most of the Solaris development burden. If they go Linux, donate
> some Sun big iron to a few universities and prominent kernel
> developers, they begin to reap immediate benefits.
What's to stop them from doing that for Open Solaris?
--
Rich Teer, SCNA, SCSA, author of "Solaris Systems Programming"
President,
Rite Online Inc.
Voice: +1 (250) 979-1638
URL: http://www.rite-group.com/rich
|
|
|
0
|
|
|
|
Reply
|
Rich
|
12/14/2004 7:10:35 PM
|
|
<thad01@dexter.glaci.com> wrote in message
news:41bf289d$0$46427$4c5efa9e@news.dias.net...
> The fact is, most servers are four processors or less,
> and Linux scales very nicely at that level. If Sun tries to 'take
> back' the commodity server market with open source Solaris, they
> have a hard struggle ahead of them. They might have gotten some
> traction five years ago... but the market has changed a lot
> since then.
But if the transition to a more stable platform is "mostly" transparent,
then the there isn't that much of a struggle.
Granted, we have no facts yet but consider IF the following are true:
o Janus lets users move "legacy" (funny -- Linux as legacy, oh well) Linux
binaries to Solaris 10
o Solaris 10 has good hardware support for server components
o Solaris 10 support compiling of Linux source code (dunno if it will or
not)
o Solaris 10 has comparable pricing model to Linux
o Solaris 10 administration, while different from Linux, changing to Sol 10
from Linux is less painful than from Windows
So, IF Solaris 10 provides reasonable support for Linux applications, and a
stable platform for vendor binaries in the future, that makes adoption of
Solaris 10 less expensive, and less risky.
Simple case, say your vendor has App 2.0 of his application. But, the old
version runs on Red Hat XYZ vs the new one on Red Hat QED. In order to use
the new version you need to upgrade your Linux box. Now, if that vendor
supports App 2.0 on Solaris, or if they will support the App running in
Janus of Sol 10, then why not upgrade to Sol 10 rather than Red Hat QED? By
doing so you have a pretty darn good chance that when App 2.5 comes out, you
won't need to update your OS to support the application.
If for no other reason than simply being able to keep your OS stable (i.e.
not for any of the other improvements in Sol 10), that seems like a
compelling concern to think about. It's one thing having to update your
application, but updating the OS is a real pain.
And vendors will think about that as well. Janus gives vendors the ability
to migrate their Linux apps to Sol 10. First they can run and support it in
Janus, then they can try to port the source code over and support a native
version (assuming they may not already have a SPARC version). And once they
make that leap, they have a more stable platform to release binary x86
applications.
Now, obviously this is not all going to happen as soon as Sol 10 hits the
market. But I do think it will happen soon, within the year we'll be seeing
movement. I think if Sun manages to push the "port to Sol 10" meme, that as
folks approach the need for either a new system, or having to upgrade their
old system, they may well consider switching over.
While there are some organizations attracted to Red Hat because of its open
source roots, I think most are attracted to its (originally) low price and
functionality. But if I can run the same commodity software (like OSS Web
server tools, etc.), as well as vendor binaries, on a more stable platform,
with similar pricing chracteristics, then all of a sudden Red Hat loses its
luster. The one thing that Red Hat can not guarantee, simply, is the
stability of its underlying platform, because it's not in control of it. Sun
can, and does, control that, and now it's coming out on the same range of
commodity server hardware, with similar performance and pricing, but with
more long term stability (as demonstrated historically -- past performance
does not guarantee future results :-)).
That makes Sol 10 a compelling option in the future for new builds and
upgrades, IMHO, and all of this is regardless of what happens with
OpenSolaris. OpenSolaris can just be icing on the cake.
This is why I think that Sun can compete, and compete well, with Linux on
the low end, but it won't happen overnight.
Regards,
Will Hartung
(willh@msoft.com)
|
|
|
0
|
|
|
|
Reply
|
Will
|
12/14/2004 7:55:42 PM
|
|
In comp.os.linux.advocacy Rich Teer <rich.teer@rite-group.com> wrote:
>> most of the Solaris development burden. If they go Linux, donate
>> some Sun big iron to a few universities and prominent kernel
>> developers, they begin to reap immediate benefits.
>
> What's to stop them from doing that for Open Solaris?
There is nothing to stop them doing it, and if they are serious
about open Solaris they will. My argument is that their return on
investment would be better in trying to tap into the existing Linux
developer pool rather than trying to build a new community around
Solaris. They seem to be clinging to the idea of using Solaris to
differentiate themselves from the other hardware vendors. That
made sense in the old days of the closed source unix wars, but is
much harder to pull off in today's fast paced open source world.
Of course I'm doing some major crystal ball gazing here. If Sun
plays it right and engages the community with the right kind of
challenges, they can certainly build a thriving developer pool
around Solaris. Given some of their past missteps, however, I'm
not so optimistic. Sun has actually done a lot of great stuff
for the open source community, yet their upper managment continues
to throw away goodwill by the truckloads with some of their
boneheaded statements and other ill-thought actions. I'm not
convinced they have the 'community building' knack. Time will
tell, and I wish them well. Sun has done some great things over
the years and continues to churn out some really great tech.
Later,
--
Thad Phetteplace
http://www.AliensForBush.com - Geeky Political Humor
|
|
|
0
|
|
|
|
Reply
|
thad01
|
12/14/2004 8:45:46 PM
|
|
Mike Cox wrote:
> That is why they post so much FUD on sites such as slashdot.org. The
> reason Linux became popular was that they were not as conservative as BSD,
> and
> would add features constantly. That meant if you wanted some enterprise
> functionality for free, you'd have to go with linux (SMP, etc.).
>
> The downside of that linux philosophy was that backwards compatiblitiy
> went
> down the drain. glibc interfaces and kernel calls would change
> significantly between versions. Sometimes the "enterprise" feature would
> be botched, like the scheduler in the 2.4 kernel.
>
> Now that Solaris 10 is going to be open sourced, Linux advocates rightly
> fear that Linux will be abandoned in favor of Solaris 10. Similar to how
> BSD was left in droves for Linux. That's right, SUN's brilliant strategy
> of making Solaris open sourced is using the linux advocates' techniques
> against them.
>
> The reason people will move to Solaris in droves is the same that caused
> BSD's user flight. It supports more enterprise features. But the
> difference this time is that Solaris is cheaper than Redhat, is completely
> backwards compatible, and scales much better on many more CPUs.
>
> Now if only SUN would release its compiler under a BSD license....
Solaris 10 was supposed to address the poor performance issues with previous
versions. After installing it, testing it I can say that "it doesn't"
address the performance issue at all... It may scale pretty well, but it
pokes around under light load as if it was carry the world...
It may be better than windows, but it pretty much looks like an ugly duck
when compared to Linux.
--
******************************************************************************
Registered Linux User Number 185956
FSF Associate Member number 2340 since 05/20/2004
Join me in chat at #linux-users on irc.freenode.net
Buy an Xbox for $149.00, run linux on it and Microsoft loses $150.00!
5:08pm up 66 days, 54 min, 8 users, load average: 0.10, 0.14, 0.09
|
|
|
0
|
|
|
|
Reply
|
Jerry
|
12/14/2004 10:17:28 PM
|
|
Rich Teer wrote:
> On Mon, 13 Dec 2004, Eltee wrote:
>
>> On the contrary. My guess is that Linux will adopt the best features of
>> Solaris and become an even better OS.
>
> Linux will doubtless "borrow" some features from Solaris, but some
> of the best features will be left out because Linus won't want them.
> I mean, if he won't accept the ability to produce a core dump on
> kernel panic, what chance has something like DTrace got?
>
Bullshit! Sun's CDDL conflicts with the GPL and no Open Source developer is
going to touch it with a 20 foot stick. And I am sure that Sun would love
to have some of its code show up in Linux so it can throw a big giant
monkey wrench into the development and adoption of Linux.
|
|
|
0
|
|
|
|
Reply
|
Freeride
|
12/15/2004 6:11:38 AM
|
|
On Tue, 14 Dec 2004, Freeride wrote:
> Bullshit! Sun's CDDL conflicts with the GPL and no Open Source developer is
Who says Open Solaris will use CDDL?
> going to touch it with a 20 foot stick. And I am sure that Sun would love
That's kind of amusing. I can name three open source developers
who are actively involved with the Open Solaris pilot. So much
for your FUD. Repeat after me until it sinks in: Open Source != GPL.
> to have some of its code show up in Linux so it can throw a big giant
> monkey wrench into the development and adoption of Linux.
I suspect that some Sun code (legally) is already in Linux.
BTW, it's bad netiquette to change the newsgroup line without
notice, so I've re-established the original one.
--
Rich Teer, SCNA, SCSA, author of "Solaris Systems Programming"
President,
Rite Online Inc.
Voice: +1 (250) 979-1638
URL: http://www.rite-group.com/rich
|
|
|
0
|
|
|
|
Reply
|
Rich
|
12/15/2004 6:29:52 AM
|
|
Jerry McBride <mcbrides9@comcast.net> writes:
>Solaris 10 was supposed to address the poor performance issues with previous
>versions. After installing it, testing it I can say that "it doesn't"
>address the performance issue at all... It may scale pretty well, but it
>pokes around under light load as if it was carry the world...
What poor performance issues? I've always found that Solaris is
faster than Linux for the things I do. But that's perhaps a function
of the things I do; there's still some catching to do when it comes
to Xserver performace; (no apgart driver available yet).
Casper
--
Expressed in this posting are my opinions. They are in no way related
to opinions held by my employer, Sun Microsystems.
Statements on Sun products included here are not gospel and may
be fiction rather than truth.
|
|
|
0
|
|
|
|
Reply
|
Casper
|
12/15/2004 2:01:28 PM
|
|
Freeride <freeride@maillinux.org> writes:
>Bullshit! Sun's CDDL conflicts with the GPL and no Open Source developer is
>going to touch it with a 20 foot stick. And I am sure that Sun would love
>to have some of its code show up in Linux so it can throw a big giant
>monkey wrench into the development and adoption of Linux.
Any license other than the GPL conflicts with the GPL. There are a lot
of freeware developers who wouldn't touch the GPL with a 30 foot stick.
(And, as the original author, you can release code under several concurrent
licenses; there exists code which is both under the GPL and the BSD license
and where you as user can chose which license you use for redistribution)
Casper
--
Expressed in this posting are my opinions. They are in no way related
to opinions held by my employer, Sun Microsystems.
Statements on Sun products included here are not gospel and may
be fiction rather than truth.
|
|
|
0
|
|
|
|
Reply
|
Casper
|
12/15/2004 2:03:34 PM
|
|
In article <Pine.SOL.4.58.0412141103300.10972@zaphod>,
Rich Teer <rich.teer@rite-group.com> wrote:
>> Sun's best strategy is to differentiate themselves on hardware,
>> fold Solaris's high end features into Linux, and save themselves
>
>Completely disagree. Why would Sun waste R&D $$$ on Linux? They
>already have an enterprise class (soon to be) open source OS, that
>is free. If you want Solaris features, why not migrate to Solaris?
I completly agree, it is easier to add the few extra features from Linux
than adding the missing features to Linux. Writing drivers is not that
hard and Sun did already start with massive driver development and a
new x86 driver group.
>> a bucketload of development cost in the process. I just don't
>> expect open source Solaris to gain the developer following
>> that Linux already has, so Sun will likely continue to carry
>
>Maybe, maybe not. You seem to forget that it's only relatively
>recently that Linux was the prime open source platform. It wasn't
>that long ago when stuff was first developed on Solaris/SunOS.
I also forsee that there are other reasons why a developer would switch
from Linux to Solaris.
One reason is that the Linux development is dominated by Linus.
It is even funny to see that Linus repeatdly publishes statements that
he cannot see how Sun may dominate the OSS market. From what I know
about Sun, I can tell that Sun does not seem to be interested in
dominating the OpenSolaris development. This may be an important factor
for OpenSolaris.
--
EMail:joerg@schily.isdn.cs.tu-berlin.de (home) J�rg Schilling D-13353 Berlin
js@cs.tu-berlin.de (uni) If you don't have iso-8859-1
schilling@fokus.fraunhofer.de (work) chars I am J"org Schilling
URL: http://www.fokus.fraunhofer.de/usr/schilling ftp://ftp.berlios.de/pub/schily
|
|
|
0
|
|
|
|
Reply
|
js
|
12/15/2004 2:35:57 PM
|
|
In article <oc5492xm0p.ln2@spinner.my.domain>,
Jerry McBride <mcbrides9@comcast.net> wrote:
>Solaris 10 was supposed to address the poor performance issues with previous
>versions. After installing it, testing it I can say that "it doesn't"
>address the performance issue at all... It may scale pretty well, but it
>pokes around under light load as if it was carry the world...
Looks like you never compared Solaris & Linux performance :-(
With my usage, Linux always has been slower or at most only as fast than
Solaris.
Linux has poor memory management and scheduling. How can such an OS
behave better than Solaris?
--
EMail:joerg@schily.isdn.cs.tu-berlin.de (home) J�rg Schilling D-13353 Berlin
js@cs.tu-berlin.de (uni) If you don't have iso-8859-1
schilling@fokus.fraunhofer.de (work) chars I am J"org Schilling
URL: http://www.fokus.fraunhofer.de/usr/schilling ftp://ftp.berlios.de/pub/schily
|
|
|
0
|
|
|
|
Reply
|
js
|
12/15/2004 2:59:57 PM
|
|
In article <41c043b8$0$19389$e4fe514c@news.xs4all.nl>,
Casper H.S. Dik <Casper.Dik@Sun.COM> wrote:
>Jerry McBride <mcbrides9@comcast.net> writes:
>
>>Solaris 10 was supposed to address the poor performance issues with previous
>>versions. After installing it, testing it I can say that "it doesn't"
>>address the performance issue at all... It may scale pretty well, but it
>>pokes around under light load as if it was carry the world...
>
>What poor performance issues? I've always found that Solaris is
>faster than Linux for the things I do. But that's perhaps a function
>of the things I do; there's still some catching to do when it comes
>to Xserver performace; (no apgart driver available yet).
I concur.
I don't compare Xserver performance, but for the typical work of a developer
and for Web Servers, Solaris is faster than Linux and already has been
faster than Linux in pre Solaris 10 times.
--
EMail:joerg@schily.isdn.cs.tu-berlin.de (home) J�rg Schilling D-13353 Berlin
js@cs.tu-berlin.de (uni) If you don't have iso-8859-1
schilling@fokus.fraunhofer.de (work) chars I am J"org Schilling
URL: http://www.fokus.fraunhofer.de/usr/schilling ftp://ftp.berlios.de/pub/schily
|
|
|
0
|
|
|
|
Reply
|
js
|
12/15/2004 3:33:38 PM
|
|
In article <uAQvd.63712$Af.6022@fed1read07>,
Freeride <freeride@maillinux.org> wrote:
I corrected your violation of the nettiquette (you did change
the Newsgroups: witout notice!)
>Bullshit! Sun's CDDL conflicts with the GPL and no Open Source developer is
>going to touch it with a 20 foot stick. And I am sure that Sun would love
>to have some of its code show up in Linux so it can throw a big giant
>monkey wrench into the development and adoption of Linux.
Well, with CDDL people like Linus have more problems with trying to dominate
a project..... but not all people behave this way.
Note that it is not the CDDL that conflichts with the GPL but it is the
GPL that conflicts with the CDDL.
The problem is that the GPL tries to reduce freedom in a way that is not
compatible with the CDDL.
The GPL does not conflict with only the BSD license because the BSD license
tolerates this freedom reduction.
The problem with the GPL is that it tries to enforce several restrictions
and that it does not allow a mix of "compatible" licenses. The fact that
it tries to impose GPL rules to any piece of software (if taken carefully)
only allows the original Author to add software to a GPL project. This is
all not true for the CDDL. So why should authors not like CDDL?
--
EMail:joerg@schily.isdn.cs.tu-berlin.de (home) J�rg Schilling D-13353 Berlin
js@cs.tu-berlin.de (uni) If you don't have iso-8859-1
schilling@fokus.fraunhofer.de (work) chars I am J"org Schilling
URL: http://www.fokus.fraunhofer.de/usr/schilling ftp://ftp.berlios.de/pub/schily
|
|
|
0
|
|
|
|
Reply
|
js
|
12/15/2004 4:02:19 PM
|
|
In article <cppjhd$1tr$1@news.cs.tu-berlin.de>,
js@cs.tu-berlin.de (Joerg Schilling) writes:
> In article <oc5492xm0p.ln2@spinner.my.domain>,
> Jerry McBride <mcbrides9@comcast.net> wrote:
>
>>Solaris 10 was supposed to address the poor performance issues with previous
>>versions. After installing it, testing it I can say that "it doesn't"
>>address the performance issue at all... It may scale pretty well, but it
>>pokes around under light load as if it was carry the world...
>
> Looks like you never compared Solaris & Linux performance :-(
>
> With my usage, Linux always has been slower or at most only as fast than
> Solaris.
>
> Linux has poor memory management and scheduling. How can such an OS
> behave better than Solaris?
Interesting question, but there are some not entirely uncommon
circumstances where it does. Particularly, process initialization -
startup, scripts, builds, configure scripts. You can get up to a factor
2 slowdown in Solaris compared to Linux.
I would agree with much of the sentiment above - push them, and Solaris
does behave much better. But under light or modest load, and there are
cases where Solaris performance really isn't quite up to par.
One problem is that it's that sort of behaviour that governs many
users' impressions of a system. It's something we need to work at.
(As part of OpenSolaris, I've tried looking at the scheduler, because I
know there are interactions there. But I don't have the familiarity
with the code or the skills yet to understand it.)
--
-Peter Tribble
MRC Rosalind Franklin Centre for Genomics Research
http://www.rfcgr.mrc.ac.uk/~ptribble/
|
|
|
0
|
|
|
|
Reply
|
ptribble
|
12/15/2004 7:54:21 PM
|
|
In article <cpq4pd$jjn$1@helium.hgmp.mrc.ac.uk>,
Peter C. Tribble <ptribble@hgmp.mrc.ac.uk> wrote:
>>>Solaris 10 was supposed to address the poor performance issues with previous
>>>versions. After installing it, testing it I can say that "it doesn't"
>>>address the performance issue at all... It may scale pretty well, but it
>>>pokes around under light load as if it was carry the world...
>>
>> Looks like you never compared Solaris & Linux performance :-(
>>
>> With my usage, Linux always has been slower or at most only as fast than
>> Solaris.
>>
>> Linux has poor memory management and scheduling. How can such an OS
>> behave better than Solaris?
>
>Interesting question, but there are some not entirely uncommon
>circumstances where it does. Particularly, process initialization -
>startup, scripts, builds, configure scripts. You can get up to a factor
>2 slowdown in Solaris compared to Linux.
For Solaris 2.6 and older I would definitely agree that startup of programs
was slower.
Solaris and Solaris apps if linked in a SVr4 compliant way need more dynamic
libs than Linux. This should have been reduced wwith never versions.
I don't believe that the Solaris schedular is related to this.
--
EMail:joerg@schily.isdn.cs.tu-berlin.de (home) J�rg Schilling D-13353 Berlin
js@cs.tu-berlin.de (uni) If you don't have iso-8859-1
schilling@fokus.fraunhofer.de (work) chars I am J"org Schilling
URL: http://www.fokus.fraunhofer.de/usr/schilling ftp://ftp.berlios.de/pub/schily
|
|
|
0
|
|
|
|
Reply
|
js
|
12/15/2004 8:03:31 PM
|
|
js@cs.tu-berlin.de (Joerg Schilling) writes:
> In article <cpq4pd$jjn$1@helium.hgmp.mrc.ac.uk>,
> Peter C. Tribble <ptribble@hgmp.mrc.ac.uk> wrote:
>
>>>>Solaris 10 was supposed to address the poor performance issues with previous
>>>>versions. After installing it, testing it I can say that "it doesn't"
>>>>address the performance issue at all... It may scale pretty well, but it
>>>>pokes around under light load as if it was carry the world...
>>>
>>> Looks like you never compared Solaris & Linux performance :-(
>>>
>>> With my usage, Linux always has been slower or at most only as fast than
>>> Solaris.
>>>
>>> Linux has poor memory management and scheduling. How can such an OS
>>> behave better than Solaris?
>>
>>Interesting question, but there are some not entirely uncommon
>>circumstances where it does. Particularly, process initialization -
>>startup, scripts, builds, configure scripts. You can get up to a factor
>>2 slowdown in Solaris compared to Linux.
>
> For Solaris 2.6 and older I would definitely agree that startup of programs
> was slower.
>
> Solaris and Solaris apps if linked in a SVr4 compliant way need more dynamic
> libs than Linux. This should have been reduced wwith never versions.
> I don't believe that the Solaris schedular is related to this.
I'd like more info.
Here's what I see, Mandrake 10.0 vs. Solaris 2.8:
linux> ldd /bin/ls
linux-gate.so.1 => (0xffffe000)
librt.so.1 => /lib/tls/librt.so.1 (0x40025000)
libtermcap.so.2 => /lib/libtermcap.so.2 (0x4003a000)
libc.so.6 => /lib/tls/libc.so.6 (0x4003e000)
libpthread.so.0 => /lib/tls/libpthread.so.0 (0x40186000)
/lib/ld-linux.so.2 => /lib/ld-linux.so.2 (0x40000000)
linux> ldd /usr/bin/make
linux-gate.so.1 => (0xffffe000)
libc.so.6 => /lib/tls/libc.so.6 (0x40025000)
/lib/ld-linux.so.2 => /lib/ld-linux.so.2 (0x40000000)
linux> ldd /bin/cat
linux-gate.so.1 => (0xffffe000)
libc.so.6 => /lib/tls/libc.so.6 (0x40025000)
/lib/ld-linux.so.2 => /lib/ld-linux.so.2 (0x40000000)
solaris> ldd /bin/ls
libc.so.1 => /usr/lib/libc.so.1
libdl.so.1 => /usr/lib/libdl.so.1
/usr/platform/SUNW,Ultra-5_10/lib/libc_psr.so.1
solaris> ldd /usr/ccs/bin/make
libintl.so.1 => /usr/lib/libintl.so.1
libnsl.so.1 => /usr/lib/libnsl.so.1
libsocket.so.1 => /usr/lib/libsocket.so.1
libw.so.1 => /usr/lib/libw.so.1
libm.so.1 => /usr/lib/libm.so.1
libc.so.1 => /usr/lib/libc.so.1
libdl.so.1 => /usr/lib/libdl.so.1
libmp.so.2 => /usr/lib/libmp.so.2
/usr/platform/SUNW,Ultra-5_10/lib/libc_psr.so.1
solaris> ldd /bin/cat
libc.so.1 => /usr/lib/libc.so.1
libdl.so.1 => /usr/lib/libdl.so.1
/usr/platform/SUNW,Ultra-5_10/lib/libc_psr.so.1
Looks like a mixed bag, but I can see some flaws in
this test, Solaris make seems to be network aware,
gnu ls provides color.
|
|
|
0
|
|
|
|
Reply
|
Dan
|
12/15/2004 8:24:20 PM
|
|
ptribble@hgmp.mrc.ac.uk (Peter C. Tribble) writes:
>(As part of OpenSolaris, I've tried looking at the scheduler, because I
>know there are interactions there. But I don't have the familiarity
>with the code or the skills yet to understand it.)
The Solaris scheduler is fairly complicated; however, it's been an
O(1) scheduler (a much touted Linux 2.something feature) since about
Solaris 2.4.
Casper
--
Expressed in this posting are my opinions. They are in no way related
to opinions held by my employer, Sun Microsystems.
Statements on Sun products included here are not gospel and may
be fiction rather than truth.
|
|
|
0
|
|
|
|
Reply
|
Casper
|
12/15/2004 8:28:51 PM
|
|
In article <41c09e83$0$21106$e4fe514c@news.xs4all.nl>,
Casper H.S. Dik <Casper.Dik@Sun.COM> writes:
> ptribble@hgmp.mrc.ac.uk (Peter C. Tribble) writes:
>
>>(As part of OpenSolaris, I've tried looking at the scheduler, because I
>>know there are interactions there. But I don't have the familiarity
>>with the code or the skills yet to understand it.)
>
> The Solaris scheduler is fairly complicated; however, it's been an
> O(1) scheduler (a much touted Linux 2.something feature) since about
> Solaris 2.4.
That it may be. The problem we saw was (as far as I recall - sorry Gary
if I mess up) due to a scheduling race in the following scenario:
Shell script runs through many short lived processes. When new process
is launched, parent sleeps. When child exits, two scheduling events are
generated - parent wakes up and tries to find a cpu to run on, and the
cpu the child was running on is now free.
Needless to say it didn't always get it right. Especially if this is a
multiprocessor system with something else happening. (The fix was to use
processor sets or even a simple pbind to stop processes wandering
around between cpus - in other words, to override some of the
scheduling decisions.)
We were seeing a slowdown of something like a factor 10 in some extreme
cases. I mean, you would think that if you had a multiprocessor machine
then you wouldn't be able to detect much of an impact on system
responsiveness provided there was at least one CPU free. The reality
was otherwise, unfortunately.
Must run some tests on build 72, to see how it's changed
recently. Things are getting much better in the last couple of Solaris
express releases.
--
-Peter Tribble
MRC Rosalind Franklin Centre for Genomics Research
http://www.rfcgr.mrc.ac.uk/~ptribble/
|
|
|
0
|
|
|
|
Reply
|
ptribble
|
12/15/2004 10:26:04 PM
|
|
On Wed, 15 Dec 2004 15:24:20 -0500, Dan Espen wrote:
> js@cs.tu-berlin.de (Joerg Schilling) writes:
>
>> In article <cpq4pd$jjn$1@helium.hgmp.mrc.ac.uk>,
>> Peter C. Tribble <ptribble@hgmp.mrc.ac.uk> wrote:
>>
>>>>>Solaris 10 was supposed to address the poor performance issues with previous
>>>>>versions. After installing it, testing it I can say that "it doesn't"
>>>>>address the performance issue at all... It may scale pretty well, but it
>>>>>pokes around under light load as if it was carry the world...
>>>>
>>>> Looks like you never compared Solaris & Linux performance :-(
>>>>
>>>> With my usage, Linux always has been slower or at most only as fast than
>>>> Solaris.
>>>>
>>>> Linux has poor memory management and scheduling. How can such an OS
>>>> behave better than Solaris?
>>>
>>>Interesting question, but there are some not entirely uncommon
>>>circumstances where it does. Particularly, process initialization -
>>>startup, scripts, builds, configure scripts. You can get up to a factor
>>>2 slowdown in Solaris compared to Linux.
>>
>> For Solaris 2.6 and older I would definitely agree that startup of programs
>> was slower.
>>
>> Solaris and Solaris apps if linked in a SVr4 compliant way need more dynamic
>> libs than Linux. This should have been reduced wwith never versions.
>> I don't believe that the Solaris schedular is related to this.
>
> I'd like more info.
>
> Here's what I see, Mandrake 10.0 vs. Solaris 2.8:
>
>
> linux> ldd /bin/ls
> linux-gate.so.1 => (0xffffe000)
> librt.so.1 => /lib/tls/librt.so.1 (0x40025000)
> libtermcap.so.2 => /lib/libtermcap.so.2 (0x4003a000)
> libc.so.6 => /lib/tls/libc.so.6 (0x4003e000)
> libpthread.so.0 => /lib/tls/libpthread.so.0 (0x40186000)
> /lib/ld-linux.so.2 => /lib/ld-linux.so.2 (0x40000000)
> linux> ldd /usr/bin/make
> linux-gate.so.1 => (0xffffe000)
> libc.so.6 => /lib/tls/libc.so.6 (0x40025000)
> /lib/ld-linux.so.2 => /lib/ld-linux.so.2 (0x40000000)
> linux> ldd /bin/cat
> linux-gate.so.1 => (0xffffe000)
> libc.so.6 => /lib/tls/libc.so.6 (0x40025000)
> /lib/ld-linux.so.2 => /lib/ld-linux.so.2 (0x40000000)
>
>
> solaris> ldd /bin/ls
> libc.so.1 => /usr/lib/libc.so.1
> libdl.so.1 => /usr/lib/libdl.so.1
> /usr/platform/SUNW,Ultra-5_10/lib/libc_psr.so.1
> solaris> ldd /usr/ccs/bin/make
> libintl.so.1 => /usr/lib/libintl.so.1
> libnsl.so.1 => /usr/lib/libnsl.so.1
> libsocket.so.1 => /usr/lib/libsocket.so.1
> libw.so.1 => /usr/lib/libw.so.1
> libm.so.1 => /usr/lib/libm.so.1
> libc.so.1 => /usr/lib/libc.so.1
> libdl.so.1 => /usr/lib/libdl.so.1
> libmp.so.2 => /usr/lib/libmp.so.2
> /usr/platform/SUNW,Ultra-5_10/lib/libc_psr.so.1
> solaris> ldd /bin/cat
> libc.so.1 => /usr/lib/libc.so.1
> libdl.so.1 => /usr/lib/libdl.so.1
> /usr/platform/SUNW,Ultra-5_10/lib/libc_psr.so.1
>
>
> Looks like a mixed bag, but I can see some flaws in
> this test, Solaris make seems to be network aware,
> gnu ls provides color.
You're not comparing like with like. The biggest issue with this
comparison is that you're comparing an x86 (based) architecture with a
SPARC arch. That does not make sense.
On an older Solaris 9 box:
$ cat /etc/release
Solaris 9 8/03 s9x_u4wos_08b x86
Copyright 2003 Sun Microsystems, Inc. All Rights Reserved.
Use is subject to license terms.
Assembled 17 June 2003
$ ldd /bin/ls
libc.so.1 => /usr/lib/libc.so.1
libdl.so.1 => /usr/lib/libdl.so.1
On a newer Solaris 9 box:
$ cat /etc/release
Solaris 9 9/04 s9x_u7wos_09 x86
Copyright 2004 Sun Microsystems, Inc. All Rights Reserved.
Use is subject to license terms.
Assembled 28 June 2004
$ ldd /bin/ls
libc.so.1 => /usr/lib/libc.so.1
libdl.so.1 => /usr/lib/libdl.so.1
On a Sol 10 box running Beta 70 in 32-bit mode:
$ cat /etc/release
Solaris 10 s10_70 X86
Copyright 2004 Sun Microsystems, Inc. All Rights Reserved.
Use is subject to license terms.
Assembled 22 October 2004
$ ldd /bin/ls
libc.so.1 => /lib/libc.so.1
libm.so.2 => /lib/libm.so.2
Conclusion: Give a more recent version of Solaris a spin on an x86
server, ideally the latest available Solaris Express release (B72)
and I think you'll be impressed.
Al Hopper Logical Approach Inc, Plano, TX. al@logical-approach.com
Voice: 972.379.2133 Fax: 972.379.2134
"It's a good thing that HP never acquired the rights to penicillin. If they had,
mankind would have perished from widespread disease while HP tried to figure out
how to integrate it with anthrax." TD (fullname unknown).
|
|
|
0
|
|
|
|
Reply
|
Al
|
12/15/2004 11:22:23 PM
|
|
On Wed, 15 Dec 2004 15:24:20 -0500, Dan Espen wrote:
> js@cs.tu-berlin.de (Joerg Schilling) writes:
>
>> In article <cpq4pd$jjn$1@helium.hgmp.mrc.ac.uk>,
>> Peter C. Tribble <ptribble@hgmp.mrc.ac.uk> wrote:
>>
>>>>>Solaris 10 was supposed to address the poor performance issues with previous
>>>>>versions. After installing it, testing it I can say that "it doesn't"
>>>>>address the performance issue at all... It may scale pretty well, but it
>>>>>pokes around under light load as if it was carry the world...
>>>>
>>>> Looks like you never compared Solaris & Linux performance :-(
>>>>
>>>> With my usage, Linux always has been slower or at most only as fast than
>>>> Solaris.
>>>>
>>>> Linux has poor memory management and scheduling. How can such an OS
>>>> behave better than Solaris?
>>>
>>>Interesting question, but there are some not entirely uncommon
>>>circumstances where it does. Particularly, process initialization -
>>>startup, scripts, builds, configure scripts. You can get up to a factor
>>>2 slowdown in Solaris compared to Linux.
>>
>> For Solaris 2.6 and older I would definitely agree that startup of programs
>> was slower.
>>
>> Solaris and Solaris apps if linked in a SVr4 compliant way need more dynamic
>> libs than Linux. This should have been reduced wwith never versions.
>> I don't believe that the Solaris schedular is related to this.
>
> I'd like more info.
>
> Here's what I see, Mandrake 10.0 vs. Solaris 2.8:
>
>
> linux> ldd /bin/ls
> linux-gate.so.1 => (0xffffe000)
> librt.so.1 => /lib/tls/librt.so.1 (0x40025000)
> libtermcap.so.2 => /lib/libtermcap.so.2 (0x4003a000)
> libc.so.6 => /lib/tls/libc.so.6 (0x4003e000)
> libpthread.so.0 => /lib/tls/libpthread.so.0 (0x40186000)
> /lib/ld-linux.so.2 => /lib/ld-linux.so.2 (0x40000000)
> linux> ldd /usr/bin/make
> linux-gate.so.1 => (0xffffe000)
> libc.so.6 => /lib/tls/libc.so.6 (0x40025000)
> /lib/ld-linux.so.2 => /lib/ld-linux.so.2 (0x40000000)
> linux> ldd /bin/cat
> linux-gate.so.1 => (0xffffe000)
> libc.so.6 => /lib/tls/libc.so.6 (0x40025000)
> /lib/ld-linux.so.2 => /lib/ld-linux.so.2 (0x40000000)
>
>
> solaris> ldd /bin/ls
> libc.so.1 => /usr/lib/libc.so.1
> libdl.so.1 => /usr/lib/libdl.so.1
> /usr/platform/SUNW,Ultra-5_10/lib/libc_psr.so.1
> solaris> ldd /usr/ccs/bin/make
> libintl.so.1 => /usr/lib/libintl.so.1
> libnsl.so.1 => /usr/lib/libnsl.so.1
> libsocket.so.1 => /usr/lib/libsocket.so.1
> libw.so.1 => /usr/lib/libw.so.1
> libm.so.1 => /usr/lib/libm.so.1
> libc.so.1 => /usr/lib/libc.so.1
> libdl.so.1 => /usr/lib/libdl.so.1
> libmp.so.2 => /usr/lib/libmp.so.2
> /usr/platform/SUNW,Ultra-5_10/lib/libc_psr.so.1
> solaris> ldd /bin/cat
> libc.so.1 => /usr/lib/libc.so.1
> libdl.so.1 => /usr/lib/libdl.so.1
> /usr/platform/SUNW,Ultra-5_10/lib/libc_psr.so.1
>
>
> Looks like a mixed bag, but I can see some flaws in
> this test, Solaris make seems to be network aware,
> gnu ls provides color.
You're not comparing like with like. The biggest issue with this
comparison is that you're comparing an x86 (based) architecture with a
SPARC arch. That does not make sense.
On an older Solaris 9 box:
$ cat /etc/release
Solaris 9 8/03 s9x_u4wos_08b x86
Copyright 2003 Sun Microsystems, Inc. All Rights Reserved.
Use is subject to license terms.
Assembled 17 June 2003
$ ldd /bin/ls
libc.so.1 => /usr/lib/libc.so.1
libdl.so.1 => /usr/lib/libdl.so.1
On a newer Solaris 9 box:
$ cat /etc/release
Solaris 9 9/04 s9x_u7wos_09 x86
Copyright 2004 Sun Microsystems, Inc. All Rights Reserved.
Use is subject to license terms.
Assembled 28 June 2004
$ ldd /bin/ls
libc.so.1 => /usr/lib/libc.so.1
libdl.so.1 => /usr/lib/libdl.so.1
On a Sol 10 box running Beta 70 in 32-bit mode:
$ cat /etc/release
Solaris 10 s10_70 X86
Copyright 2004 Sun Microsystems, Inc. All Rights Reserved.
Use is subject to license terms.
Assembled 22 October 2004
$ ldd /bin/ls
libc.so.1 => /lib/libc.so.1
libm.so.2 => /lib/libm.so.2
Conclusion: Give a more recent version of Solaris a spin on an x86
server, ideally the latest available Solaris Express release (B72)
and I think you'll be impressed.
Al Hopper Logical Approach Inc, Plano, TX. al@logical-approach.com
Voice: 972.379.2133 Fax: 972.379.2134
"It's a good thing that HP never acquired the rights to penicillin. If they had,
mankind would have perished from widespread disease while HP tried to figure out
how to integrate it with anthrax." TD (fullname unknown).
|
|
|
0
|
|
|
|
Reply
|
Al
|
12/15/2004 11:29:36 PM
|
|
Dan Espen <daneNO@SPAM.mk.telcordia.com> writes in comp.unix.solaris:
|linux> ldd /usr/bin/make
| linux-gate.so.1 => (0xffffe000)
| libc.so.6 => /lib/tls/libc.so.6 (0x40025000)
| /lib/ld-linux.so.2 => /lib/ld-linux.so.2 (0x40000000)
|solaris> ldd /usr/ccs/bin/make
| libintl.so.1 => /usr/lib/libintl.so.1
| libnsl.so.1 => /usr/lib/libnsl.so.1
| libsocket.so.1 => /usr/lib/libsocket.so.1
| libw.so.1 => /usr/lib/libw.so.1
| libm.so.1 => /usr/lib/libm.so.1
| libc.so.1 => /usr/lib/libc.so.1
| libdl.so.1 => /usr/lib/libdl.so.1
| libmp.so.2 => /usr/lib/libmp.so.2
| /usr/platform/SUNW,Ultra-5_10/lib/libc_psr.so.1
Eh, not sure how useful a comparison that is - if you link with -z
lazyload, many of those will never get loaded at startup (and some,
never at all). The linker guys have done some amazing things to
speed up loads when the right options are used.
--
________________________________________________________________________
Alan Coopersmith * alanc@alum.calberkeley.org * Alan.Coopersmith@Sun.COM
http://www.csua.berkeley.edu/~alanc/ * http://blogs.sun.com/alanc/
Working for, but definitely not speaking for, Sun Microsystems, Inc.
|
|
|
0
|
|
|
|
Reply
|
Alan
|
12/16/2004 3:35:08 AM
|
|
Al Hopper <al@logical-approach.com> writes:
> On Wed, 15 Dec 2004 15:24:20 -0500, Dan Espen wrote:
>
>> js@cs.tu-berlin.de (Joerg Schilling) writes:
>>
>>> In article <cpq4pd$jjn$1@helium.hgmp.mrc.ac.uk>,
>>> Peter C. Tribble <ptribble@hgmp.mrc.ac.uk> wrote:
....
>>> Solaris and Solaris apps if linked in a SVr4 compliant way need more dynamic
>>> libs than Linux. This should have been reduced wwith never versions.
>>
>> I'd like more info.
>>
>> Here's what I see, Mandrake 10.0 vs. Solaris 2.8:
>>
>>
>> linux> ldd /bin/ls
>> linux-gate.so.1 => (0xffffe000)
>> librt.so.1 => /lib/tls/librt.so.1 (0x40025000)
>> libtermcap.so.2 => /lib/libtermcap.so.2 (0x4003a000)
>> libc.so.6 => /lib/tls/libc.so.6 (0x4003e000)
>> libpthread.so.0 => /lib/tls/libpthread.so.0 (0x40186000)
>> /lib/ld-linux.so.2 => /lib/ld-linux.so.2 (0x40000000)
>> solaris> ldd /bin/ls
>> libc.so.1 => /usr/lib/libc.so.1
>> libdl.so.1 => /usr/lib/libdl.so.1
>> /usr/platform/SUNW,Ultra-5_10/lib/libc_psr.so.1
>>
>> Looks like a mixed bag, but I can see some flaws in
>> this test, gnu ls provides color.
>
> You're not comparing like with like. The biggest issue with this
> comparison is that you're comparing an x86 (based) architecture with a
> SPARC arch. That does not make sense.
It looked to me like the biggest issue
was the lack of equivalent function.
> On an older Solaris 9 box:
>
> $ cat /etc/release
> Solaris 9 8/03 s9x_u4wos_08b x86
> Copyright 2003 Sun Microsystems, Inc. All Rights Reserved.
> Use is subject to license terms.
> Assembled 17 June 2003
> $ ldd /bin/ls
> libc.so.1 => /usr/lib/libc.so.1
> libdl.so.1 => /usr/lib/libdl.so.1
All your examples show 2 shared libs needed on a basic
Solaris x86 executables where the minimum for linux seemed to be 3.
Above it says Solaris needs more dynamic libs than linux,
but for the simple cases, the opposite seems to be true.
|
|
|
0
|
|
|
|
Reply
|
Dan
|
12/16/2004 3:38:32 AM
|
|
In article <Pine.SOL.4.58.0412142224230.10972@zaphod> (Wed, 15 Dec 2004
06:29:52 +0000), Rich Teer wrote:
> Who says Open Solaris will use CDDL?
What's the point, if any, of the CDDL? Is Sun just testing the waters
to see how much negative feedback is generated? Why do they go to the
expense of creating a new "open source" [sic] license if they don't intend
to use it?
Who says Open [sic] Solaris won't use the CDDL?
> Repeat after me until it sinks in: Open Source != GPL.
The proper relation is, and remains:
Open Source" < GPL
> I suspect that some Sun code (legally) is already in Linux.
There may be. However, there isn't a single Sun Microsystems copyright
notice in the 2.6.9 kernel, so Sun's contribution, if there is one, is
miniscule. (OTOH, so what?)
--
"I think [Microsoft] have a pretty good story, but I tell you, game's on.
We've got to prove ourselves, and some people are choosing Linux."
-- Steve Ballmer. http://news.com.com/2008-1082-998297.html. 25 Apr 2003
|
|
|
0
|
|
|
|
Reply
|
Hamilcar
|
12/16/2004 4:14:52 AM
|
|
In article <cppn6b$aoc$1@news.cs.tu-berlin.de> (Wed, 15 Dec 2004 16:02:19
+0000), Joerg Schilling wrote:
> Well, with CDDL people like Linus have more problems with trying to dominate
> a project.....
Do you then claim that all arguments in favor of the CDDL [sic] devolve to
ad hominem attacks?
> [...] but not all people behave this way.
But some do:
| Warning: you are not allowed to modify or to remove the
| Copyright and version printing code below!
| See also GPL <A7> 2 subclause c)
Have you failed to read the GPL? You cannot require, even though you so
desire, the retention of "version printing code". You seem to have made a
mistake in using the GPL at any time.
| I am sorry for the inconvenience but I am forced to do this because
| some people create inofficial branches. These branches create
| problems but the initiators do not give support and thus cause the
| development of the official cdrecord versions to slow down because
| I am loaded with unneeded work.
Your attempt to dominate cdrecord hasn't been completely successful?
| Please note that this is a memorandum on how I interpret the GPL.
| If you use/modify/redistribute cdrecord, you need to accept it
| this way.
You don't understand that your interpretation of a license is not binding
on anyone if it's a misinterpretation?
> Note that it is not the CDDL that conflichts with the GPL but it is the
> GPL that conflicts with the CDDL.
Note that Sun was under no obligation to create a new license that
conflicted with the GPL. They did so anyway. Could this have been
an accidental oversight?
> The problem is that the GPL tries to reduce freedom in a way that is not
> compatible with the CDDL.
The problem is that the CDDL [sic] tries to reduce the freedoms for all
granted by the GPL, such that Sun may maintain sufficient control over its
code to prevent it from being used in certain other products -- notably
the Linux kernel.
> The problem with the GPL [...]
What you've attempted to label as a problem [sic] with the GPL is its
main strength. Its license cannot be diluted by such marketing
press releases from companies like Sun Microsystems.
> So why should authors not like CDDL?
Some programmers do not want their work to be converted to a proprietary
status. (Some want to allow this, and have the right to do so. Some
probably don't care.)
--
"It turns out Luddites don't know how to use software properly,
so you should look into that."
-- Bill Gates. Chairman, Microsoft. FOCUS Magazine interview. 10/23/1995
|
|
|
0
|
|
|
|
Reply
|
Hamilcar
|
12/16/2004 4:29:59 AM
|
|
In article <cppi4d$s9c$1@news.cs.tu-berlin.de> (Wed, 15 Dec 2004 14:35:57
+0000), Joerg Schilling wrote:
> One reason is that the Linux development is dominated by Linus.
Your penis envy is becoming obvious.
|
|
|
0
|
|
|
|
Reply
|
Hamilcar
|
12/16/2004 4:30:24 AM
|
|
On Wed, 15 Dec 2004, Hamilcar Barca wrote:
> What's the point, if any, of the CDDL? Is Sun just testing the waters
> to see how much negative feedback is generated? Why do they go to the
> expense of creating a new "open source" [sic] license if they don't intend
> to use it?
Sun intend to use CDDL. I'm just saying that they haven't commited to
using it for Open Solaris (nor have they publically said that they aren't).
Anyone outside Sun who states that Sun will use the CDDL for Open Solaris
is either speculating, or breaking the terms of an NDA.
> Who says Open [sic] Solaris won't use the CDDL?
Sun haven't publically commited to ANY license for Open Solaris yet. They've
only stated that it will be OSI approved.
Given that I'm a member of the Open Solaris pilot project I'm privvy
to more info than I can say here. And no, I'm not about to break
the terms of the NDA I signed!
--
Rich Teer, SCNA, SCSA, author of "Solaris Systems Programming"
. * * . * .* .
. * . .*
President, * . . /\ ( . . *
Rite Online Inc. . . / .\ . * .
.*. / * \ . .
. /* o \ .
Voice: +1 (250) 979-1638 * '''||''' .
URL: http://www.rite-online.net ******************
|
|
|
0
|
|
|
|
Reply
|
Rich
|
12/16/2004 4:43:39 AM
|
|
Hamilcar Barca <hamilcar@tld.always.invalid> writes in comp.unix.solaris:
|What's the point, if any, of the CDDL?
To provide an update to the commonly accepted Mozilla Public License to
deal with some concerns people had about the MPL. If the OSI review
goes well (which it seems to be, providing useful feedback for improving
the license further), there are a variety of projects it may be used
for, but exactly which ones is not being announced yet, to allow the
discussion to focus on the license itself, not the projects that may use
it.
--
________________________________________________________________________
Alan Coopersmith * alanc@alum.calberkeley.org * Alan.Coopersmith@Sun.COM
http://www.csua.berkeley.edu/~alanc/ * http://blogs.sun.com/alanc/
Working for, but definitely not speaking for, Sun Microsystems, Inc.
|
|
|
0
|
|
|
|
Reply
|
Alan
|
12/16/2004 7:36:09 AM
|
|
Hamilcar Barca <hamilcar@tld.always.invalid> writes in comp.unix.solaris:
|Note that Sun was under no obligation to create a new license that
|conflicted with the GPL. They did so anyway. Could this have been
|an accidental oversight?
Or could it have simply been anticipating the compatibility issue being
resolved by the GPL version 3, given Richard Stallman's statements on
patent grants in licenses and plans to update the GPL soon? (I don't
know, I wasn't involved, but it seems just as reasonable a possibility
as any claims of trying to intentionally be GPL incompatible.)
--
________________________________________________________________________
Alan Coopersmith * alanc@alum.calberkeley.org * Alan.Coopersmith@Sun.COM
http://www.csua.berkeley.edu/~alanc/ * http://blogs.sun.com/alanc/
Working for, but definitely not speaking for, Sun Microsystems, Inc.
|
|
|
0
|
|
|
|
Reply
|
Alan
|
12/16/2004 7:41:26 AM
|
|
In article <20041215232949.209$Y7@news.newsreader.com>,
Hamilcar Barca <hamilcar@tld.always.invalid> wrote:
>In article <cppn6b$aoc$1@news.cs.tu-berlin.de> (Wed, 15 Dec 2004 16:02:19
>+0000), Joerg Schilling wrote:
>
>> Well, with CDDL people like Linus have more problems with trying to dominate
>> a project.....
>
>Do you then claim that all arguments in favor of the CDDL [sic] devolve to
>ad hominem attacks?
???
It seems that you do not follow the discussions about the CDDL on the net :-(
You repeatedly read quotes like this from Linus:
- From osnews:
-----
Linus' comments are better. Sun does face a huge problem in building
up a community. The CDDL also allows linking with proprietary stuff
- like the LGPL. He says that is bad. His stance on binary drivers
would seem to contradict that stance, but whatever. He also says that
this license helps Sun retain control of the code more than the GPL
does and that it doesn't make everyone an owner as much as the GPL
does. That's just not true. In fact, because of the patent grant, it
makes people greater owners. There is no way that Sun can keep control
of stuff licenced under the CDDL. I could very easily take all their
CDDL licenced code and start a project with a licence that was
GPL-like as long as that licence included a patent grant and stuff.
-----
This nicely gives the impression on domination on Linux you get when
reading what he writes on Linux. The problem with Linus control on
Linux is that he likes to control everything even things he does not
understand, and this is a lot.
Linus did even prove this inability recently when the Linux kernel
guys did start stating that they are unable to create a usable kernel.
Since a few months, the Linux kernel people tell you that it is not
their intention that Linux kernels from ftp.kernel.org are useful
anymore. They instead point you to Linux kernels from Linux
distributions. This looks like: "Hey, we are unable to create a useful
work (as we only like to play with kernel software). If you like to
use our software try to get a version from somebody who fixes our bugs
first"...
Let me give you an example: A few months ago, people discovered a
minor possible security bug in the Linux Generic SCSI code that has
been the result of a typical beginners mistake in the code. With
Linux-2.6.8.1 (something that has been declared "stable"), Linus did
introduce an incompatible interface change in the Generic SCSI
interface. I and many other people pointed him to a way to fix the
problem without introducing kernel incompatibilities. Linus did tell
all of us that we will never change his mind.
- From eweek:
-----
Torvalds said he does not expect developers clamoring to start playing
with that source code.
-----
This is looks like a result of the fact that Linus seems to be:
- unteachable (he does not react in a useful way if
you send him mail that proves that his
claims are wrong).
- unwilling to learn (he refuses to read background
information he is pointed to after he
asks for proves that he is wrong).
N.B. I did point him more than once to the Solaris 8 community source
and to the FreeBSD source when he did get exited with his strange SCSI
kernel driver ideas. I was in hope that a contact to the reallity would
help him, but it rather seems that he only likes to look at his Linux
belly button :-(
Note that I do not like to start a CDDL discussion but only to comment on the
concerns people from the Linux camp have with Sun.....so I have no problem with
my NDA :-)
What the people from the Linux camp need to learn is that it _is_ possible
to develop a big project without being dominated by a single person.
The Solaris development from the past shows that it has always been possible
for single persons to bring up and integrate good new ideas that are not
understood by the management and not wanted in the beginning. Just look
at Bryan Cantrill and dtrace....
>> [...] but not all people behave this way.
>
>But some do:
Are you really trying to compare a large project like the Linux kernel
with a medium sized project like cdrtools that is mostly run by a single
person?
>| Warning: you are not allowed to modify or to remove the
>| Copyright and version printing code below!
>| See also GPL <A7> 2 subclause c)
>
>Have you failed to read the GPL? You cannot require, even though you so
>desire, the retention of "version printing code". You seem to have made a
>mistake in using the GPL at any time.
Please read and understand the GPL first.....
This requirement is excatly what the GPL claims for programs like cdrecord:
c) If the modified program normally reads commands interactively
when run, you must cause it, when started running for such
interactive use in the most ordinary way, to print or display an
announcement including an appropriate copyright notice and a
notice that there is no warranty (or else, saying that you provide
a warranty) and that users may redistribute the program under
these conditions, and telling the user how to view a copy of this
License. (Exception: if the Program itself is interactive but
does not normally print such an announcement, your work based on
the Program is not required to print an announcement.)
The extra comments in my code just like to tell people who did not understand
the GPL what the GPL means. Note that this is a problem with nearly _all_ major
Linux distributions. They more or less do all break the GPL. The Debian
people seem to be the only exception (after some discussions I did have with
them).
Also note that I am _definitely_ not trying to avoid real forks of my
work. The problem with Linux based companies like RedHat & SuSE is that
they are proprietary software companies that just use other peoples
code and modify it _without_ even trying to give feedback to the original
authors. Their intention is not to deploy good and unique free software
but rather to have something that differs from their competitiors. This is
something from the stone age of software companies. Well one might argue that
these companies are young and still need to learn.....
The result of the changes on cdrtools made by e.g. SuSE has _not_ been
better software but rather _broken_ software that is no longer useful
for even the basic intention of this software :-(
If you _read_ the GPL, then you will see that the GPL does not allow you
to make changes that discredit the original Author. Fact is, that
companies like SuSE did break my software but did use my name for
advertising....
Also, for each author's protection and ours, we want to make certain
that everyone understands that there is no warranty for this free
software. If the software is modified by someone else and passed on, we
want its recipients to know that what they have is not the original, so
that any problems introduced by others will not reflect on the original
authors' reputations.
This is exactly what SuSE did: they published a modified and obviously
defective version of cdrecord but let it look as it it was the original
cdrecord.
As a result, a lot of SuSE customers did send bug reports via mail to _me_
and have been angry because of the bugs introduced not my me but by SuSE.
At this point, it looks like a good idea in the current CDDL draft to
introduce paragraph 3.4 which would have allowed me to send an invoice to
SuSE to get a compensation for the work that SuSE did force me to do.
--
EMail:joerg@schily.isdn.cs.tu-berlin.de (home) J�rg Schilling D-13353 Berlin
js@cs.tu-berlin.de (uni) If you don't have iso-8859-1
schilling@fokus.fraunhofer.de (work) chars I am J"org Schilling
URL: http://www.fokus.fraunhofer.de/usr/schilling ftp://ftp.berlios.de/pub/schily
|
|
|
0
|
|
|
|
Reply
|
js
|
12/16/2004 10:38:13 AM
|
|
In comp.unix.solaris Joerg Schilling <js@cs.tu-berlin.de> wrote:
> You repeatedly read quotes like this from Linus:
none of what JS is citing is attributed to Linus (who may very well agree
with this). Perhaps he's only citing his own anonymous posting.
--
Thomas E. Dickey
http://invisible-island.net
ftp://invisible-island.net
|
|
|
0
|
|
|
|
Reply
|
Thomas
|
12/16/2004 11:39:27 AM
|
|
ptribble@hgmp.mrc.ac.uk (Peter C. Tribble) writes:
> In article <41c09e83$0$21106$e4fe514c@news.xs4all.nl>,
> Casper H.S. Dik <Casper.Dik@Sun.COM> writes:
>> ptribble@hgmp.mrc.ac.uk (Peter C. Tribble) writes:
>>
>>>(As part of OpenSolaris, I've tried looking at the scheduler, because I
>>>know there are interactions there. But I don't have the familiarity
>>>with the code or the skills yet to understand it.)
>>
>> The Solaris scheduler is fairly complicated; however, it's been an
>> O(1) scheduler (a much touted Linux 2.something feature) since about
>> Solaris 2.4.
>
> That it may be. The problem we saw was (as far as I recall - sorry Gary
> if I mess up) due to a scheduling race in the following scenario:
>
> Shell script runs through many short lived processes. When new process
> is launched, parent sleeps. When child exits, two scheduling events are
> generated - parent wakes up and tries to find a cpu to run on, and the
> cpu the child was running on is now free.
>
> Needless to say it didn't always get it right. Especially if this is a
> multiprocessor system with something else happening.
Yep, that refreshed my memory on the problem with "configure" style
scripts running *extremly* slow on Solaris when there is some low
priority background activity.
Here's a link to the old google archived usenet article; it includes a
simple test case:
http://groups.google.de/groups?selm=wy4rjf3tn9.fsf%40tools.de
> (The fix was to use
> processor sets or even a simple pbind to stop processes wandering
> around between cpus - in other words, to override some of the
> scheduling decisions.)
>
> We were seeing a slowdown of something like a factor 10 in some extreme
> cases. I mean, you would think that if you had a multiprocessor machine
> then you wouldn't be able to detect much of an impact on system
> responsiveness provided there was at least one CPU free. The reality
> was otherwise, unfortunately.
>
> Must run some tests on build 72, to see how it's changed
> recently. Things are getting much better in the last couple of Solaris
> express releases.
Well, I just re-run the test included in my old usenet article and I
don't see any changes or improvements for this particular test case,
comparing between Solaris 8 x86 and Solaris 10 b69 x86...
vfork()ing short lived subprocesses is 10x faster than fork()ing them,
in case there is some background activity on the system.
|
|
|
0
|
|
|
|
Reply
|
Juergen
|
12/16/2004 12:08:17 PM
|
|
In article <10s2svfjhvbml4c@corp.supernews.com>,
Thomas Dickey <dickey@saltmine.radix.net> wrote:
>In comp.unix.solaris Joerg Schilling <js@cs.tu-berlin.de> wrote:
>
>> You repeatedly read quotes like this from Linus:
>
>none of what JS is citing is attributed to Linus (who may very well agree
>with this). Perhaps he's only citing his own anonymous posting.
The next time you are confused, try to avoid to post.....
But hey, getting a reply from you proves that my posting is correct and
important :-)
--
EMail:joerg@schily.isdn.cs.tu-berlin.de (home) J�rg Schilling D-13353 Berlin
js@cs.tu-berlin.de (uni) If you don't have iso-8859-1
schilling@fokus.fraunhofer.de (work) chars I am J"org Schilling
URL: http://www.fokus.fraunhofer.de/usr/schilling ftp://ftp.berlios.de/pub/schily
|
|
|
0
|
|
|
|
Reply
|
js
|
12/16/2004 12:36:47 PM
|
|
In comp.unix.solaris Joerg Schilling <js@cs.tu-berlin.de> wrote:
> But hey, getting a reply from you proves that my posting is correct and
> important :-)
neither.
slightly amusing (will provide useful followups for your fellow trolls).
--
Thomas E. Dickey
http://invisible-island.net
ftp://invisible-island.net
|
|
|
0
|
|
|
|
Reply
|
Thomas
|
12/16/2004 12:48:43 PM
|
|
Dan Espen wrote:
> solaris> ldd /usr/ccs/bin/make
> libintl.so.1 => /usr/lib/libintl.so.1
This is a bug. There's no need to link against /usr/lib/libintl.so.1
(it's just a filter and the implementation is in libc).
--
.-. .-. Yes, I am an agent of Satan, but my duties are largely
(_ \ / _) ceremonial.
|
| dave@fly.srk.fer.hr
|
|
|
0
|
|
|
|
Reply
|
Drazen
|
12/16/2004 1:24:05 PM
|
|
Alan Coopersmith wrote:
> Eh, not sure how useful a comparison that is - if you link with -z
> lazyload, many of those will never get loaded at startup (and some,
> never at all). The linker guys have done some amazing things to
> speed up loads when the right options are used.
Whether lazy loading is useful depends on the way the application and the
libraries are written. If something calls foolib_init() and then never
uses anything from foolib, the library will be loaded anyway, symbol
resolution will be somewhat slower and the binary will be somewhat larger.
You get the same effect if something's referencing the library data
directly, eg. gtk_major_version.
--
.-. .-. Yes, I am an agent of Satan, but my duties are largely
(_ \ / _) ceremonial.
|
| dave@fly.srk.fer.hr
|
|
|
0
|
|
|
|
Reply
|
Drazen
|
12/16/2004 1:31:44 PM
|
|
In article <20041215232949.209$Y7@news.newsreader.com>,
Hamilcar Barca <hamilcar@tld.always.invalid> writes:
[...]
> The problem is that the CDDL [sic] tries to reduce the freedoms for all
> granted by the GPL, such that Sun may maintain sufficient control over its
> code to prevent it from being used in certain other products -- notably
> the Linux kernel.
>
[...]
The problem is that not everyone agrees that what GPL does and
{BSD,CDDL,...} doesn't do makes GPL freer.
The problem is also that GPL can be an obstacle to using free and
proprietary software together to meet the end user's needs. Or
have we forgotten about them altogether?
--
mailto:rlhamil@smart.net http://www.smart.net/~rlhamil
Lasik/PRK theme music:
"In the Hall of the Mountain King", from "Peer Gynt"
|
|
|
0
|
|
|
|
Reply
|
Richard
|
12/16/2004 1:34:33 PM
|
|
Joerg Schilling wrote:
> The result of the changes on cdrtools made by e.g. SuSE has _not_ been
> better software but rather _broken_ software that is no longer useful
> for even the basic intention of this software :-(
You're in Germany, SuSE is based in Germany. I was under impression that
the German copyright law has integrity rights under the author's moral
rights. If that's the case you're perfectly capable of stopping the
distribution of such modifications.
> As a result, a lot of SuSE customers did send bug reports via mail to _me_
> and have been angry because of the bugs introduced not my me but by SuSE.
So you can even prove the damage to the reputation.
--
.-. .-. Yes, I am an agent of Satan, but my duties are largely
(_ \ / _) ceremonial.
|
| dave@fly.srk.fer.hr
|
|
|
0
|
|
|
|
Reply
|
Drazen
|
12/16/2004 1:54:48 PM
|
|
Drazen Kacar <dave@fly.srk.fer.hr> schrieb:
> Joerg Schilling wrote:
>
>> The result of the changes on cdrtools made by e.g. SuSE has _not_ been
>> better software but rather _broken_ software that is no longer useful
>> for even the basic intention of this software :-(
>
> You're in Germany, SuSE is based in Germany. I was under impression that
> the German copyright law has integrity rights under the author's moral
> rights. If that's the case you're perfectly capable of stopping the
> distribution of such modifications.
Joerg is not the owner of all the code basing cdrtools.
So he can't change the current license of cdrtools.
cu
ulf
--
Ulf Volmer
u.volmer@u-v.de
|
|
|
0
|
|
|
|
Reply
|
Ulf
|
12/16/2004 2:23:21 PM
|
|
Ulf Volmer wrote:
> Drazen Kacar <dave@fly.srk.fer.hr> schrieb:
> > Joerg Schilling wrote:
> >
> >> The result of the changes on cdrtools made by e.g. SuSE has _not_ been
> >> better software but rather _broken_ software that is no longer useful
> >> for even the basic intention of this software :-(
> >
> > You're in Germany, SuSE is based in Germany. I was under impression that
> > the German copyright law has integrity rights under the author's moral
> > rights. If that's the case you're perfectly capable of stopping the
> > distribution of such modifications.
>
> Joerg is not the owner of all the code basing cdrtools.
>
> So he can't change the current license of cdrtools.
Moral rights are usually inalienable, ie. they don't depend on the license
or the contract or other legal business. They are granted by the copyright
law (usually not transferable) and in some jurisdictions can't even be
vaiwed off.
--
.-. .-. Yes, I am an agent of Satan, but my duties are largely
(_ \ / _) ceremonial.
|
| dave@fly.srk.fer.hr
|
|
|
0
|
|
|
|
Reply
|
Drazen
|
12/16/2004 3:11:38 PM
|
|
In article <slrncs36ip.fmf.u.volmer@xerxes.is-a-geek.net>,
Ulf Volmer <u.volmer@BMer.de> wrote:
>> You're in Germany, SuSE is based in Germany. I was under impression that
>> the German copyright law has integrity rights under the author's moral
>> rights. If that's the case you're perfectly capable of stopping the
>> distribution of such modifications.
This is not as easy as it may look...
>Joerg is not the owner of all the code basing cdrtools.
>
>So he can't change the current license of cdrtools.
I am not the owner of enough code from mkisofs and I am not the owner
of cdda2wav. I am however the owner of cdrecord. While I most likely would
get the permission for cdda2wav from Heiko, mkisofs is a problem...
--
EMail:joerg@schily.isdn.cs.tu-berlin.de (home) J�rg Schilling D-13353 Berlin
js@cs.tu-berlin.de (uni) If you don't have iso-8859-1
schilling@fokus.fraunhofer.de (work) chars I am J"org Schilling
URL: http://www.fokus.fraunhofer.de/usr/schilling ftp://ftp.berlios.de/pub/schily
|
|
|
0
|
|
|
|
Reply
|
js
|
12/16/2004 3:49:45 PM
|
|
Richard L. Hamilton wrote:
> The problem is that not everyone agrees that what GPL does and
> {BSD,CDDL,...} doesn't do makes GPL freer.
> The problem is also that GPL can be an obstacle to using free and
> proprietary software together to meet the end user's needs. Or have
> we forgotten about them altogether?
PalmSource to build Palm OS on Linux
Tony Smith Dec 08 2004
"PalmSource is to create a new version of the Palm OS with Linux at its
core, the company said today after announcing a plan to buy Chinese
phone software company China MobileSoft (CMS)."
"Like Apple with Mac OS X, PalmSource will keep all the top-layer code
proprietary, but it will release any changes it makes to the underlying
Linux code - for faster boot times and battery life preservation
systems, for example - available to the open source community."
http://www.theregister.co.uk/2004/12/08/palmsource_linux/
IBM does Linux-only dance on Power
Ashlee Vance Sep 13 2004 Chicago
"IBM continues to refine its Linux server strategy, announcing today a
new set of kit that runs the open source OS only on its Power5 processor
... IBM will offer both Red Hat and SuSE's enterprise Linux operating
systems on the new box."
http://www.theregister.co.uk/2004/09/13/ibm_linuxonly_powerbox/
"SUSE Linux Desktop 1.0 was chosen by Sun to be part of the Java Desktop
System primarily to act as a platform for GNOME development which Sun is
working on. In addition, SUSE is a premier operating environment with
significant customer base and development efforts."
http://wwws.sun.com/software/javadesktopsystem/faq.html#4q3
"Xandros uses a hybrid open sourcecode with proprietary features that
provide significantly better security than other popular operating systems."
http://www.xandros.com/about/corporatebackground.html
"Xandros, the leading developer of easy-to-use Linux solutions, today
announced that Wal-Mart.com is carrying a fully loaded Linux desktop
computer with the pre-installed Xandros Desktop Operating System (OS)
for only $199.98."
"Built by Microtel and available at Wal-Mart's online store,* the new
desktop PC offers a complete suite of pre-installed software for home,
school, and small office desktop use"
http://www.linuxelectrons.com/article.php/20041209213249573
"The combined Linux market - desktops, servers, and packaged software
running on Linux - will exceed $35 billion by 2008, with $10 billion of
that coming from Linux PC sales, says an IDC Linux market share and
forecast study announced yesterday by OSDL, the Open Source Development
Labs."
http://p2pnet.net/story/3308
"HP first major PC maker to launch Linux-based notebook"
http://www.sfgate.com/cgi-bin/article.cgi?file=/news/archive/2004/08/03/financial1657EDT0259.DTL
See also ...
http://www.novell.com
http://www.redhat.com
http://www.xandros.com
http://www.hp.com
http://www.linspire.com
|
|
|
0
|
|
|
|
Reply
|
Daeron
|
12/16/2004 7:22:04 PM
|
|
In article <cproil$lvf$1@news.cs.tu-berlin.de> (Thu, 16 Dec 2004 10:38:13
+0000), Joerg Schilling wrote:
> In article <20041215232949.209$Y7@news.newsreader.com>,
> Hamilcar Barca <hamilcar@tld.always.invalid> wrote:
>>
> It seems that you do not follow the discussions about the CDDL on the net :-(
It's certain you have no evidence for your claims
> You repeatedly read quotes like this from Linus:
>
> [snip 9 paragraphs without a single quotation from Linus Torvalds]
You imagine quotes where there are absolutely none.
>>But some do:
>
> Are you really trying to compare a large project like the Linux kernel
> with a medium sized project like cdrtools that is mostly run by a single
> person?
Are you really denying that you're guilty of the charges you leveled
against Linux, when your admission against interest is sufficient proof?
>>| Warning: you are not allowed to modify or to remove the
>>| Copyright and version printing code below!
>>| See also GPL <A7> 2 subclause c)
>>
>>Have you failed to read the GPL? You cannot require, even though you so
>>desire, the retention of "version printing code". You seem to have made a
>>mistake in using the GPL at any time.
>
> Please read and understand the GPL first.....
I have read and understood the GPL. Your quotation proves you wrong
beyond any doubt. No section, subsection, paragraph, and/or phrase denies
licensees from removing advertising clauses.
> This requirement is excatly what the GPL claims for programs like cdrecord:
>
> c) If the modified program normally reads commands interactively
> when run, you must cause it, when started running for such
> interactive use in the most ordinary way, to print or display an
> announcement including an appropriate copyright notice and a
> notice that there is no warranty (or else, saying that you provide
> a warranty) and that users may redistribute the program under
> these conditions, and telling the user how to view a copy of this
> License. (Exception: if the Program itself is interactive but
> does not normally print such an announcement, your work based on
> the Program is not required to print an announcement.)
Which of these words requires GPL licensees to maintain your advertising
clause? Not a single one.
> The extra comments in my code just like to tell people who did not understand
> the GPL what the GPL means.
No, they don't. They attempt to misinform licensees about their rights
and obligations under the GPL.
> Note that this is a problem with nearly _all_ major
> Linux distributions.
Note that the misinterpretations of others do not excuse yours. Clearly,
you are able to see your advertising requirement is a violation of the GPL.
> The result of the changes on cdrtools made by e.g. SuSE has _not_ been
> better software but rather _broken_ software that is no longer useful
> for even the basic intention of this software :-(
That's unfortunate, but it's a risk when you use the GPL.
> If you _read_ the GPL, then you will see that the GPL does not allow you
> to make changes that discredit the original Author.
It certainly does not. It requires only the original author's copyright
notice be preserved. If I add a line to cdrecord that says
printf("Joerg Schilling produces inferior software\n") ;
I have violated no portion of the GPL. NOTE: I use cdrecord and it is not
inferior software.
> Fact is, that companies like SuSE did break my software [...]
This may be so. It in no way violates the GPL.
> [...] but did use my name for advertising....
This may be so. It in no way violates the GPL.
> Also, for each author's protection and ours, we want to make certain
> that everyone understands that there is no warranty for this free
> software. If the software is modified by someone else and passed on, we
> want its recipients to know that what they have is not the original, so
> that any problems introduced by others will not reflect on the original
> authors' reputations.
>
> This is exactly what SuSE did: they published a modified and obviously
> defective version of cdrecord but let it look as it it was the original
> cdrecord.
This may be so. If it is, it's a violation of the GPL.
> At this point, it looks like a good idea in the current CDDL draft to
> introduce paragraph 3.4 which would have allowed me to send an invoice to
> SuSE to get a compensation for the work that SuSE did force me to do.
SuSE did not force you to do any work -- none at all. If you don't want
to support software you released under the GPL, you are free to do so.
Whether the defects are yours or SuSE's, you undertook the work on your
own initiative.
The CDDL is just a "red herring".
--
"It is unthinkable that the European Competition authority could fail to
deal with anti-competitive practices in the EU with such huge effects
simply because [Microsoft] has its headquarters in the U.S."
-- Mario Monti. EU Competition Commissioner.
|
|
|
0
|
|
|
|
Reply
|
Hamilcar
|
12/16/2004 8:56:40 PM
|
|
In article <10s33n9oosu16e6@corp.supernews.com> (Thu, 16 Dec 2004 13:34:33
+0000), Richard L. Hamilton wrote:
> The problem is also that GPL can be an obstacle to using free and
> proprietary software together to meet the end user's needs.
It does not. It prohibits the mixing of proprietary and GPL code only if
they are "distributed". End users are free to mix, match, hack, break,
and/or innovate [sic] to their heart's content.
> Or have we forgotten about them altogether?
Who is "we"?
--
"As the majority of hobbyists must be aware, most of you
steal your software."
-- William Henry Gates III. An Open Letter to Hobbyists. 3 Feb 1976.
|
|
|
0
|
|
|
|
Reply
|
Hamilcar
|
12/16/2004 8:58:54 PM
|
|
In article <Pine.SOL.4.58.0412152035560.15621@zaphod> (Thu, 16 Dec 2004
04:43:39 +0000), Rich Teer wrote:
> On Wed, 15 Dec 2004, Hamilcar Barca wrote:
>
>> What's the point, if any, of the CDDL? Is Sun just testing the waters
>> to see how much negative feedback is generated? Why do they go to the
>> expense of creating a new "open source" [sic] license if they don't intend
>> to use it?
>
> Sun intend to use CDDL. I'm just saying that they haven't commited to
> using it for Open Solaris (nor have they publically said that they aren't).
Let me rephrase a portion of my article:
* What's the point, if any, of the CDDL?
* Is Sun just testing the waters to see how much negative feedback is
generated?
>> Who says Open [sic] Solaris won't use the CDDL?
>
> Sun haven't publically commited to ANY license for Open Solaris yet. They've
> only stated that it will be OSI approved.
This claim from Sun seems to have no merit. Why yet another license?
What if Sun change their mind?
> Given that I'm a member of the Open Solaris pilot project I'm privvy
> to more info than I can say here. And no, I'm not about to break
> the terms of the NDA I signed!
You seem to be claiming to possess facts not in evidence. These are not
useful.
--
"I've hired quite a few people from the Linux open source world to run
a research and development center. Then we bring consultants from the
Linux open source space to help us build solutions."
-- Martin Taylor. Platform Strategy. Microsoft Corporation.
|
|
|
0
|
|
|
|
Reply
|
Hamilcar
|
12/16/2004 9:04:08 PM
|
|
In article <cprdt9$mln$1@agate.berkeley.edu> (Thu, 16 Dec 2004 07:36:09
+0000), Alan Coopersmith wrote:
> Hamilcar Barca <hamilcar@tld.always.invalid> writes in comp.unix.solaris:
> |What's the point, if any, of the CDDL?
>
> To provide an update to the commonly accepted Mozilla Public License to
> deal with some concerns people had about the MPL.
Why is this useful?
> If the OSI review goes well
This OSI?
http://www.opensource.org/docs/board.php
Eric S. Raymond (President) (PGP key)
Eric designed the language and marketing tactics around which the
OSI was formed. He and Bruce Perens co-founded the organization. Eric
lives in Malvern, Pennsylvania, USA.
> (which it seems to be, providing useful
> feedback for improving the license further), [...]
This kind of feedback?
"Sun can vapor on about voting and committees all it wants, but [...]
JCP is still a single point of control, [and] the Java reference imple-
mentation and class libraries are under a proprietary license [...]"
-- Eric S. Raymond
"Rather than blowing smoke about the bazaar model, I think [Sun's]
Mr. Schwartz's time would be better spent explaining why he thinks
those reasons don't apply to Java [...]"
-- Eric S. Raymond
"Any time [Sun Microsystems tries] to use my work to justify retaining
proprietary control or argue that Linux is somehow less open than Java,
that's either culpable stupidity or dishonesty [...]"
-- Eric S. Raymond
Sun's new CDDL seems to be nothing more than a marketing announcement.
--
"Ironically, Microsoft's efforts to deny interoperability of Windows with
legitimate non-Microsoft applications have created an environment in which
Microsoft's programs interoperate efficiently only with Internet viruses."
-- Daniel Geer.
|
|
|
0
|
|
|
|
Reply
|
Hamilcar
|
12/16/2004 9:09:42 PM
|
|
In article <20041216155625.499$Se@news.newsreader.com>,
Hamilcar Barca <hamilcar@tld.always.invalid> wrote:
>In article <cproil$lvf$1@news.cs.tu-berlin.de> (Thu, 16 Dec 2004 10:38:13
>+0000), Joerg Schilling wrote:
>
>> In article <20041215232949.209$Y7@news.newsreader.com>,
>> Hamilcar Barca <hamilcar@tld.always.invalid> wrote:
>>>
>> It seems that you do not follow the discussions about the CDDL on the net :-(
>
>It's certain you have no evidence for your claims
Looks like a typical reply from an uninformed person :-(
If you would have a minimum amount of knowledge on how to use google,
it yould take you a few seconds to find the sources I did quote by just feeding
a few words from the qwuoted text into google....
http://www.osnews.com/comment.php?news_id=9074
http://www.eweek.com/article2/0,1759,1739000,00.asp?kc=EWRSS03119TX1K0000594
It may be intersting for you to see that the sources I quote are
from the same anti Sun league as you :-)
But people like you are not interested in a useful discussion if they find
themselves on the side of the loosers.
The rest of your useless posting has been disposed to save the environment....
--
EMail:joerg@schily.isdn.cs.tu-berlin.de (home) J�rg Schilling D-13353 Berlin
js@cs.tu-berlin.de (uni) If you don't have iso-8859-1
schilling@fokus.fraunhofer.de (work) chars I am J"org Schilling
URL: http://www.fokus.fraunhofer.de/usr/schilling ftp://ftp.berlios.de/pub/schily
|
|
|
0
|
|
|
|
Reply
|
js
|
12/16/2004 9:44:19 PM
|
|
js@cs.tu-berlin.de (Joerg Schilling) writes:
>
> The rest of your useless posting has been disposed to save the environment....
I think the whole discussion should be disposed to save the environment...
I am kill-filling this subject. It's getting silly.
Dragan
--
Dragan Cvetkovic,
To be or not to be is true. G. Boole No it isn't. L. E. J. Brouwer
!!! Sender/From address is bogus. Use reply-to one !!!
|
|
|
0
|
|
|
|
Reply
|
Dragan
|
12/16/2004 9:47:22 PM
|
|
In article <20041216160354.664$1F@news.newsreader.com>,
Hamilcar Barca <hamilcar@tld.always.invalid> wrote:
>> Sun intend to use CDDL. I'm just saying that they haven't commited to
>> using it for Open Solaris (nor have they publically said that they aren't).
>
>Let me rephrase a portion of my article:
>
>* What's the point, if any, of the CDDL?
>
>* Is Sun just testing the waters to see how much negative feedback is
> generated?
Sun is interested in constructive feedback, something that you seem to be
unable to give :-(
--
EMail:joerg@schily.isdn.cs.tu-berlin.de (home) J�rg Schilling D-13353 Berlin
js@cs.tu-berlin.de (uni) If you don't have iso-8859-1
schilling@fokus.fraunhofer.de (work) chars I am J"org Schilling
URL: http://www.fokus.fraunhofer.de/usr/schilling ftp://ftp.berlios.de/pub/schily
|
|
|
0
|
|
|
|
Reply
|
js
|
12/16/2004 9:50:26 PM
|
|
In article <20041216160927.922$y9@news.newsreader.com>,
Hamilcar Barca <hamilcar@tld.always.invalid> wrote:
>
>This kind of feedback?
>
> "Sun can vapor on about voting and committees all it wants, but [...]
> JCP is still a single point of control, [and] the Java reference imple-
> mentation and class libraries are under a proprietary license [...]"
> -- Eric S. Raymond
>
> "Rather than blowing smoke about the bazaar model, I think [Sun's]
> Mr. Schwartz's time would be better spent explaining why he thinks
> those reasons don't apply to Java [...]"
> -- Eric S. Raymond
Looks like the reply of a mugger that did just receive wristwatch
as a gift and then becomes angry for not receiving the tie as a throw in....
--
EMail:joerg@schily.isdn.cs.tu-berlin.de (home) J�rg Schilling D-13353 Berlin
js@cs.tu-berlin.de (uni) If you don't have iso-8859-1
schilling@fokus.fraunhofer.de (work) chars I am J"org Schilling
URL: http://www.fokus.fraunhofer.de/usr/schilling ftp://ftp.berlios.de/pub/schily
|
|
|
0
|
|
|
|
Reply
|
js
|
12/16/2004 9:55:35 PM
|
|
In article <lmpt193ogl.fsf@privacy.net>,
Dragan Cvetkovic <dragan_usenet@gmx.net> wrote:
>js@cs.tu-berlin.de (Joerg Schilling) writes:
>
>>
>> The rest of your useless posting has been disposed to save the environment....
>
>I think the whole discussion should be disposed to save the environment...
Most people did write useful postings during the last days.
My impression is that the problems are caused by a few people from cola that
are just unable write useful things. Something similar as you see on LKML.
There are only about 3-5 people on LKML who repeatedly write offensive mails
but these people emboss the impresion that most people have from LKML.
--
EMail:joerg@schily.isdn.cs.tu-berlin.de (home) J�rg Schilling D-13353 Berlin
js@cs.tu-berlin.de (uni) If you don't have iso-8859-1
schilling@fokus.fraunhofer.de (work) chars I am J"org Schilling
URL: http://www.fokus.fraunhofer.de/usr/schilling ftp://ftp.berlios.de/pub/schily
|
|
|
0
|
|
|
|
Reply
|
js
|
12/16/2004 10:03:22 PM
|
|
In article <20041216160927.922$y9@news.newsreader.com>,
Hamilcar Barca <hamilcar@tld.always.invalid> writes:
> In article <cprdt9$mln$1@agate.berkeley.edu> (Thu, 16 Dec 2004 07:36:09
> +0000), Alan Coopersmith wrote:
>
>> Hamilcar Barca <hamilcar@tld.always.invalid> writes in comp.unix.solaris:
>> |What's the point, if any, of the CDDL?
>>
>> To provide an update to the commonly accepted Mozilla Public License to
>> deal with some concerns people had about the MPL.
>
> Why is this useful?
>
>> If the OSI review goes well
>
> This OSI?
>
> http://www.opensource.org/docs/board.php
>
> Eric S. Raymond (President) (PGP key)
> Eric designed the language and marketing tactics around which the
> OSI was formed. He and Bruce Perens co-founded the organization. Eric
> lives in Malvern, Pennsylvania, USA.
>
No, the one with Oscar Goldman, Rudy Wells, Col. Steve Austen, Jaimie
Summers, ...
What a maroon!
>> (which it seems to be, providing useful
>> feedback for improving the license further), [...]
>
> This kind of feedback?
>
> "Sun can vapor on about voting and committees all it wants, but [...]
> JCP is still a single point of control, [and] the Java reference imple-
> mentation and class libraries are under a proprietary license [...]"
> -- Eric S. Raymond
>
> "Rather than blowing smoke about the bazaar model, I think [Sun's]
> Mr. Schwartz's time would be better spent explaining why he thinks
> those reasons don't apply to Java [...]"
> -- Eric S. Raymond
>
> "Any time [Sun Microsystems tries] to use my work to justify retaining
> proprietary control or argue that Linux is somehow less open than Java,
> that's either culpable stupidity or dishonesty [...]"
> -- Eric S. Raymond
>
> Sun's new CDDL seems to be nothing more than a marketing announcement.
>
So, you think Sun should simply surrender control of everything they create
or surrender any claim to being open?
Did you grow up commie, or is it something you went out of your way to
become? That's not a fair position, is it? And yours is?
--
mailto:rlhamil@smart.net http://www.smart.net/~rlhamil
Lasik/PRK theme music:
"In the Hall of the Mountain King", from "Peer Gynt"
|
|
|
0
|
|
|
|
Reply
|
Richard
|
12/16/2004 11:14:19 PM
|
|
In article <32e5iuF3m8349U1@individual.net>,
Daeron <daeron@demon.com> writes:
> Richard L. Hamilton wrote:
>
>> The problem is that not everyone agrees that what GPL does and
>> {BSD,CDDL,...} doesn't do makes GPL freer.
>
>> The problem is also that GPL can be an obstacle to using free and
>> proprietary software together to meet the end user's needs. Or have
>> we forgotten about them altogether?
[counter-examples snipped]
Good counter-examples. But I did say "can be", not "always is".
My impression is that one of the things Sun may not love about GPL is
that it could complicate some business relationships with other companies.
FWIW, I suppose Sun _could_ multi-license: say your choice of
GPL or {BSD, CDDL, whatever} for unsupported use, or a more typical
commercial license for supported use. Unless that would cause them
serious problems too, I suppose that would at least cut the ignorant
whiners off at the knees; and it sure would be fun to see them hobbling
around on stumps!
--
mailto:rlhamil@smart.net http://www.smart.net/~rlhamil
Lasik/PRK theme music:
"In the Hall of the Mountain King", from "Peer Gynt"
|
|
|
0
|
|
|
|
Reply
|
Richard
|
12/16/2004 11:24:14 PM
|
|
In article <20041216155840.132$KA@news.newsreader.com>,
Hamilcar Barca <hamilcar@tld.always.invalid> writes:
> In article <10s33n9oosu16e6@corp.supernews.com> (Thu, 16 Dec 2004 13:34:33
> +0000), Richard L. Hamilton wrote:
>
>> The problem is also that GPL can be an obstacle to using free and
>> proprietary software together to meet the end user's needs.
>
> It does not. It prohibits the mixing of proprietary and GPL code only if
> they are "distributed". End users are free to mix, match, hack, break,
> and/or innovate [sic] to their heart's content.
And I thought that other discussion already established that most end users
don't _want_ to do anything more than install and forget.
>
>> Or have we forgotten about them altogether?
>
> Who is "we"?
That would be those who are wrapped up in ideology and keep ignoring
the strictly practical.
Even within the long-time open-source community, not everyone agrees on
GPL!
--
mailto:rlhamil@smart.net http://www.smart.net/~rlhamil
Lasik/PRK theme music:
"In the Hall of the Mountain King", from "Peer Gynt"
|
|
|
0
|
|
|
|
Reply
|
Richard
|
12/16/2004 11:27:15 PM
|
|
begin Richard L. Hamilton wrote:
> In article <20041216155840.132$KA@news.newsreader.com>,
> Hamilcar Barca <hamilcar@tld.always.invalid> writes:
>> In article <10s33n9oosu16e6@corp.supernews.com> (Thu, 16 Dec 2004
>> 13:34:33 +0000), Richard L. Hamilton wrote:
>>
>>> The problem is also that GPL can be an obstacle to using free and
>>> proprietary software together to meet the end user's needs.
>>
>> It does not. It prohibits the mixing of proprietary and GPL code only if
>> they are "distributed". End users are free to mix, match, hack, break,
>> and/or innovate [sic] to their heart's content.
>
> And I thought that other discussion already established that most end
> users don't _want_ to do anything more than install and forget.
>
The discussion is not about what "most end users" want
Other discussions also established that Solaris trolls are braindead,
impotent and generally disliked. Just like wintrolls
>>
>>> Or have we forgotten about them altogether?
>>
>> Who is "we"?
>
>
> That would be those who are wrapped up in ideology and keep ignoring
> the strictly practical.
>
> Even within the long-time open-source community, not everyone agrees on
> GPL!
>
So what? Even GPL proponents (like me) see that it does not fit every
purpose. Same is true for other licenses.
--
Windows isn't unstable. It's spontaneous.
|
|
|
0
|
|
|
|
Reply
|
Peter
|
12/16/2004 11:48:18 PM
|
|
In article <10s45mbt8u3gl4b@corp.supernews.com> (Thu, 16 Dec 2004 23:14:19
+0000), Richard L. Hamilton wrote:
> In article <20041216160927.922$y9@news.newsreader.com>,
> Hamilcar Barca <hamilcar@tld.always.invalid> writes:
>>
>> Sun's new CDDL seems to be nothing more than a marketing announcement.
>
> So, you think Sun should simply surrender control of everything they create
> or surrender any claim to being open?
No. I think Sun -- specifically including Jonathan Schwartz -- should
stop spewing their hypocritical marketing drivel. "Open standards are
better than open source", "Linux is Red Hat is Linux", "Java Community
[sic] Process", "Open [sic] Solaris", and "C[sic]DDL" represent the sort
of vacuous posturing for which Microsoft is noted.
> Did you grow up commie, or is it something you went out of your way to
> become?
"Commie" might have been a barbed insult in the 1950s but its use by a
C[sic]DDL advocate simply demonstrates its uselessness.
> That's not a fair position, is it?
Of course it isn't, but you took it anyway.
--
"Ironically, Microsoft's efforts to deny interoperability of Windows with
legitimate non-Microsoft applications have created an environment in which
Microsoft's programs interoperate efficiently only with Internet viruses."
-- Daniel Geer.
|
|
|
0
|
|
|
|
Reply
|
Hamilcar
|
12/17/2004 12:11:24 AM
|
|
In article <10s46ej8iv97i04@corp.supernews.com> (Thu, 16 Dec 2004 23:27:15
+0000), Richard L. Hamilton wrote:
> In article <20041216155840.132$KA@news.newsreader.com>,
> Hamilcar Barca <hamilcar@tld.always.invalid> writes:
>> In article <10s33n9oosu16e6@corp.supernews.com> (Thu, 16 Dec 2004 13:34:33
>> +0000), Richard L. Hamilton wrote:
>>
>>> The problem is also that GPL can be an obstacle to using free and
>>> proprietary software together to meet the end user's needs.
>>
>> It does not. It prohibits the mixing of proprietary and GPL code only if
>> they are "distributed". End users are free to mix, match, hack, break,
>> and/or innovate [sic] to their heart's content.
>
> And I thought that other discussion already established that most end users
> don't _want_ to do anything more than install and forget.
If this is the case, then "that other discussion already established"
that the GPL is absolutely no barrier to mixing GPL and proprietary
software, as "most end users don't _want_ to do anything more than install
and forget", and thus are not involved in (re-)distribution.
> Even within the long-time open-source community, not everyone agrees on
> GPL!
There's no need for everyone to agree that the GPL ends world hunger.
It's useful on the other hand to dispel misapprehensions such as "[the]
GPL can be an obstacle to using free and proprietary software together to
meet the end user's needs", misapprehensions unfounded in reality.
--
"I've hired quite a few people from the Linux open source world to run
a research and development center. Then we bring consultants from the
Linux open source space to help us build solutions."
-- Martin Taylor. Platform Strategy. Microsoft Corporation.
|
|
|
0
|
|
|
|
Reply
|
Hamilcar
|
12/17/2004 12:11:27 AM
|
|
On Thu, 16 Dec 2004, Hamilcar Barca wrote:
> No. I think Sun -- specifically including Jonathan Schwartz -- should
> stop spewing their hypocritical marketing drivel. "Open standards are
Open standards ARE better than open source. If the closed standard
that an open source project implements changes, that project is screwed
until they can reverse-engineer the new standard. For example, suppose
M$ decides to change the SMB protocol, so that it doesn't have the faintest
resembalence to the protocol that Samba currently implements. Further
suppose that they chose not to document or open that "standard". How long
would it be before the Samba project could interact with the new SMB
standard? Oh, and just to make it harder, M$ decides to encrypt all the
protocol traffic, so reverse-engineering it by sniffing the network is
made difficult. People like you would say "what's the problem? Samba is
open source, so this problem will be easy to fix!"...
When a standard is open, multiple implementations of it can compete, each
on their own merits.
> better than open source", "Linux is Red Hat is Linux", "Java Community
You're repeating your Linux lover's diatribe. Schwartz has never said
(though he has often been accused of it) that Linux = Red Hat. To quote
this page: http://blogs.sun.com/roller/page/jonathan/20040901#innovation_is_not_a_threat
They are a problem for Red Hat. And Red Hat is not linux, despite what
they say, and despite what the media (and IBM's ads) seem to conflate.
I'll say that again for the hard of reading: "And Red Hat is not linux..."
Now, what part of that sentence do you understand as reading "Red Hat is
Linux"?
Now, run along little boy. Your daddy wants his computer back,
and don't forget to wipe your drool off the keyboard before you
go.
--
Rich Teer, SCNA, SCSA, author of "Solaris Systems Programming"
. * * . * .* .
. * . .*
President, * . . /\ ( . . *
Rite Online Inc. . . / .\ . * .
.*. / * \ . .
. /* o \ .
Voice: +1 (250) 979-1638 * '''||''' .
URL: http://www.rite-online.net ******************
|
|
|
0
|
|
|
|
Reply
|
Rich
|
12/17/2004 12:56:46 AM
|
|
On Thu, 16 Dec 2004, Hamilcar Barca wrote:
> It's useful on the other hand to dispel misapprehensions such as "[the]
> GPL can be an obstacle to using free and proprietary software together to
> meet the end user's needs", misapprehensions unfounded in reality.
Hmm. I am an end user, and I would like Sun to distribute Donald Becker's
network device drivers with Solaris. But wait! Becker claims that the
GPL forbids the linking of his drivers with a closed-source kernel (in this
case, Solaris), and I don't want to rummage around trying to find his
drivers and get them to work. I want to install and forget.
Looks like in this scenario the GPL is indeed "an obstacle to using free and
proprietary software together to meet the end user's needs".
--
Rich Teer, SCNA, SCSA, author of "Solaris Systems Programming"
. * * . * .* .
. * . .*
President, * . . /\ ( . . *
Rite Online Inc. . . / .\ . * .
.*. / * \ . .
. /* o \ .
Voice: +1 (250) 979-1638 * '''||''' .
URL: http://www.rite-online.net ******************
|
|
|
0
|
|
|
|
Reply
|
Rich
|
12/17/2004 1:01:44 AM
|
|
In article <Pine.SOL.4.58.0412161632510.15621@zaphod> (Fri, 17 Dec 2004
00:56:46 +0000), Rich Teer wrote:
> On Thu, 16 Dec 2004, Hamilcar Barca wrote:
>
>> No. I think Sun -- specifically including Jonathan Schwartz -- should
>> stop spewing their hypocritical marketing drivel. "Open standards are
>
> Open standards ARE better than open source. If the closed standard
> that an open source project implements changes, that project is screwed
> until they can reverse-engineer the new standard.
If the source code was freely available, it documents the actual protocol,
as implemented. The bigger problem is the closed source, compounded by
the closed protocol.
>> better than open source", "Linux is Red Hat is Linux", "Java Community
>
> You're repeating your Linux lover's diatribe.
I recall now, as you point out, that Schwartz never said Linux is Red Hat
(and vice versa). I do believe he meant to imply it, but that's
immaterial. He didn't say it.
Schwartz is still full of lies, which you accidentally failed to quote.
Here's another:
http://blogs.sun.com/roller/page/jonathan/20040818
Red Hat's managed to dupe the linux community with their proprietary
distro [...] Not exactly a pro-open stance to take.
I assume that when Mr. Schwartz writes "proprietary", he mean to write
"100% licensed under the GPL". How non-proprietary is Solaris? It's
been 100% closed since the day Sun (legally) combined BSD and System V.
How non-proprietary is Java? Pro-open means having to license Java
protocols, and having implementation requirements forced on the developer?
How does purveying a Java Community [sic] Process under the control of Sun
amount to "pro-open" [sic]?
Schwartz is a pure, unadulterated hypocrite -- as is Bill Gates. What's
the difference?
> Now, run along little boy. Your daddy wants his computer back,
> and don't forget to wipe your drool off the keyboard before you
> go.
That's even more juvenile as the "commie" insult. The Sun worshipers are
becoming as obsessed as the Steve Jobs personality cult.
--
"...SCO and Microsoft actually think that Linux customers are so dumb to
take [EV1.com] (with direct ties to Microsoft Corporation) as some kind
of suggestion that the SCO extortion money should be paid."
-- Lewis A. Mettler, Esq. http://www.lamlaw.com/
|
|
|
0
|
|
|
|
Reply
|
Hamilcar
|
12/17/2004 3:31:44 AM
|
|
In article <Pine.SOL.4.58.0412161656511.15621@zaphod> (Fri, 17 Dec 2004
01:01:44 +0000), Rich Teer wrote:
> On Thu, 16 Dec 2004, Hamilcar Barca wrote:
>
> Hmm. I am an end user, and I would like Sun to distribute Donald Becker's
> network device drivers with Solaris.
Then do it.
> But wait!
No!
> Becker claims that the GPL forbids the linking of his drivers with a
> closed-source kernel (in this case, Solaris)
Do you have some evidence to support this claim? Even if you do, the
claim is irrelevant. When you stop ranting and actually read the GPL,
you'll find that it does not in any way prohibit linking GPL code with
proprietary code -- it prohibits only the distribution of such a merged
product.
> and I don't want to rummage around trying to find his
> drivers and get them to work. I want to install and forget.
I guess you're out of luck.
> Looks like in this scenario the GPL is indeed "an obstacle to using free and
> proprietary software together to meet the end user's needs".
It's clear only that your claims about the GPL are unfounded, and that
in any case your conclusions are entirely false.
|
|
|
0
|
|
|
|
Reply
|
Hamilcar
|
12/17/2004 3:35:37 AM
|
|
Freeride wrote:
> Rich Teer wrote:
>
>> On Mon, 13 Dec 2004, Eltee wrote:
>>
>>> On the contrary. My guess is that Linux will adopt the best features of
>>> Solaris and become an even better OS.
>>
>> Linux will doubtless "borrow" some features from Solaris, but some
>> of the best features will be left out because Linus won't want them.
>> I mean, if he won't accept the ability to produce a core dump on
>> kernel panic, what chance has something like DTrace got?
>>
>
>
> Bullshit! Sun's CDDL conflicts with the GPL and no Open Source developer
> is going to touch it with a 20 foot stick. And I am sure that Sun would
> love to have some of its code show up in Linux so it can throw a big giant
> monkey wrench into the development and adoption of Linux.
In light of the way they've snuggled up to Microsoft... I'd say that's
probably what this is all about... Think about it... it fodder for the
canon...
-
******************************************************************************
Registered Linux User Number 185956
FSF Associate Member number 2340 since 05/20/2004
Join me in chat at #linux-users on irc.freenode.net
Buy an Xbox for $149.00, run linux on it and Microsoft loses $150.00!
10:37pm up 68 days, 6:23, 8 users, load average: 0.10, 0.14, 0.09
|
|
|
0
|
|
|
|
Reply
|
Jerry
|
12/17/2004 3:45:45 AM
|
|
Hamilcar Barca wrote:
> 01:01:44 +0000), Rich Teer wrote:
>> Becker claims that the GPL forbids the linking of his drivers with a
>> closed-source kernel (in this case, Solaris)
>
> Do you have some evidence to support this claim? Even if you do, the
> claim is irrelevant. When you stop ranting and actually read the GPL,
> you'll find that it does not in any way prohibit linking GPL code with
> proprietary code -- it prohibits only the distribution of such a merged
> product.
http://beowulf.es.embnet.org/listarchives/linux-tulip/1999/04/0076.html
True, one can port all you want, but if no one else can benefit
from your work, how is that helpful? I make no assertions that
I 100% understand the GPL, or other issues involved here, but I
fail to see how this is a good thing.
--
Eric Enright /"\
ericAtiptsoftDcom \ / ASCII Ribbon Campaign
X Against HTML E-Mail
Public Key: 0xBEDF636F / \
|
|
|
0
|
|
|
|
Reply
|
Eric
|
12/17/2004 8:43:30 AM
|
|
Hamilcar Barca <hamilcar@tld.always.invalid> writes:
[ Quotes deleted ]
> -- Eric S. Raymond
>Sun's new CDDL seems to be nothing more than a marketing announcement.
Certainly you don't think that any European takes Eric Raymond seriously
as he completely fails to understand how cathedrals were build?
Casper
|
|
|
0
|
|
|
|
Reply
|
Casper
|
12/17/2004 2:00:31 PM
|
|
begin Casper H.S. Dik wrote:
> Hamilcar Barca <hamilcar@tld.always.invalid> writes:
>
> [ Quotes deleted ]
>> -- Eric S. Raymond
>
>>Sun's new CDDL seems to be nothing more than a marketing announcement.
>
>
> Certainly you don't think that any European takes Eric Raymond seriously
> as he completely fails to understand how cathedrals were build?
>
Well, perhaps he liked more than in the orient, since he seems to know
something about bazaars
--
It's not about, 'Where do you want to go today?' It's more like,
'Where am I allowed to go today?'
|
|
|
0
|
|
|
|
Reply
|
Peter
|
12/17/2004 2:05:25 PM
|
|
Hamilcar Barca <hamilcar@tld.always.invalid> writes:
>If the source code was freely available, it documents the actual protocol,
>as implemented. The bigger problem is the closed source, compounded by
>the closed protocol.
That way madness lies. We've seen that with many protocols where
the implementation was the definition.
It stifles protocol development (if the source is the documentation,
there's a limit to what you can change in the source)
>I assume that when Mr. Schwartz writes "proprietary", he mean to write
>"100% licensed under the GPL". How non-proprietary is Solaris? It's
>been 100% closed since the day Sun (legally) combined BSD and System V.
"proprietary" means "not based on open standards"; that's what it always
meant until it was hijacked by the open source crowd.
>How non-proprietary is Java? Pro-open means having to license Java
>protocols, and having implementation requirements forced on the developer?
>How does purveying a Java Community [sic] Process under the control of Sun
>amount to "pro-open" [sic]?
Sun believes in "Open" standards; and standards require conformance
and interoperability testing.
Standards are good; they allow for interoperability, forward portability
and binary compatibility.
Standards do impose restrictions; failing to understand that that is
the reason for the restrictions and that those restrictions are there
for *very* good reasons is what lies at the heart of the debate between
those who believe "openess" just means that your source needs to be
open and those of us who believe that openess is much more than that.
Casper
--
Expressed in this posting are my opinions. They are in no way related
to opinions held by my employer, Sun Microsystems.
Statements on Sun products included here are not gospel and may
be fiction rather than truth.
|
|
|
0
|
|
|
|
Reply
|
Casper
|
12/17/2004 2:41:49 PM
|
|
>> Becker claims that the GPL forbids the linking of his drivers with a
>> closed-source kernel (in this case, Solaris)
>Do you have some evidence to support this claim? Even if you do, the
>claim is irrelevant. When you stop ranting and actually read the GPL,
>you'll find that it does not in any way prohibit linking GPL code with
>proprietary code -- it prohibits only the distribution of such a merged
>product.
Sun released code which allowed our customers to run Linux drivers
compiled from source under Solaris; Becker threatened to sue. End of
story.
Casper
|
|
|
0
|
|
|
|
Reply
|
Casper
|
12/17/2004 2:43:04 PM
|
|
In article <Pine.SOL.4.58.0412161632510.15621@zaphod>,
Rich Teer <rich.teer@rite-group.com> writes:
> On Thu, 16 Dec 2004, Hamilcar Barca wrote:
>
>> No. I think Sun -- specifically including Jonathan Schwartz -- should
>> stop spewing their hypocritical marketing drivel. "Open standards are
>
> Open standards ARE better than open source. If the closed standard
> that an open source project implements changes, that project is screwed
> until they can reverse-engineer the new standard. For example, suppose
> M$ decides to change the SMB protocol, so that it doesn't have the faintest
> resembalence to the protocol that Samba currently implements. Further
> suppose that they chose not to document or open that "standard". How long
> would it be before the Samba project could interact with the new SMB
> standard? Oh, and just to make it harder, M$ decides to encrypt all the
> protocol traffic, so reverse-engineering it by sniffing the network is
> made difficult. People like you would say "what's the problem? Samba is
> open source, so this problem will be easy to fix!"...
Well, to be fair, to be consistent he could also say that
Windows should be either open sourced or shouldn't be used. Not realistic,
but that's to be expected.
> When a standard is open, multiple implementations of it can compete, each
> on their own merits.
>
[...]
At least _somebody_ gets it!
--
mailto:rlhamil@smart.net http://www.smart.net/~rlhamil
Lasik/PRK theme music:
"In the Hall of the Mountain King", from "Peer Gynt"
|
|
|
0
|
|
|
|
Reply
|
Richard
|
12/17/2004 4:26:20 PM
|
|
In article <20041216223127.826$D5@news.newsreader.com>,
Hamilcar Barca <hamilcar@tld.always.invalid> writes:
> In article <Pine.SOL.4.58.0412161632510.15621@zaphod> (Fri, 17 Dec 2004
> 00:56:46 +0000), Rich Teer wrote:
>
>> On Thu, 16 Dec 2004, Hamilcar Barca wrote:
>>
>>> No. I think Sun -- specifically including Jonathan Schwartz -- should
>>> stop spewing their hypocritical marketing drivel. "Open standards are
>>
>> Open standards ARE better than open source. If the closed standard
>> that an open source project implements changes, that project is screwed
>> until they can reverse-engineer the new standard.
>
> If the source code was freely available, it documents the actual protocol,
> as implemented. The bigger problem is the closed source, compounded by
> the closed protocol.
Counter-example: something that (except for one branch for awhile) was
open for a very long time: the X11 windowing system. There's a _sample_
implementation, but they're very clear that it's not a _reference_
implementation. Putting the standards first insures interoperability,
while putting the code (however open and under whatever license) first
doesn't.
[...]
>> Now, run along little boy. Your daddy wants his computer back,
>> and don't forget to wipe your drool off the keyboard before you
>> go.
>
> That's even more juvenile as the "commie" insult. The Sun worshipers are
> becoming as obsessed as the Steve Jobs personality cult.
>
Yeah, both like something that's better than Linux!
--
mailto:rlhamil@smart.net http://www.smart.net/~rlhamil
Lasik/PRK theme music:
"In the Hall of the Mountain King", from "Peer Gynt"
|
|
|
0
|
|
|
|
Reply
|
Richard
|
12/17/2004 4:34:36 PM
|
|
In article <20041216191108.842$rd@news.newsreader.com>,
Hamilcar Barca <hamilcar@tld.always.invalid> writes:
> In article <10s45mbt8u3gl4b@corp.supernews.com> (Thu, 16 Dec 2004 23:14:19
> +0000), Richard L. Hamilton wrote:
>
>> In article <20041216160927.922$y9@news.newsreader.com>,
>> Hamilcar Barca <hamilcar@tld.always.invalid> writes:
>>>
>>> Sun's new CDDL seems to be nothing more than a marketing announcement.
>>
>> So, you think Sun should simply surrender control of everything they create
>> or surrender any claim to being open?
>
> No. I think Sun -- specifically including Jonathan Schwartz -- should
> stop spewing their hypocritical marketing drivel. "Open standards are
> better than open source", "Linux is Red Hat is Linux", "Java Community
> [sic] Process", "Open [sic] Solaris", and "C[sic]DDL" represent the sort
> of vacuous posturing for which Microsoft is noted.
So "vacuous posturing" just means you don't agree. Got it.
I personally _do_ agree that open standards are more important than
open source; any damnfool can write code, but if interoperability
standards either don't exist or are proprietary, then they can only
use that code one place, without a whole lot more work. Using the
code as the reference is simply not as disciplined; when the standard
has to evolve, that needs to be a matter of design, not just a matter
of a marathon hacking session!
WTF with the Java Community Process? Just because they don't give
up control or give everything away doesn't mean they're not inviting
(and when it suits them, listening to) input and participation. Remember,
a business is about return on investment for the investors, _not_ about
making the world a better place according to any other agenda. With
that as a priority, I'd say they're not doing badly at all. When I look
at http://www.jcp.org/en/introduction/faq#2.5
I see plenty of room for folks to clone stuff, as long as they do it
in an interoperable and disciplined manner And I see other things that
mean that while Sun controls the process, they don't totally dominate
the agenda. That's plenty good enough for me, for now anyway.
And the CDDL doesn't look any worse than a number of other open source
licenses. To me, it looks way better than that ideolog's dream of a GPL!
>> Did you grow up commie, or is it something you went out of your way to
>> become?
>
> "Commie" might have been a barbed insult in the 1950s but its use by a
> C[sic]DDL advocate simply demonstrates its uselessness.
>
>> That's not a fair position, is it?
>
> Of course it isn't, but you took it anyway.
And you missed the point; that you're at least as unreasonable in
your allegedly reasoned arguments as mere name calling would be in
response.
--
mailto:rlhamil@smart.net http://www.smart.net/~rlhamil
Lasik/PRK theme music:
"In the Hall of the Mountain King", from "Peer Gynt"
|
|
|
0
|
|
|
|
Reply
|
Richard
|
12/17/2004 4:58:14 PM
|
|
In article <cpt6pi$78i$00$1@news.t-online.com>,
Peter =?UTF-8?B?S8O2aGxtYW5u?= <Peter.Koehlmann@t-online.de> writes:
> begin Richard L. Hamilton wrote:
>
>> In article <20041216155840.132$KA@news.newsreader.com>,
>> Hamilcar Barca <hamilcar@tld.always.invalid> writes:
>>> In article <10s33n9oosu16e6@corp.supernews.com> (Thu, 16 Dec 2004
>>> 13:34:33 +0000), Richard L. Hamilton wrote:
>>>
>>>> The problem is also that GPL can be an obstacle to using free and
>>>> proprietary software together to meet the end user's needs.
>>>
>>> It does not. It prohibits the mixing of proprietary and GPL code only if
>>> they are "distributed". End users are free to mix, match, hack, break,
>>> and/or innovate [sic] to their heart's content.
>>
>> And I thought that other discussion already established that most end
>> users don't _want_ to do anything more than install and forget.
>>
>
> The discussion is not about what "most end users" want
>
> Other discussions also established that Solaris trolls are braindead,
> impotent and generally disliked. Just like wintrolls
>
But of course nobody that worships on the Holy Altar of Linux could
possibly be a troll. Yeah, right.
>>>> Or have we forgotten about them altogether?
>>>
>>> Who is "we"?
>>
>>
>> That would be those who are wrapped up in ideology and keep ignoring
>> the strictly practical.
>>
>> Even within the long-time open-source community, not everyone agrees on
>> GPL!
>>
>
> So what? Even GPL proponents (like me) see that it does not fit every
> purpose. Same is true for other licenses.
Woohoo! So does that mean that you at least admit that other licenses
aren't necessarily Evil(tm)?
--
mailto:rlhamil@smart.net http://www.smart.net/~rlhamil
Lasik/PRK theme music:
"In the Hall of the Mountain King", from "Peer Gynt"
|
|
|
0
|
|
|
|
Reply
|
Richard
|
12/17/2004 5:02:35 PM
|
|
In article <9c1a92xbur.ln2@spinner.my.domain>,
Jerry McBride <mcbrides9@comcast.net> writes:
> Freeride wrote:
>
>> Rich Teer wrote:
>>
>>> On Mon, 13 Dec 2004, Eltee wrote:
>>>
>>>> On the contrary. My guess is that Linux will adopt the best features of
>>>> Solaris and become an even better OS.
>>>
>>> Linux will doubtless "borrow" some features from Solaris, but some
>>> of the best features will be left out because Linus won't want them.
>>> I mean, if he won't accept the ability to produce a core dump on
>>> kernel panic, what chance has something like DTrace got?
>>>
>>
>>
>> Bullshit! Sun's CDDL conflicts with the GPL and no Open Source developer
>> is going to touch it with a 20 foot stick. And I am sure that Sun would
>> love to have some of its code show up in Linux so it can throw a big giant
>> monkey wrench into the development and adoption of Linux.
>
> In light of the way they've snuggled up to Microsoft... I'd say that's
> probably what this is all about... Think about it... it fodder for the
> canon...
How is cooperating where there are mutual interests automatically a
conspiracy to crush Linux? Why would Sun really care one way or
the other about Linux, as long as people buy their hardware or support
services?
--
mailto:rlhamil@smart.net http://www.smart.net/~rlhamil
Lasik/PRK theme music:
"In the Hall of the Mountain King", from "Peer Gynt"
|
|
|
0
|
|
|
|
Reply
|
Richard
|
12/17/2004 5:06:40 PM
|
|
Casper H.S. Dik wrote:
> Hamilcar Barca <hamilcar@tld.always.invalid> writes:
<snip>
> It stifles protocol development (if the source is the documentation,
> there's a limit to what you can change in the source)
How so ? You write up your modifications as a RFC document. Then produce
the code. Did TCP/IP being non-proprietary stifle its spread ?
>> I assume that when Mr. Schwartz writes "proprietary", he mean to
>> write "100% licensed under the GPL". How non-proprietary is
>> Solaris? It's been 100% closed since the day Sun (legally)
>> combined BSD and System V.
> "proprietary" means "not based on open standards"; that's what it
> always meant until it was hijacked by the open source crowd.
In the context of source code, non "proprietary" means that the
purchaser is free to use the source code as he sees fit as long as he
includes the source of any additions he makes to the code in subsequent
sales.
That is what it has always meant. At least until the closed source crowd
attempted to hijack the concept and pollute its meaning to the public.
By bending the argument into a 'open standards' rather than 'open
source' debate. A bit like you're doing here.
<snip>
> Standards do impose restrictions; failing to understand that that is
> the reason for the restrictions and that those restrictions are there
> for *very* good reasons is what lies at the heart of the debate
> between those who believe "openess" just means that your source needs
> to be open and those of us who believe that openess is much more than
> that.
I see arguments like this as an attempt by the 'closed source crowd' to
hijack the concept. See its got 'open' in the title. And what could be
more open than that. This reference to protocols is just distraction fud.
|
|
|
0
|
|
|
|
Reply
|
Daeron
|
12/17/2004 7:36:25 PM
|
|
Linux Advocates Fear Solaris 10.
Casper H.S. Dik wrote:
> Sun released code which allowed our customers to run Linux drivers
> compiled from source under Solaris; Becker threatened to sue. End of
> story.
So the problem is with Suns' restrictive license, is it not ?
According to this article Sun took GPL source and compiled it into
Solaris binaries. According to Bruce Perens Sun is within its legal
rights to do so. Becker argues that the GPL does not permit distribution
as part of non-GPL'd code for commercial use.
www.angelfire.com/blog/trollarchive/sun.finds.hole.in.gpl.html
|
|
|
0
|
|
|
|
Reply
|
Daeron
|
12/17/2004 7:46:25 PM
|
|
Daeron <daeron@demon.com> writes:
>> It stifles protocol development (if the source is the documentation,
>> there's a limit to what you can change in the source)
>How so ? You write up your modifications as a RFC document. Then produce
>the code. Did TCP/IP being non-proprietary stifle its spread ?
You forget to mention that the source wasn't the documentation to
begin with when it comes to TCP/IP.
And that for TCP/IP it was the *open*standards* RFC process which
made development possible; and it was the fact that multiple,
separate implementations existed that made the implementations
honest, caused the standards to be clarified.
If all we had was, say, BSD networking, then it would have been
very hard to write an interoperable TCP/IP stack for any OS.
>In the context of source code, non "proprietary" means that the
>purchaser is free to use the source code as he sees fit as long as he
>includes the source of any additions he makes to the code in subsequent
>sales.
Proprietary has one meaning only; it has nothing to do with source code.
Even "open source" can be owned by someone; in computing proprietary means
one thing (CIFS, not NFS)
>That is what it has always meant. At least until the closed source crowd
>attempted to hijack the concept and pollute its meaning to the public.
>By bending the argument into a 'open standards' rather than 'open
>source' debate. A bit like you're doing here.
Proprietary has *no* meaning in the context of source code and the
concept of proprietary (meaning one of a kind, vendor specific like
Windows, VMS, AOS/VS) was used to set apart the "Open Standard"
based (read: Unix) based implementations.
Proprietary was only recently hijacked by the open source proponents
*when they went commercial*.
>I see arguments like this as an attempt by the 'closed source crowd' to
>hijack the concept. See its got 'open' in the title. And what could be
>more open than that. This reference to protocols is just distraction fud.
Perhaps you should read up on computer history; you obviously weren't
there.
Casper
--
Expressed in this posting are my opinions. They are in no way related
to opinions held by my employer, Sun Microsystems.
Statements on Sun products included here are not gospel and may
be fiction rather than truth.
|
|
|
0
|
|
|
|
Reply
|
Casper
|
12/17/2004 8:09:57 PM
|
|
Daeron <daeron@demon.com> writes:
> Linux Advocates Fear Solaris 10.
>Casper H.S. Dik wrote:
>> Sun released code which allowed our customers to run Linux drivers
>> compiled from source under Solaris; Becker threatened to sue. End of
>> story.
>So the problem is with Suns' restrictive license, is it not ?
Ah, so when the GPL restricts something from being distributed with
it, it's never the GPL's fault? Right.
I guess we're now debating a religion rather than a license.
Casper
--
Expressed in this posting are my opinions. They are in no way related
to opinions held by my employer, Sun Microsystems.
Statements on Sun products included here are not gospel and may
be fiction rather than truth.
|
|
|
0
|
|
|
|
Reply
|
Casper
|
12/17/2004 8:12:49 PM
|
|
On Fri, 17 Dec 2004, Casper H.S. Dik wrote:
> Proprietary has *no* meaning in the context of source code and the
> concept of proprietary (meaning one of a kind, vendor specific like
> Windows, VMS, AOS/VS) was used to set apart the "Open Standard"
> based (read: Unix) based implementations.
>
> Proprietary was only recently hijacked by the open source proponents
> *when they went commercial*.
What I find most amusing in these rants by GPL worhipers, who openly
despise anything they deem as "proprietory", is the fact that most of
them run Linux on proprietory x86-based machines yet don't give Sun
credit for their SPARC CPU, which is an open standard. Essentially,
they're hypocrites. (I hasten to add that those running Linux on SPARC
are NOT hypocrites, even though I personally don't see the point of doing
so.)
> Perhaps you should read up on computer history; you obviously weren't
> there.
Indeed.
--
Rich Teer, SCNA, SCSA, author of "Solaris Systems Programming"
. * * . * .* .
. * . .*
President, * . . /\ ( . . *
Rite Online Inc. . . / .\ . * .
.*. / * \ . .
. /* o \ .
Voice: +1 (250) 979-1638 * '''||''' .
URL: http://www.rite-online.net ******************
|
|
|
0
|
|
|
|
Reply
|
Rich
|
12/17/2004 8:31:44 PM
|
|
begin Rich Teer wrote:
> On Fri, 17 Dec 2004, Casper H.S. Dik wrote:
>
>> Proprietary has *no* meaning in the context of source code and the
>> concept of proprietary (meaning one of a kind, vendor specific like
>> Windows, VMS, AOS/VS) was used to set apart the "Open Standard"
>> based (read: Unix) based implementations.
>>
>> Proprietary was only recently hijacked by the open source proponents
>> *when they went commercial*.
>
> What I find most amusing in these rants by GPL worhipers, who openly
> despise anything they deem as "proprietory", is the fact that most of
> them run Linux on proprietory x86-based machines yet don't give Sun
> credit for their SPARC CPU, which is an open standard.
Idiot
Or do <ou have even dumber "arguments" and aspire for "cretin"?
< snip >
--
"Last I checked, it wasn't the power cord for the Clue Generator that
was sticking up your ass." - John Novak, rasfwrj
|
|
|
0
|
|
|
|
Reply
|
Peter
|
12/17/2004 8:39:07 PM
|
|
On 17 Dec 2004 14:41:49 GMT, Casper H.S Dik <Casper.Dik@Sun.COM> wrote:
> Hamilcar Barca <hamilcar@tld.always.invalid> writes:
>
>>If the source code was freely available, it documents the actual protocol,
>>as implemented. The bigger problem is the closed source, compounded by
>>the closed protocol.
>
> That way madness lies. We've seen that with many protocols where
> the implementation was the definition.
>
> It stifles protocol development (if the source is the documentation,
> there's a limit to what you can change in the source)
Well designed protocols incorporate extensibility (I'm thinking of
ASN.1, with which you can write protocol version 1, and then version 2,
and it should be both forwards and backwards compatible). I have the
impression that most open source protocols are based on ASCII parsing,
and are either difficult to extend, or very verbose, or both.
A bientot
Paul
--
Paul Floyd http://paulf.free.fr (for what it's worth)
Surgery: ennobled Gerald.
|
|
|
0
|
|
|
|
Reply
|
Paul
|
12/17/2004 9:16:12 PM
|
|
Hamilcar Barca <hamilcar@tld.always.invalid> writes in comp.unix.solaris:
|In article <cprdt9$mln$1@agate.berkeley.edu> (Thu, 16 Dec 2004 07:36:09
|+0000), Alan Coopersmith wrote:
|
|> Hamilcar Barca <hamilcar@tld.always.invalid> writes in comp.unix.solaris:
|> |What's the point, if any, of the CDDL?
|>
|> To provide an update to the commonly accepted Mozilla Public License to
|> deal with some concerns people had about the MPL.
|
|Why is this useful?
Because it fits better what the authors/owners of various code want/need
in order to release it to open source. No one license is ever likely to
satisfy everyone. If this satisfies some enough to release code to open
source that otherwise wouldn't be, or which would go out under a more
restrictive license, isn't that a good thing? (Yes, in the eyes of GPL
purists, it may not be the best outcome, but it's better than nothing.)
|> If the OSI review goes well
|
|This OSI?
|
| http://www.opensource.org/docs/board.php
Yes.
|> (which it seems to be, providing useful
|> feedback for improving the license further), [...]
|
|This kind of feedback?
No, the kind of rational discussion being found on their mailing lists,
not sound bites chosen by journalists to make their articles more
interesting by trying to show controversy whereever possible.
--
________________________________________________________________________
Alan Coopersmith * alanc@alum.calberkeley.org * Alan.Coopersmith@Sun.COM
http://www.csua.berkeley.edu/~alanc/ * http://blogs.sun.com/alanc/
Working for, but definitely not speaking for, Sun Microsystems, Inc.
|
|
|
0
|
|
|
|
Reply
|
Alan
|
12/17/2004 9:45:18 PM
|
|
On Fri, 17 Dec 2004, Peter [UTF-8] Köhlmann wrote:
> > What I find most amusing in these rants by GPL worhipers, who openly
> > despise anything they deem as "proprietory", is the fact that most of
> > them run Linux on proprietory x86-based machines yet don't give Sun
> > credit for their SPARC CPU, which is an open standard.
>
> Idiot
Ah yes. When well reasoned logic, backed up by independant data,
fails, resort to personal insults. You really are graping at straws
now, Peter.
But I'm game: where's your proof that SPARC is not an open standard,
and that x86 is not proprietory?
Or are you disputing my statement that most GPL advocates run
Linux on x86 hardware?
Or perhaps, as is becoming increasingly evidient, you're not
capable of holding a serious debate with anyone who's views
differ from your own, and in fact, you're a troll.
--
Rich Teer, SCNA, SCSA, author of "Solaris Systems Programming"
. * * . * .* .
. * . .*
President, * . . /\ ( . . *
Rite Online Inc. . . / .\ . * .
.*. / * \ . .
. /* o \ .
Voice: +1 (250) 979-1638 * '''||''' .
URL: http://www.rite-online.net ******************
|
|
|
0
|
|
|
|
Reply
|
Rich
|
12/17/2004 10:42:49 PM
|
|
begin Rich Teer wrote:
> On Fri, 17 Dec 2004, Peter [UTF-8] Köhlmann wrote:
>
>> > What I find most amusing in these rants by GPL worhipers, who openly
>> > despise anything they deem as "proprietory", is the fact that most of
>> > them run Linux on proprietory x86-based machines yet don't give Sun
>> > credit for their SPARC CPU, which is an open standard.
>>
>> Idiot
>
> Ah yes. When well reasoned logic, backed up by independant data,
> fails, resort to personal insults. You really are graping at straws
> now, Peter.
>
> But I'm game: where's your proof that SPARC is not an open standard,
> and that x86 is not proprietory?
>
> Or are you disputing my statement that most GPL advocates run
> Linux on x86 hardware?
>
> Or perhaps, as is becoming increasingly evidient, you're not
> capable of holding a serious debate with anyone who's views
> differ from your own, and in fact, you're a troll.
>
As already said, you are an idiot
--
Who the fuck is General Failure, and why is he reading my harddisk?
|
|
|
0
|
|
|
|
Reply
|
Peter
|
12/17/2004 10:49:15 PM
|
|
Paul Floyd <root@127.0.0.1> writes:
>Well designed protocols incorporate extensibility (I'm thinking of
>ASN.1, with which you can write protocol version 1, and then version 2,
>and it should be both forwards and backwards compatible). I have the
>impression that most open source protocols are based on ASCII parsing,
>and are either difficult to extend, or very verbose, or both.
There are many errors to make in protocol specifications and
design; one would be lack of extensibility, but the main problem with
"it's documented by the source code" type of protocols is the fact
that typical implementations lack sufficient error checking and
then clients will rely on that. There are interesting examples
of that in the BSD LP protocol.
The protocol can be then condensed in a RFC but that is no guarantee
that subsequent implementation from that RFC can interoperate; again
the BSD printer protocol demonstrates that; it's also not extensible.
Examples of good protocols include "telnet" which was defined in the
more formal era of computing; bad protocols are some of the early
more-or-less open source ones: BSD rsh/rlogin/rexec and coupled with
that rmt and rdist.
Casper
--
Expressed in this posting are my opinions. They are in no way related
to opinions held by my employer, Sun Microsystems.
Statements on Sun products included here are not gospel and may
be fiction rather than truth.
|
|
|
0
|
|
|
|
Reply
|
Casper
|
12/17/2004 10:57:38 PM
|
|
On Fri, 17 Dec 2004, Peter [UTF-8] Köhlmann wrote:
> As already said, you are an idiot
Using my "fsckwit to English" translator, I read that as "you are
100% correct; I give up and bow to your intellectual superiority".
*Plonk*
--
Rich Teer, SCNA, SCSA, author of "Solaris Systems Programming"
. * * . * .* .
. * . .*
President, * . . /\ ( . . *
Rite Online Inc. . . / .\ . * .
.*. / * \ . .
. /* o \ .
Voice: +1 (250) 979-1638 * '''||''' .
URL: http://www.rite-online.net ******************
|
|
|
0
|
|
|
|
Reply
|
Rich
|
12/17/2004 11:07:41 PM
|
|
In article <Pine.SOL.4.58.0412171435230.19783@zaphod>,
Rich Teer <rich.teer@rite-group.com> wrote:
>On Fri, 17 Dec 2004, Peter [UTF-8] Köhlmann wrote:
nothing but nonsense and insults that should better not
quoted to save the environment.....
an Rich replied:
>Ah yes. When well reasoned logic, backed up by independant data,
>fails, resort to personal insults. You really are graping at straws
>now, Peter.
.....
>Or perhaps, as is becoming increasingly evidient, you're not
>capable of holding a serious debate with anyone who's views
>differ from your own, and in fact, you're a troll.
Rich,
it does not make sense to even try a discussion with this troll.
I've made a similar experience a few days ago and I decided that
any single word you try write in hope to have a serios discussion with him
is just lost time :-(
He does not have a clue about Open Standards vs. proprietary software
(e.g. such as made by the FSF in some cases) and he seems to be unable to
understand Copyright laws and licensing issues. We should better give up.
At least we could make the experience that there are other people on cola
who _are_ able to have serious discussions. Let us continue a decent
discussion with the people who are able to do so.....
--
EMail:joerg@schily.isdn.cs.tu-berlin.de (home) J�rg Schilling D-13353 Berlin
js@cs.tu-berlin.de (uni) If you don't have iso-8859-1
schilling@fokus.fraunhofer.de (work) chars I am J"org Schilling
URL: http://www.fokus.fraunhofer.de/usr/schilling ftp://ftp.berlios.de/pub/schily
|
|
|
0
|
|
|
|
Reply
|
js
|
12/17/2004 11:12:50 PM
|
|
begin Rich Teer wrote:
> On Fri, 17 Dec 2004, Peter [UTF-8] Köhlmann wrote:
>
>> As already said, you are an idiot
>
> Using my "fsckwit to English" translator, I read that as "you are
> 100% correct; I give up and bow to your intellectual superiority".
>
> *Plonk*
>
Good. If that means you stop to post your drivel in groups which won't have
it, fine
And you are still an idiot
You may continue to think that SPARC is "open" and x86 is "closed"
You just reinforce the appearance of a cretin too dumb to spill abucket of
water
Pray tell, are all the posters in the solaris group as stupid as you are, or
are there even dumber people (sounds not very convincing)
--
Klingon function calls do not have 'parameters' -
they have 'arguments' - and they ALWAYS WIN THEM.
|
|
|
0
|
|
|
|
Reply
|
Peter
|
12/17/2004 11:27:59 PM
|
|
On Fri, 17 Dec 2004, Joerg Schilling wrote:
> Rich,
>
> it does not make sense to even try a discussion with this troll.
I think you're right.
> He does not have a clue about Open Standards vs. proprietary software
> (e.g. such as made by the FSF in some cases) and he seems to be unable to
> understand Copyright laws and licensing issues. We should better give up.
I agree. Anyone who is not even capable of using google to determine
the open nature of SPARC (hint: you can find out all about it on SPARC
International's web site, www.sparc.com) isn't worth wasting breath on.
> At least we could make the experience that there are other people on cola
> who _are_ able to have serious discussions. Let us continue a decent
> discussion with the people who are able to do so.....
Indeed.
--
Rich Teer, SCNA, SCSA, author of "Solaris Systems Programming"
. * * . * .* .
. * . .*
President, * . . /\ ( . . *
Rite Online Inc. . . / .\ . * .
.*. / * \ . .
. /* o \ .
Voice: +1 (250) 979-1638 * '''||''' .
URL: http://www.rite-online.net ******************
|
|
|
0
|
|
|
|
Reply
|
Rich
|
12/18/2004 12:29:35 AM
|
|
In comp.os.linux.advocacy, Rich Teer
<rich.teer@rite-group.com>
wrote
on Fri, 17 Dec 2004 20:31:44 GMT
<Pine.SOL.4.58.0412171225480.19783@zaphod>:
> On Fri, 17 Dec 2004, Casper H.S. Dik wrote:
>
>> Proprietary has *no* meaning in the context of source code and the
>> concept of proprietary (meaning one of a kind, vendor specific like
>> Windows, VMS, AOS/VS) was used to set apart the "Open Standard"
>> based (read: Unix) based implementations.
>>
>> Proprietary was only recently hijacked by the open source proponents
>> *when they went commercial*.
>
> What I find most amusing in these rants by GPL worhipers, who openly
> despise anything they deem as "proprietory", is the fact that most of
> them run Linux on proprietory x86-based machines yet don't give Sun
> credit for their SPARC CPU, which is an open standard. Essentially,
> they're hypocrites. (I hasten to add that those running Linux on SPARC
> are NOT hypocrites, even though I personally don't see the point of doing
> so.)
Exactly. We should of course all abandon the GPL for this reason,
and go back to the tried and true methods of software development
houses such as Microsoft.
(BTW: 'x86' is kinda open-source as well -- Cyrix, AMD, and others
compete in this space. I don't know if there's a standard for it,
but they work well enough... :-) )
>
>> Perhaps you should read up on computer history; you obviously weren't
>> there.
>
> Indeed.
>
--
#191, ewill3@earthlink.net
It's still legal to go .sigless.
|
|
|
0
|
|
|
|
Reply
|
The
|
12/18/2004 1:00:08 AM
|
|
In article <32gqppF3l62c0U1@individual.net>,
Daeron <daeron@demon.com> writes:
> Casper H.S. Dik wrote:
[...]
>> "proprietary" means "not based on open standards"; that's what it
>> always meant until it was hijacked by the open source crowd.
>
> In the context of source code, non "proprietary" means that the
> purchaser is free to use the source code as he sees fit as long as he
> includes the source of any additions he makes to the code in subsequent
> sales.
The BSD license predates that viral nonsense about whether or not you
redistribute your changes along with binaries by a long time; I've never
seen anyone say it wasn't open, although some people for whatever reason
do prefer GPL.
> That is what it has always meant. At least until the closed source crowd
> attempted to hijack the concept and pollute its meaning to the public.
> By bending the argument into a 'open standards' rather than 'open
> source' debate. A bit like you're doing here.
GPL is not the only possible open source license. What part of that don't
some folks get?
>> Standards do impose restrictions; failing to understand that that is
>> the reason for the restrictions and that those restrictions are there
>> for *very* good reasons is what lies at the heart of the debate
>> between those who believe "openess" just means that your source needs
>> to be open and those of us who believe that openess is much more than
>> that.
>
> I see arguments like this as an attempt by the 'closed source crowd' to
> hijack the concept. See its got 'open' in the title. And what could be
> more open than that. This reference to protocols is just distraction fud.
Bull. Standards are more important than code simply as a matter of
engineering discipline. Whether they're all open or all deep dark secrets
is secondary, although if the objective is to allow and encourage
alternative implementations (rather than have one magic hackfest product
rule the world), the standards need to be open.
--
mailto:rlhamil@smart.net http://www.smart.net/~rlhamil
Lasik/PRK theme music:
"In the Hall of the Mountain King", from "Peer Gynt"
|
|
|
0
|
|
|
|
Reply
|
Richard
|
12/18/2004 1:01:44 AM
|
|
In article <cpvpv6$nv0$05$1@news.t-online.com>,
Peter =?UTF-8?B?S8O2aGxtYW5u?= <Peter.Koehlmann@t-online.de> writes:
[...]
> You may continue to think that SPARC is "open" and x86 is "closed"
> You just reinforce the appearance of a cretin too dumb to spill abucket of
> water
If you can get the full tapes for an (admittedly not current) SPARC chip
for free, and you can't for an x86 chip, which is more open?
people (sounds not very convincing)
--
mailto:rlhamil@smart.net http://www.smart.net/~rlhamil
Lasik/PRK theme music:
"In the Hall of the Mountain King", from "Peer Gynt"
|
|
|
0
|
|
|
|
Reply
|
Richard
|
12/18/2004 1:05:22 AM
|
|
Rich Teer poked his little head through the XP firewall and said:
> On Fri, 17 Dec 2004, Casper H.S. Dik wrote:
>
>> Proprietary has *no* meaning in the context of source code and the
>> concept of proprietary (meaning one of a kind, vendor specific like
>> Windows, VMS, AOS/VS) was used to set apart the "Open Standard"
>> based (read: Unix) based implementations.
>>
>> Proprietary was only recently hijacked by the open source proponents
>> *when they went commercial*.
>
> What I find most amusing in these rants
Who's ranting here?
> by GPL worhipers
Who's worshiping here?
> who openly despise anything they deem as "proprietory",
Painting with a broad brush.
> is the fact that most of them run Linux on proprietory x86-based machines
All computers are proprietary. Except maybe the "Freedom CPU" (see
http://www.f-cpu.org).
> yet don't give Sun credit for their SPARC CPU, which is an open standard.
The x86 architecture is open also, even if the implementations are
proprietary.
> Essentially, they're hypocrites.
Maybe. Except that chip fabrication is not yet something that can be done
by individuals, though software is amenable to solo artists.
> (I hasten to add that those running Linux on SPARC are NOT hypocrites,
Why not? They may be running on an "open CPU", as you seem to be claiming,
but there are still many other proprietary things that they object to.
> even though I personally don't see the point of doing so.)
Why not? Is the SPARC optimized to run Sun operating systems?
>> Perhaps you should read up on computer history; you obviously weren't
>> there.
>
> Indeed.
--
Tux: "If you bought a computer with Windows, sorry, you paid too much!"
Ballmer: "We need to get paid!"
|
|
|
0
|
|
|
|
Reply
|
iso
|
12/18/2004 3:19:54 AM
|
|
Rich Teer poked his little head through the XP firewall and said:
> On Fri, 17 Dec 2004, Peter [UTF-8] Köhlmann wrote:
>
>> Idiot
>
> Ah yes. When well reasoned logic, backed up by independant data,
> fails, resort to personal insults. You really are graping at straws
> now, Peter.
You need to take it for what Peter meant it, Rich. He despises your
argument.
>
> But I'm game: where's your proof that SPARC is not an open standard,
> and that x86 is not proprietory?
SPARC: http://www.sparc.com/v8.shtml
x86: http://www.x86.org (site not responding, though)
>
> Or are you disputing my statement that most GPL advocates run
> Linux on x86 hardware?
Disputing that statement, as well as making it in the first place, requires
some hard data. Got any? If you do, does it matter?
--
Tux: "If you bought a computer with Windows, sorry, you paid too much!"
Ballmer: "We need to get paid!"
|
|
|
0
|
|
|
|
Reply
|
iso
|
12/18/2004 3:28:19 AM
|
|
Richard L. Hamilton poked his little head through the XP firewall and said:
> Bull. Standards are more important than code simply as a matter of
> engineering discipline. Whether they're all open or all deep dark secrets
> is secondary, although if the objective is to allow and encourage
> alternative implementations (rather than have one magic hackfest product
> rule the world), the standards need to be open.
Code is important. Why? Because, you if have code, you can then build it
and run it.
If you have an open standard, now you must write code based on that
standard. Many people cannot write code. Therefore, the open standard does
them no good, unless someone writes the code for them. If there's a charge
for that code, that may prevent them from using the code. That is why open
source is important.
Also, standards are not really easy to test. Code is much easier to test.
--
Tux: "If you bought a computer with Windows, sorry, you paid too much!"
Ballmer: "We need to get paid!"
|
|
|
0
|
|
|
|
Reply
|
iso
|
12/18/2004 3:35:07 AM
|
|
Casper H.S Dik poked his little head through the XP firewall and said:
>>> Becker claims that the GPL forbids the linking of his drivers with a
>>> closed-source kernel (in this case, Solaris)
>
>>Do you have some evidence to support this claim? Even if you do, the
>>claim is irrelevant. When you stop ranting and actually read the GPL,
>>you'll find that it does not in any way prohibit linking GPL code with
>>proprietary code -- it prohibits only the distribution of such a merged
>>product.
>
> Sun released code which allowed our customers to run Linux drivers
> compiled from source under Solaris; Becker threatened to sue. End of
> story.
Actually, it sounds too simple. Surely there must be more to the story.
--
Tux: "If you bought a computer with Windows, sorry, you paid too much!"
Ballmer: "We need to get paid!"
|
|
|
0
|
|
|
|
Reply
|
iso
|
12/18/2004 3:37:31 AM
|
|
Richard L. Hamilton poked his little head through the XP firewall and said:
>> That's even more juvenile as the "commie" insult. The Sun worshipers are
>> becoming as obsessed as the Steve Jobs personality cult.
>
> Yeah, both like something that's better than Linux!
For some meanings of "better". I'm disinclined, for example, to download
Solaris and install it on one of my x86 boxes, because of its poor
reputation on that hardware.
Maybe that reputation is now better. So, where can I get Solaris to
download and try? Ah, found it. Damn, now I can't rag on Sun until I
actually try this out. Hmmmm, I only have one box with 512Mb on it, and
that's this one. Still, maybe it's worth trying.
Damn, what an ordeal just to get to the license!
3.0 LICENSE RESTRICTIONS
3.1 You agree to reproduce any copyright and other proprietary right notices
on any copy of Software. Except as otherwise provided by law, You may not
modify or create derivative works of Software, or reverse engineer,
disassemble or decompile binary portions of Software, or otherwise
attempt to derive the source code from such portions.
Yeah, that definitely reminds me why I chose Linux. Thanks, guys!
Anyway, since I'm technically interested, and not dependent on Solaris to do
computing, I can afford the luxury of trying it out.
Okay, got the four CDs downloading now, it should take about an hour.
It will be quite awhile before I get the hardware ready for it, though.
--
Tux: "If you bought a computer with Windows, sorry, you paid too much!"
Ballmer: "We need to get paid!"
|
|
|
0
|
|
|
|
Reply
|
iso
|
12/18/2004 3:59:46 AM
|
|
On Fri, 17 Dec 2004 21:59:46 -0600, Lin�nut <lin�nut@bone.com> wrote:
>Richard L. Hamilton poked his little head through the XP firewall and said:
>
>>> That's even more juvenile as the "commie" insult. The Sun worshipers are
>>> becoming as obsessed as the Steve Jobs personality cult.
>>
>> Yeah, both like something that's better than Linux!
>
>For some meanings of "better". I'm disinclined, for example, to download
>Solaris and install it on one of my x86 boxes, because of its poor
>reputation on that hardware.
>
>Maybe that reputation is now better. So, where can I get Solaris to
>download and try? Ah, found it. Damn, now I can't rag on Sun until I
>actually try this out. Hmmmm, I only have one box with 512Mb on it, and
>that's this one. Still, maybe it's worth trying.
I'm running it on an x86 with 128mb ram and it works just fine.
|
|
|
0
|
|
|
|
Reply
|
Gary
|
12/18/2004 4:03:28 AM
|
|
In article <41c2f02d$0$78228$e4fe514c@news.xs4all.nl> (Fri, 17 Dec 2004
14:41:49 +0000), Casper H. S. Dik wrote:
>>I assume that when Mr. Schwartz writes "proprietary", he mean to write
>>"100% licensed under the GPL". How non-proprietary is Solaris? It's
>>been 100% closed since the day Sun (legally) combined BSD and System V.
>
> "proprietary" means "not based on open standards"; that's what it always
> meant until it was hijacked by the open source crowd.
Sure, just like DEC's claim that renaming VMS to OpenVMS was justified
because running on two single-source platforms was the "definition of open."
Proprietary means closed.
> the debate between
> those who believe "openess" just means that your source needs to be
> open and those of us who believe that openess is much more than that.
And those who believe openness is much less than that, in complete control
over so-called standards.
--
"[P]eople who use MSN's search engine probably...expose Microstupidware
to the Internet and so try to flood my mailbox with virus and worm
spoors. They probably also send HTML encrypted mail."
-- Vernon Schryver, in <c7ocrs@calcite.rhyolite.com>
|
|
|
0
|
|
|
|
Reply
|
Hamilcar
|
12/18/2004 7:51:01 AM
|
|
In article <Pine.SOL.4.58.0412171225480.19783@zaphod> (Fri, 17 Dec 2004
20:31:44 +0000), Rich Teer wrote:
> is the fact that most of them run Linux on proprietory x86-based
> machines
It is? There are CPUs from Intel, AMD, Cyrix (National Semiconductor?),
VIA, and TransMeta. Are these proprietary.
> yet don't give Sun credit for their SPARC CPU, which is an
> open standard.
Who's the accrediting body?
> Essentially, they're hypocrites.
Essentially, you had no basis for your claims and were forced to resort to
argument ad hominem.
--
"But mark this: all the silly names by all the silly Linux morons
in the world won't stop Longhorn from being a HUGE commercial and
technical success for MS."
-- DFS
|
|
|
0
|
|
|
|
Reply
|
Hamilcar
|
12/18/2004 8:00:39 AM
|
|
In article <Pine.SOL.4.58.0412171435230.19783@zaphod> (Fri, 17 Dec 2004
22:42:49 +0000), Rich Teer wrote:
> But I'm game: where's your proof that SPARC is not an open standard,
> and that x86 is not proprietory?
You made the claims; you provide the proof.
> Or are you disputing my statement that most GPL advocates run
> Linux on x86 hardware?
This is irrelevant, until you've shown your premises to be true.
--
"McBride said he sometimes carries a gun...and travels with armed
guards. The gun is licensed [and] security officials [sic] have
told him that convicted felons are behind the death threats."
-- Bloomberg News. 8 March 2004.
|
|
|
0
|
|
|
|
Reply
|
Hamilcar
|
12/18/2004 8:02:07 AM
|
|
In article <41c2f078$0$78228$e4fe514c@news.xs4all.nl> (Fri, 17 Dec 2004
14:43:04 +0000), Casper H. S. Dik wrote:
>>> Becker claims that the GPL forbids the linking of his drivers with a
>>> closed-source kernel (in this case, Solaris)
>
>>Do you have some evidence to support this claim? Even if you do, the
>>claim is irrelevant. When you stop ranting and actually read the GPL,
>>you'll find that it does not in any way prohibit linking GPL code with
>>proprietary code -- it prohibits only the distribution of such a merged
>>product.
>
> Sun released code which allowed our customers to run Linux drivers
> compiled from source under Solaris; Becker threatened to sue.
SCOX is suing IBM, sometimes for copyright violations, restraint of
trade, Un-Constutional and illegal use of a licensed called GPL, quite
apparently without any cause.
Neither a threat to sue or an actual lawsuit is evidence of GPL violation.
> End of story.
Perhaps it was. A similar ending was found in the story of Kodak v. Sun.
No issues of any kind were decided in either of these situations, except
that of corporations doing business.
|
|
|
0
|
|
|
|
Reply
|
Hamilcar
|
12/18/2004 8:08:00 AM
|
|
In article <10s6416rb1crd68@corp.supernews.com> (Fri, 17 Dec 2004 16:58:14
+0000), Richard L. Hamilton wrote:
> In article <20041216191108.842$rd@news.newsreader.com>,
> Hamilcar Barca <hamilcar@tld.always.invalid> writes:
>
> So "vacuous posturing" just means you don't agree. Got it.
I don't agree with vacuous posturing. Such posturers include Bill Gates,
Steve Ballmer, Darl McBride, and now Jonathan Schwartz. They're either
dimwitted or deliberate liars. I don't agree with their lack of ethics.
> WTF with the Java Community Process?
The Community [sic] in JCP is marketing drivel and vacuous posturing.
> Just because they don't give
> up control or give everything away
It's your statement that they don't "give everything away". I said its a
deliberately misnamed process, similar but not identical to Microsoft's
"Shared [sic] Source".
> Remember, a business is about return on investment for the investors,
> _not_ about making the world a better place according to any other agenda.
You keep changing the subject. If Sun wants to maintain control over Java
or Solaris or whatever, let them do it. I object to their fluff and lies,
and personally I don't see why anyone would trust such a company enough to
do business with them. It's not like they're the electric monopoly.
> I see plenty of room for folks to clone stuff, as long as they do it
> in an interoperable and disciplined manner
In other words, as long as they do it Sun's way.
> And the CDDL doesn't look any worse than a number of other open source
> licenses. To me, it looks way better than that ideolog's dream of a GPL!
It doesn't. My original point, you'll note, was that it was entirely
unnecessary.
> And you missed the point; that you're at least as unreasonable in
> your allegedly reasoned arguments as mere name calling would be in
> response.
Were you offended that I posted some misstatements by a Sun executive?
--
ITB: How worried are you about the number of attacks on Microsoft software?
BG: Actually in a sense it's very good to have this maturity ...
-- IT Business interview with Bill Gates (Chairman, Microsoft). 10/29/2003
http://www.itbusiness.ca/index.asp?theaction=61&sid=53897
|
|
|
0
|
|
|
|
Reply
|
Hamilcar
|
12/18/2004 8:18:03 AM
|
|
Hamilcar Barca <hamilcar@tld.always.invalid> writes:
>In article <Pine.SOL.4.58.0412171225480.19783@zaphod> (Fri, 17 Dec 2004
>20:31:44 +0000), Rich Teer wrote:
>> is the fact that most of them run Linux on proprietory x86-based
>> machines
>It is? There are CPUs from Intel, AMD, Cyrix (National Semiconductor?),
>VIA, and TransMeta. Are these proprietary.
You seem to have missed the wars about this; there are several
reasons why others can make x86 CPUs; one being the near monopoly
Intel has as well as some licensing deals Intel did in the past.
Certainly Intel felt they owned x86; they did sue AMD.
>> yet don't give Sun credit for their SPARC CPU, which is an
>> open standard.
>Who's the accrediting body?
SPARC international.
There are several different SPARC implementation (Sun, Fujitsu)
Casper
--
Expressed in this posting are my opinions. They are in no way related
to opinions held by my employer, Sun Microsystems.
Statements on Sun products included here are not gospel and may
be fiction rather than truth.
|
|
|
0
|
|
|
|
Reply
|
Casper
|
12/18/2004 8:29:15 AM
|
|
Hamilcar Barca <hamilcar@tld.always.invalid> writes:
>Perhaps it was. A similar ending was found in the story of Kodak v. Sun.
>No issues of any kind were decided in either of these situations, except
>that of corporations doing business.
But the fact remains that the GPL restricts people from using "free"
software; you cannot dispute that because I have proven that.
It does not need a completed lawsuit, the mere threat of one is sometimes
sufficient.
Casper
--
Expressed in this posting are my opinions. They are in no way related
to opinions held by my employer, Sun Microsystems.
Statements on Sun products included here are not gospel and may
be fiction rather than truth.
|
|
|
0
|
|
|
|
Reply
|
Casper
|
12/18/2004 8:35:24 AM
|
|
In article <pJ6dnWOB2Y5HPF7cRVn-sQ@comcast.com>,
=?iso-8859-1?Q?Lin=F8nut?= <=?iso-8859-1?Q?lin=F8nut?=@bone.com> writes:
> Rich Teer poked his little head through the XP firewall and said:
[...]
> Maybe. Except that chip fabrication is not yet something that can be done
> by individuals, though software is amenable to solo artists.
I'd guess that there are some firms that fabricate low-volume custom chips
for others. I know for a fact that even 20+ years ago, some universities
would routinely have student designs put on silicon. So while fabrication
may not be directly possible for an individual, in principle, design and
contracted-out fabrication would be, and depending on whether there's some
usable free chip-design (and simulation) software compatible with what the
fabs accept for manufacturing, it might not even take an incredibly
wealthy person to have a wafer's worth made up.
>> (I hasten to add that those running Linux on SPARC are NOT hypocrites,
>
> Why not? They may be running on an "open CPU", as you seem to be claiming,
> but there are still many other proprietary things that they object to.
>
>> even though I personally don't see the point of doing so.)
>
> Why not? Is the SPARC optimized to run Sun operating systems?
Not so much that as that gcc doesn't generate particularly well optimized
code for SPARC; that's particularly important on a RISC processor, where
much of the point of the design is to trade off a more difficult
compile-time job for less run-time overhead.
(It's also a pain that there's no C++ ABI; i.e. straight C will link
between gcc and Sun's (or some other) compiler, but C++ typically won't.)
One could also just as easily say that since x86 is the primary Linux
platform, unless some computer vendor that uses a different processor
family has put great effort into optimizing the Linux port for that family,
Linux is likely to be more optimized to run on x86 than on that other
family.
--
mailto:rlhamil@smart.net http://www.smart.net/~rlhamil
Lasik/PRK theme music:
"In the Hall of the Mountain King", from "Peer Gynt"
|
|
|
0
|
|
|
|
Reply
|
Richard
|
12/18/2004 9:56:24 AM
|
|
Casper H.S. Dik <Casper.Dik@Sun.COM> writes:
>Hamilcar Barca <hamilcar@tld.always.invalid> writes:
>
>[ Quotes deleted ]
>> -- Eric S. Raymond
>
>>Sun's new CDDL seems to be nothing more than a marketing announcement.
>
>
>Certainly you don't think that any European takes Eric Raymond seriously
>as he completely fails to understand how cathedrals were build?
Actually, I don't take him seriously because he's madder than a box of frogs.
--
"The road to Paradise is through Intercourse."
[email me at huge [at] huge [dot] org [dot] uk]
|
|
|
0
|
|
|
|
Reply
|
huge
|
12/18/2004 10:17:27 AM
|
|
Richard.L.Hamilton@mindwarp.smart.net (Richard L. Hamilton) writes:
>(It's also a pain that there's no C++ ABI; i.e. straight C will link
>between gcc and Sun's (or some other) compiler, but C++ typically won't.)
C++ is a pain. Period. Avoid it. (And there another problem disappaers).
What's worse, there's no stable C++ ABI between gcc versions and
Sun C++ major versions (though there's a backward compatibility switch)
Casper
--
Expressed in this posting are my opinions. They are in no way related
to opinions held by my employer, Sun Microsystems.
Statements on Sun products included here are not gospel and may
be fiction rather than truth.
|
|
|
0
|
|
|
|
Reply
|
Casper
|
12/18/2004 10:20:05 AM
|
|
Peter =?UTF-8?B?S8O2aGxtYW5u?= <Peter.Koehlmann@t-online.de> writes:
>begin Rich Teer wrote:
>
>> On Fri, 17 Dec 2004, Casper H.S. Dik wrote:
>>
>>> Proprietary has *no* meaning in the context of source code and the
>>> concept of proprietary (meaning one of a kind, vendor specific like
>>> Windows, VMS, AOS/VS) was used to set apart the "Open Standard"
>>> based (read: Unix) based implementations.
>>>
>>> Proprietary was only recently hijacked by the open source proponents
>>> *when they went commercial*.
>>
>> What I find most amusing in these rants by GPL worhipers, who openly
>> despise anything they deem as "proprietory", is the fact that most of
>> them run Linux on proprietory x86-based machines yet don't give Sun
>> credit for their SPARC CPU, which is an open standard.
>
>Idiot
Indeed you are. But did you have something to post other than your sig?
--
"The road to Paradise is through Intercourse."
[email me at huge [at] huge [dot] org [dot] uk]
|
|
|
0
|
|
|
|
Reply
|
huge
|
12/18/2004 10:20:53 AM
|
|
begin Casper H.S. Dik wrote:
> Hamilcar Barca <hamilcar@tld.always.invalid> writes:
>
>>Perhaps it was. A similar ending was found in the story of Kodak v. Sun.
>>No issues of any kind were decided in either of these situations, except
>>that of corporations doing business.
>
> But the fact remains that the GPL restricts people from using "free"
> software; you cannot dispute that because I have proven that.
>
> It does not need a completed lawsuit, the mere threat of one is sometimes
> sufficient.
>
You guys always fail to mention that GPL restricts people from using free
software when they don't adhere to the licence. Otherwise, no, it does not
restrict them. And again: Nobody forces anyone to use GPLed code. If
you /want/ to use it, then by all means do it. And adhere to its licence
--
All things are possible, except skiing thru a revolving door.
|
|
|
0
|
|
|
|
Reply
|
Peter
|
12/18/2004 10:22:16 AM
|
|
Peter =?UTF-8?B?S8O2aGxtYW5u?= <Peter.Koehlmann@t-online.de> writes:
>begin Rich Teer wrote:
[20 lines snipped]
>> Or perhaps, as is becoming increasingly evidient, you're not
>> capable of holding a serious debate with anyone who's views
>> differ from your own, and in fact, you're a troll.
>>
>
>
>As already said, you are an idiot
Oh, the irony.
--
"The road to Paradise is through Intercourse."
[email me at huge [at] huge [dot] org [dot] uk]
|
|
|
0
|
|
|
|
Reply
|
huge
|
12/18/2004 10:22:18 AM
|
|
Peter =?UTF-8?B?S8O2aGxtYW5u?= <Peter.Koehlmann@t-online.de> writes:
[17 lines snipped]
>You may continue to think that SPARC is "open" and x86 is "closed"
>You just reinforce the appearance of a cretin too dumb to spill abucket of
>water
Except he's right and you're wrong.
And you're posting non-ASCII characrters in Usenet headers.
And changing the Newsgroups line without saying so.
All of which makes you a moron, as well as a troll.
*plonk*
--
"The road to Paradise is through Intercourse."
[email me at huge [at] huge [dot] org [dot] uk]
|
|
|
0
|
|
|
|
Reply
|
huge
|
12/18/2004 10:24:49 AM
|
|
In article <20041218030016.752$M5@news.newsreader.com>,
Hamilcar Barca <hamilcar@tld.always.invalid> wrote:
>In article <Pine.SOL.4.58.0412171225480.19783@zaphod> (Fri, 17 Dec 2004
>20:31:44 +0000), Rich Teer wrote:
>
>> is the fact that most of them run Linux on proprietory x86-based
>> machines
>
>It is? There are CPUs from Intel, AMD, Cyrix (National Semiconductor?),
>VIA, and TransMeta. Are these proprietary.
The fact that star is able to read and write a subset of the proprietray
GNU tar format does not make the GNU tar archive format open.....
in most cases it is even esier to reverse engineer the format than to
look into the source! Well this is a result of the fact that there is not
much struture in the GNU tar source.
The fact that somtheing like the TransMeta CPU exists is not a fact of x86 being
open but just because it is possible to reverse engineer.
--
EMail:joerg@schily.isdn.cs.tu-berlin.de (home) J�rg Schilling D-13353 Berlin
js@cs.tu-berlin.de (uni) If you don't have iso-8859-1
schilling@fokus.fraunhofer.de (work) chars I am J"org Schilling
URL: http://www.fokus.fraunhofer.de/usr/schilling ftp://ftp.berlios.de/pub/schily
|
|
|
0
|
|
|
|
Reply
|
js
|
12/18/2004 10:53:23 AM
|
|
In article <_vednWnKCsBOPl7cRVn-jw@comcast.com>,
=?iso-8859-1?Q?Lin=F8nut?= <=?iso-8859-1?Q?lin=F8nut@bone.com?=> wrote:
>Rich Teer poked his little head through the XP firewall and said:
>> But I'm game: where's your proof that SPARC is not an open standard,
>> and that x86 is not proprietory?
>
>SPARC: http://www.sparc.com/v8.shtml
This proves that Sparc _is_ open Standard.
>x86: http://www.x86.org (site not responding, though)
This does not prove anything except when you are able to prove that
if this site announces new features, Intel will implement them.....
--
EMail:joerg@schily.isdn.cs.tu-berlin.de (home) J�rg Schilling D-13353 Berlin
js@cs.tu-berlin.de (uni) If you don't have iso-8859-1
schilling@fokus.fraunhofer.de (work) chars I am J"org Schilling
URL: http://www.fokus.fraunhofer.de/usr/schilling ftp://ftp.berlios.de/pub/schily
|
|
|
0
|
|
|
|
Reply
|
js
|
12/18/2004 11:24:47 AM
|
|
In comp.unix.solaris Lin�nut <=?iso-8859-1?Q?lin=F8nut?=@bone.com> wrote:
> Richard L. Hamilton poked his little head through the XP firewall and said:
>
>> Bull. Standards are more important than code simply as a matter of
>> engineering discipline. Whether they're all open or all deep dark secrets
>> is secondary, although if the objective is to allow and encourage
>> alternative implementations (rather than have one magic hackfest product
>> rule the world), the standards need to be open.
>
> Code is important. Why? Because, you if have code, you can then build it
> and run it.
>
> If you have an open standard, now you must write code based on that
> standard. Many people cannot write code. Therefore, the open standard does
> them no good, unless someone writes the code for them. If there's a charge
> for that code, that may prevent them from using the code. That is why open
> source is important.
>
> Also, standards are not really easy to test. Code is much easier to test.
>
I have seen *lots* of GPL'ed code, yes I can read it and test it, but
despite it, sometimes I can't use it. why? because it does not follow some
standards. and because of this I can't compile the code and it will take
*too* much of time to fix the code. so, there is useless code witch called
to be open, but it's really proprietary. *GNU* proprietary.
If the code is there, it will *help* to make life simpler, but open standards
are only what counts. if I don't have a documented interface to build my
program against, there is no point to write program at all.
toomas
--
When the only tool you have is a hammer, every problem starts to look
like a nail.
|
|
|
0
|
|
|
|
Reply
|
Toomas
|
12/18/2004 11:28:49 AM
|
|
In article <T82dnZ4DPtn2OF7cRVn-1w@comcast.com>,
=?iso-8859-1?Q?Lin=F8nut?= <=?iso-8859-1?Q?lin=F8nut@bone.com?=> wrote:
>Richard L. Hamilton poked his little head through the XP firewall and said:
>If you have an open standard, now you must write code based on that
>standard. Many people cannot write code. Therefore, the open standard does
>them no good, unless someone writes the code for them. If there's a charge
Open Source does not help those people either.
And charging is completely unrelated....
>Also, standards are not really easy to test. Code is much easier to test.
It is just the other way!
Fact is that code, not based on a standard, in many cases is not even well defined.
The most important issue with a written standard is that it defines how the
implementation is expected to work. This makes it definitely easier to test as you
know what to test.
Let me give an example:
GNU tar did start claiming it is able to do incremental backups and restores
in 1992. This has been done using a proprietary (non standard) implementation.
Even the latest version of GNU tar fails miserably with simple test cases
for backups & restores. It is still completely unproven if is possible to
implement working backups & restores using the GNU tar archive format because
this format is widely undocumented. So we know that it does not work and we
don't know whether the bugs may be fixed at all.
Note that we now have 2004 - 12 years later....
OK, the extensions on the TAR archive format used used by star to implement
true incemental dumps & restores are also proprietary but that are at least
documented. This makes it easier to prove the used method for correctness and
this is a requirement for making it a standard.
Even though a final star version that offers true incemental dumps & restores
has not yet released, I would trust the current version (star-1.5a54) more
than GNU tar....
Fom my point of view, it is important to implement code in a way that would
allow future standardisation if there is not yet an open standard that coveres
the code. A lot of the code in the free software used on Linux (and not already
covered by open standards) does not meet this.
--
EMail:joerg@schily.isdn.cs.tu-berlin.de (home) J�rg Schilling D-13353 Berlin
js@cs.tu-berlin.de (uni) If you don't have iso-8859-1
schilling@fokus.fraunhofer.de (work) chars I am J"org Schilling
URL: http://www.fokus.fraunhofer.de/usr/schilling ftp://ftp.berlios.de/pub/schily
|
|
|
0
|
|
|
|
Reply
|
js
|
12/18/2004 11:44:56 AM
|
|
In article <cq103n$or8$3@anubis.demon.co.uk>, Huge <huge@ukmisc.org.uk> wrote:
>Casper H.S. Dik <Casper.Dik@Sun.COM> writes:
>>Hamilcar Barca <hamilcar@tld.always.invalid> writes:
>>
>>[ Quotes deleted ]
>>> -- Eric S. Raymond
>>
>>>Sun's new CDDL seems to be nothing more than a marketing announcement.
>>
>>
>>Certainly you don't think that any European takes Eric Raymond seriously
>>as he completely fails to understand how cathedrals were build?
>
>Actually, I don't take him seriously because he's madder than a box of frogs.
Here in Europe many people have problems to take his OSS activism for serious
because he is a gun fetishist. OSS and rifles do not fit well together....
--
EMail:joerg@schily.isdn.cs.tu-berlin.de (home) J�rg Schilling D-13353 Berlin
js@cs.tu-berlin.de (uni) If you don't have iso-8859-1
schilling@fokus.fraunhofer.de (work) chars I am J"org Schilling
URL: http://www.fokus.fraunhofer.de/usr/schilling ftp://ftp.berlios.de/pub/schily
|
|
|
0
|
|
|
|
Reply
|
js
|
12/18/2004 11:51:47 AM
|
|
Peter =?UTF-8?B?S8O2aGxtYW5u?= <Peter.Koehlmann@t-online.de> writes:
>You guys always fail to mention that GPL restricts people from using free
>software when they don't adhere to the licence. Otherwise, no, it does not
>restrict them. And again: Nobody forces anyone to use GPLed code. If
>you /want/ to use it, then by all means do it. And adhere to its licence
Then don't claim it's "free". Free doesn't carry cost or restrictions.
It's also not "free" as in "freedom"; BSDL is.
The restrictions it places on the use of "free" software carry a cost;
a cost that some are not willing to pay.
Now, I'm not complaining that GPL'ed software isn't free; I'm just
complaining that saying that it is "free" doesn't make it so.
Casper
--
Expressed in this posting are my opinions. They are in no way related
to opinions held by my employer, Sun Microsystems.
Statements on Sun products included here are not gospel and may
be fiction rather than truth.
|
|
|
0
|
|
|
|
Reply
|
Casper
|
12/18/2004 12:00:51 PM
|
|
js@cs.tu-berlin.de (Joerg Schilling) writes:
>The fact that star is able to read and write a subset of the proprietray
>GNU tar format does not make the GNU tar archive format open.....
>in most cases it is even esier to reverse engineer the format than to
>look into the source! Well this is a result of the fact that there is not
>much struture in the GNU tar source.
Indeed; many protocols are more easily reversed from "the wire" than
the source code.
When I started at Sun I hacked "sadmind/admind"; I spend a few days
to a week trying to figure out the protocol from the source and failed.
In the end it took me less than a day to reverse engineer it from the
wire.
>The fact that somtheing like the TransMeta CPU exists is not a fact of x86 being
>open but just because it is possible to reverse engineer.
There have been lawsuits that proof that assertion: Intel trying
to stop the competition. It's also the sole reason of the Itanic's
existance; make something so weird and so encumbered that it is impossible
to clone.
Casper
--
Expressed in this posting are my opinions. They are in no way related
to opinions held by my employer, Sun Microsystems.
Statements on Sun products included here are not gospel and may
be fiction rather than truth.
|
|
|
0
|
|
|
|
Reply
|
Casper
|
12/18/2004 12:04:26 PM
|
|
In article <cq10hh$or8$6@anubis.demon.co.uk>, Huge <huge@ukmisc.org.uk> wrote:
>Peter =?UTF-8?B?S8O2aGxtYW5u?= <Peter.Koehlmann@t-online.de> writes:
.....
>Except he's right and you're wrong.
>
>And you're posting non-ASCII characrters in Usenet headers.
Be careful, using escape codes to represent non-ASCII characrters
is allowed in headers.
But of course with respect to all other claims from him, he is definitely
wrong and beyond belief.
--
EMail:joerg@schily.isdn.cs.tu-berlin.de (home) J�rg Schilling D-13353 Berlin
js@cs.tu-berlin.de (uni) If you don't have iso-8859-1
schilling@fokus.fraunhofer.de (work) chars I am J"org Schilling
URL: http://www.fokus.fraunhofer.de/usr/schilling ftp://ftp.berlios.de/pub/schily
|
|
|
0
|
|
|
|
Reply
|
js
|
12/18/2004 12:05:53 PM
|
|
js@cs.tu-berlin.de (Joerg Schilling) writes:
>Fom my point of view, it is important to implement code in a way that would
>allow future standardisation if there is not yet an open standard that coveres
>the code. A lot of the code in the free software used on Linux (and not already
>covered by open standards) does not meet this.
See, e.g., the DCE "pthreads" debacle; how come Solaris pthreads aren't
compatible with DCE pthreads? Well, that's because DCE pthreads aren't
pthreads at all.
(At least the people at UI that gave us Solaris thread had the foresight
not to call all functions pthread_* even though they looked remarkably like
a certain incantation of the thread standard)
If you compare the UI/Solaris threads with the POSIX threads you will
notice interesting differences such as all kinds of *_init functions in
pthreads; once you've been doing software engineering for a while
you'll understand that difference and why pthreads are superior even
though UI threads are much easier to use.
Casper
--
Expressed in this posting are my opinions. They are in no way related
to opinions held by my employer, Sun Microsystems.
Statements on Sun products included here are not gospel and may
be fiction rather than truth.
|
|
|
0
|
|
|
|
Reply
|
Casper
|
12/18/2004 12:08:41 PM
|
|
begin Toomas Soome wrote:
> In comp.unix.solaris Linønut <=?iso-8859-1?Q?lin=F8nut?=@bone.com> wrote:
>> Richard L. Hamilton poked his little head through the XP firewall and
>> said:
>>
>>> Bull. Standards are more important than code simply as a matter of
>>> engineering discipline. Whether they're all open or all deep dark
>>> secrets is secondary, although if the objective is to allow and
>>> encourage alternative implementations (rather than have one magic
>>> hackfest product rule the world), the standards need to be open.
>>
>> Code is important. Why? Because, you if have code, you can then build
>> it and run it.
>>
>> If you have an open standard, now you must write code based on that
>> standard. Many people cannot write code. Therefore, the open standard
>> does
>> them no good, unless someone writes the code for them. If there's a
>> charge
>> for that code, that may prevent them from using the code. That is why
>> open source is important.
>>
>> Also, standards are not really easy to test. Code is much easier to
>> test.
>>
>
> I have seen *lots* of GPL'ed code, yes I can read it and test it, but
> despite it, sometimes I can't use it. why? because it does not follow some
> standards. and because of this I can't compile the code
I am certain you have lots and lots of examples where you could not compile
GPLed code *because* it was not following some standard. Lots
> and it will take
> *too* much of time to fix the code. so, there is useless code witch called
> to be open, but it's really proprietary. *GNU* proprietary.
>
> If the code is there, it will *help* to make life simpler, but open
> standards are only what counts. if I don't have a documented interface to
> build my program against, there is no point to write program at all.
>
So you are another Sun-advocate who, for lack of anything remotely
resembling the truth, makes up his horror experiences with the GPL out of
whole cloth
--
What happens if a big asteroid hits Earth? Judging from realistic
simulations involving a sledge hammer and a common laboratory frog,
we can assume it will be pretty bad. --- Dave Barry
|
|
|
0
|
|
|
|
Reply
|
Peter
|
12/18/2004 12:09:23 PM
|
|
begin Huge wrote:
> Peter =?UTF-8?B?S8O2aGxtYW5u?= <Peter.Koehlmann@t-online.de> writes:
>
> [17 lines snipped]
>
>>You may continue to think that SPARC is "open" and x86 is "closed"
>>You just reinforce the appearance of a cretin too dumb to spill abucket of
>>water
>
> Except he's right and you're wrong.
>
If *you* say so it is certainly that way. I am sure of that
> And you're posting non-ASCII characrters in Usenet headers.
>
You are posting drivel. And you have no idea what you are blubbering about
> And changing the Newsgroups line without saying so.
>
Which is exactly the right thing to do. I set a follow-up. Automatically
And you know what: You can't do a thing about it
> All of which makes you a moron, as well as a troll.
>
> *plonk*
>
This coming from someone too afraid people might later see his cretinous
rants and therefor uses X-No-Archive in his headers
You are certainly right. Your posts could as well have been written by a
drunk monkey and so google should not preserve them. Perhaps they are
--
A fool-proof method for sculpting an elephant:
first, get a huge block of marble; then you chip
away everything that doesn't look like an elephant.
|
|
|
0
|
|
|
|
Reply
|
Peter
|
12/18/2004 12:15:15 PM
|
|
In article <T82dnZkDPtlmOF7cRVn-1w@comcast.com>,
=?iso-8859-1?Q?Lin=F8nut?= <=?iso-8859-1?Q?lin=F8nut@bone.com?=> wrote:
>Casper H.S Dik poked his little head through the XP firewall and said:
>
>>>> Becker claims that the GPL forbids the linking of his drivers with a
>>>> closed-source kernel (in this case, Solaris)
>>
>>>Do you have some evidence to support this claim? Even if you do, the
>>>claim is irrelevant. When you stop ranting and actually read the GPL,
>>>you'll find that it does not in any way prohibit linking GPL code with
>>>proprietary code -- it prohibits only the distribution of such a merged
>>>product.
>>
>> Sun released code which allowed our customers to run Linux drivers
>> compiled from source under Solaris; Becker threatened to sue. End of
>> story.
>
>Actually, it sounds too simple. Surely there must be more to the story.
Of course, Becker is just unable to understand the GPL.
Sun did not publish binaries compiled from Beckers sources and for this reason,
Sun did not even make an attempt to link against non OSS code.
Sun did publish the GPLd source for a wrapper (written by Sun) that did allow
people to compile and run Becker's drivers in Solaris x86.
Note that the GPL allowd anybody to do anything with the sources as long
as they don't publish the resulting work. Sun did not publish a resulting
work but a kit that did allow people to do things in their own privacy.
It is intersting to see that People like Becker attack companies like Sun that
respect the GPL but at the same time Veritas has no problems with their
questionable hack on GNU tar....
--
EMail:joerg@schily.isdn.cs.tu-berlin.de (home) J�rg Schilling D-13353 Berlin
js@cs.tu-berlin.de (uni) If you don't have iso-8859-1
schilling@fokus.fraunhofer.de (work) chars I am J"org Schilling
URL: http://www.fokus.fraunhofer.de/usr/schilling ftp://ftp.berlios.de/pub/schily
|
|
|
0
|
|
|
|
Reply
|
js
|
12/18/2004 12:35:40 PM
|
|
In article <y4mdnVaR-6qvNl7cRVn-jg@comcast.com>,
=?iso-8859-1?Q?Lin=F8nut?= <=?iso-8859-1?Q?lin=F8nut@bone.com?=> wrote:
>Richard L. Hamilton poked his little head through the XP firewall and said:
>
>>> That's even more juvenile as the "commie" insult. The Sun worshipers are
>>> becoming as obsessed as the Steve Jobs personality cult.
>>
>> Yeah, both like something that's better than Linux!
>
>For some meanings of "better". I'm disinclined, for example, to download
>Solaris and install it on one of my x86 boxes, because of its poor
>reputation on that hardware.
Do you really run such poor exotic Hardware?
Of do you are just too lazy to test a recent version of Solaris?
It is interesting to see that because it Linux zelots did act the same way
on Linux, Linux currently would have no more than 1000 users.
It is many years ago, that I really had problems with hardware
support for X86 from Solaris.
>Maybe that reputation is now better. So, where can I get Solaris to
>download and try? Ah, found it. Damn, now I can't rag on Sun until I
>actually try this out. Hmmmm, I only have one box with 512Mb on it, and
>that's this one. Still, maybe it's worth trying.
Solaris even works on boxes with much less memory.
>Damn, what an ordeal just to get to the license!
What is your problem?
As it is a matter of fact that > 90% of the Linux users also run Win32,
most people do not even have problems with mush more restrictive licenses.
Note that this is for the _current_ _betas_, not for the OpenSolaris
you will see soon.
--
EMail:joerg@schily.isdn.cs.tu-berlin.de (home) J�rg Schilling D-13353 Berlin
js@cs.tu-berlin.de (uni) If you don't have iso-8859-1
schilling@fokus.fraunhofer.de (work) chars I am J"org Schilling
URL: http://www.fokus.fraunhofer.de/usr/schilling ftp://ftp.berlios.de/pub/schily
|
|
|
0
|
|
|
|
Reply
|
js
|
12/18/2004 12:42:21 PM
|
|
In article <20041218025039.103$GM@news.newsreader.com>,
Hamilcar Barca <hamilcar@tld.always.invalid> wrote:
>In article <41c2f02d$0$78228$e4fe514c@news.xs4all.nl> (Fri, 17 Dec 2004
>14:41:49 +0000), Casper H. S. Dik wrote:
>
>>>I assume that when Mr. Schwartz writes "proprietary", he mean to write
>>>"100% licensed under the GPL". How non-proprietary is Solaris? It's
>>>been 100% closed since the day Sun (legally) combined BSD and System V.
>>
>> "proprietary" means "not based on open standards"; that's what it always
>> meant until it was hijacked by the open source crowd.
>
>Sure, just like DEC's claim that renaming VMS to OpenVMS was justified
>because running on two single-source platforms was the "definition of open."
>Proprietary means closed.
It looks like you fail to understand the word open :-(
If you did understand what's going on with OpenSolaris, you would not make
such rediculous comparison.
--
EMail:joerg@schily.isdn.cs.tu-berlin.de (home) J�rg Schilling D-13353 Berlin
js@cs.tu-berlin.de (uni) If you don't have iso-8859-1
schilling@fokus.fraunhofer.de (work) chars I am J"org Schilling
URL: http://www.fokus.fraunhofer.de/usr/schilling ftp://ftp.berlios.de/pub/schily
|
|
|
0
|
|
|
|
Reply
|
js
|
12/18/2004 12:43:59 PM
|
|
In article <41c3ebcc$0$85990$e4fe514c@news.xs4all.nl>,
Casper H.S. Dik <Casper.Dik@Sun.COM> wrote:
>Hamilcar Barca <hamilcar@tld.always.invalid> writes:
>
>>Perhaps it was. A similar ending was found in the story of Kodak v. Sun.
>>No issues of any kind were decided in either of these situations, except
>>that of corporations doing business.
>
>But the fact remains that the GPL restricts people from using "free"
>software; you cannot dispute that because I have proven that.
When trying to analyse the postings from GPL zelots, it seems that RMS did
do a really good marketing job with redefining the meaning of the word "free"
and "proprietary". It seems that most GPL zelots do ot even remember the
official meaning anymore :-(
--
EMail:joerg@schily.isdn.cs.tu-berlin.de (home) J�rg Schilling D-13353 Berlin
js@cs.tu-berlin.de (uni) If you don't have iso-8859-1
schilling@fokus.fraunhofer.de (work) chars I am J"org Schilling
URL: http://www.fokus.fraunhofer.de/usr/schilling ftp://ftp.berlios.de/pub/schily
|
|
|
0
|
|
|
|
Reply
|
js
|
12/18/2004 12:49:54 PM
|
|
In article <10s7vm8q8lseu1d@corp.supernews.com>,
Richard L. Hamilton <Richard.L.Hamilton@mindwarp.smart.net> wrote:
>In article <pJ6dnWOB2Y5HPF7cRVn-sQ@comcast.com>,
> =?iso-8859-1?Q?Lin=F8nut?= <=?iso-8859-1?Q?lin=F8nut?=@bone.com> writes:
>> Rich Teer poked his little head through the XP firewall and said:
>[...]
>> Maybe. Except that chip fabrication is not yet something that can be done
>> by individuals, though software is amenable to solo artists.
>
>I'd guess that there are some firms that fabricate low-volume custom chips
>for others. I know for a fact that even 20+ years ago, some universities
>would routinely have student designs put on silicon. So while fabrication
If you have access to the silicon bakery os a nuniversity.....
But sometimes (e.g. with the TU-Berlin) there is even the will to make the
silicon bakery of the university accessible to start up companies.
If you don't have the right connections to a local silicon bakery, you fail.
>> Why not? Is the SPARC optimized to run Sun operating systems?
>
>Not so much that as that gcc doesn't generate particularly well optimized
>code for SPARC; that's particularly important on a RISC processor, where
>much of the point of the design is to trade off a more difficult
>compile-time job for less run-time overhead.
This is only a result of the fact that the GCC developers do not put much
effort in Sparc optmimization, not a proof for Sparc being not open.
....
>One could also just as easily say that since x86 is the primary Linux
>platform, unless some computer vendor that uses a different processor
>family has put great effort into optimizing the Linux port for that family,
>Linux is likely to be more optimized to run on x86 than on that other
>family.
Then tell me why gcc in most cases creates less performant x86 code than
Sun's C-compiler.
--
EMail:joerg@schily.isdn.cs.tu-berlin.de (home) J�rg Schilling D-13353 Berlin
js@cs.tu-berlin.de (uni) If you don't have iso-8859-1
schilling@fokus.fraunhofer.de (work) chars I am J"org Schilling
URL: http://www.fokus.fraunhofer.de/usr/schilling ftp://ftp.berlios.de/pub/schily
|
|
|
0
|
|
|
|
Reply
|
js
|
12/18/2004 12:55:39 PM
|
|
In article <41c41bf3$0$90761$e4fe514c@news.xs4all.nl>,
Casper H.S. Dik <Casper.Dik@Sun.COM> wrote:
>The restrictions it places on the use of "free" software carry a cost;
>a cost that some are not willing to pay.
>
>Now, I'm not complaining that GPL'ed software isn't free; I'm just
>complaining that saying that it is "free" doesn't make it so.
I would call this effective marketing ,-)
--
EMail:joerg@schily.isdn.cs.tu-berlin.de (home) J�rg Schilling D-13353 Berlin
js@cs.tu-berlin.de (uni) If you don't have iso-8859-1
schilling@fokus.fraunhofer.de (work) chars I am J"org Schilling
URL: http://www.fokus.fraunhofer.de/usr/schilling ftp://ftp.berlios.de/pub/schily
|
|
|
0
|
|
|
|
Reply
|
js
|
12/18/2004 12:59:05 PM
|
|
js@cs.tu-berlin.de (Joerg Schilling) writes:
>In article <cq10hh$or8$6@anubis.demon.co.uk>, Huge <huge@ukmisc.org.uk> wrote:
>>Peter =?UTF-8?B?S8O2aGxtYW5u?= <Peter.Koehlmann@t-online.de> writes:
>
>.....
>
>>Except he's right and you're wrong.
>>
>>And you're posting non-ASCII characrters in Usenet headers.
>
>Be careful, using escape codes to represent non-ASCII characrters
>is allowed in headers.
Looks like a duck, walks like a duck, quacks like a duck - it's a duck.
>But of course with respect to all other claims from him, he is definitely
>wrong and beyond belief.
Quite.
--
"The road to Paradise is through Intercourse."
[email me at huge [at] huge [dot] org [dot] uk]
|
|
|
0
|
|
|
|
Reply
|
huge
|
12/18/2004 1:11:12 PM
|
|
js@cs.tu-berlin.de (Joerg Schilling) writes:
>Here in Europe many people have problems to take his OSS activism for serious
>because he is a gun fetishist. OSS and rifles do not fit well together....
Nah, all Americans are gun fetishist[1]. You've just upset all
the gun toting OSSers; better not fly their for a bit :-)
Casper
[1] Fortunately, the Americans likely to be upset by being tarred with this
brush are not likely to own guns.
--
Expressed in this posting are my opinions. They are in no way related
to opinions held by my employer, Sun Microsystems.
Statements on Sun products included here are not gospel and may
be fiction rather than truth.
|
|
|
0
|
|
|
|
Reply
|
Casper
|
12/18/2004 1:14:11 PM
|
|
js@cs.tu-berlin.de (Joerg Schilling) writes:
>>Maybe that reputation is now better. So, where can I get Solaris to
>>download and try? Ah, found it. Damn, now I can't rag on Sun until I
>>actually try this out. Hmmmm, I only have one box with 512Mb on it, and
>>that's this one. Still, maybe it's worth trying.
>Solaris even works on boxes with much less memory.
See, e.g.,
http://blogs.sun.com/roller/page/casper/20041129#fujtisu_lifebook_b112_running_solaris
>Note that this is for the _current_ _betas_, not for the OpenSolaris
>you will see soon.
And not the same for the free binary license.
Casper
--
Expressed in this posting are my opinions. They are in no way related
to opinions held by my employer, Sun Microsystems.
Statements on Sun products included here are not gospel and may
be fiction rather than truth.
|
|
|
0
|
|
|
|
Reply
|
Casper
|
12/18/2004 1:17:20 PM
|
|
In comp.unix.solaris Peter K�hlmann <Peter.Koehlmann@t-online.de> wrote:
>> I have seen *lots* of GPL'ed code, yes I can read it and test it, but
>> despite it, sometimes I can't use it. why? because it does not follow some
>> standards. and because of this I can't compile the code
>
> I am certain you have lots and lots of examples where you could not compile
> GPLed code *because* it was not following some standard. Lots
>
>> and it will take
>> *too* much of time to fix the code. so, there is useless code witch called
>> to be open, but it's really proprietary. *GNU* proprietary.
>>
>> If the code is there, it will *help* to make life simpler, but open
>> standards are only what counts. if I don't have a documented interface to
>> build my program against, there is no point to write program at all.
>>
>
> So you are another Sun-advocate who, for lack of anything remotely
> resembling the truth, makes up his horror experiences with the GPL out of
> whole cloth
oh, it has nothing to do with Sun. I'm happy with some things about Sun and
not happy at all with others. And the very same will apply to GNU/GPL
as well - I'm working in IT since 94, most time in university and I have
spent quite many hours with free software. I really hate to see it's
becoming more and more "free".
the point behind of free software is to make it avalable for everyone.
despite ths OS you are using, despite the compiler you are using.
if you are trying to lock me into some OS (be it solaris, linux, windows,...),
it's not a free software any more, even if it will include GPL as a license
or sources I can read. thanks, but no thanks, it's not better in any way than
alternatives offered by Sun or Microsoft or whomever.
toomas
--
A chicken is an egg's way of producing more eggs.
|
|
|
0
|
|
|
|
Reply
|
Toomas
|
12/18/2004 1:39:11 PM
|
|
In article <cq15kj$bs9$1@news.cs.tu-berlin.de>,
js@cs.tu-berlin.de (Joerg Schilling) writes:
> In article <cq103n$or8$3@anubis.demon.co.uk>, Huge <huge@ukmisc.org.uk> wrote:
>>Casper H.S. Dik <Casper.Dik@Sun.COM> writes:
>>>Hamilcar Barca <hamilcar@tld.always.invalid> writes:
>>>
>>>[ Quotes deleted ]
>>>> -- Eric S. Raymond
>>>
>>>>Sun's new CDDL seems to be nothing more than a marketing announcement.
>>>
>>>
>>>Certainly you don't think that any European takes Eric Raymond seriously
>>>as he completely fails to understand how cathedrals were build?
>>
>>Actually, I don't take him seriously because he's madder than a box of frogs.
>
> Here in Europe many people have problems to take his OSS activism for serious
> because he is a gun fetishist. OSS and rifles do not fit well together....
Why? Is there some technical reason that a computer-controlled rifle
couldn't run a stripped down version of Linux? :-)
(I thought I was being funny, and then I remembered reading about
www.live-shot.com; although I have no idea whether or not they run Linux.)
Seriously, one could argue that both are freedoms that help resist
domination, whether by particularly aggressive businesses like MS, or as
something that by design was the last-resort check and balance against the
sort of power that governments have held.
I'm sure there are counter-arguments; and I'm not trying to start
something. I'm just saying there's more than one way of looking at it.
--
mailto:rlhamil@smart.net http://www.smart.net/~rlhamil
Lasik/PRK theme music:
"In the Hall of the Mountain King", from "Peer Gynt"
|
|
|
0
|
|
|
|
Reply
|
Richard
|
12/18/2004 3:19:17 PM
|
|
In article <10s8ijle8pu15d0@corp.supernews.com>,
Richard L. Hamilton <Richard.L.Hamilton@mindwarp.smart.net> wrote:
>>>>Certainly you don't think that any European takes Eric Raymond seriously
>>>>as he completely fails to understand how cathedrals were build?
>>>
>>>Actually, I don't take him seriously because he's madder than a box of frogs.
>>
>> Here in Europe many people have problems to take his OSS activism for serious
>> because he is a gun fetishist. OSS and rifles do not fit well together....
>
>Why? Is there some technical reason that a computer-controlled rifle
>couldn't run a stripped down version of Linux? :-)
No, the answer is much simpler:
OSS gives you freedom while weapons (in the hand of others) take your freedom.
So what does ESR like?
- fight for your freedom
- threaten your freedom with his guns?
BTW: even weapons in your hand have nothing in common with freedom. They just
give you force about others. If you don't like to threaten the freedom of others
you don't need weapons. What kind of threat weapons (specially in the USA) press
on others may easily be proved by comparing fatalities caused by weapons in the
USA to the numbers in the rest of the world.
--
EMail:joerg@schily.isdn.cs.tu-berlin.de (home) J�rg Schilling D-13353 Berlin
js@cs.tu-berlin.de (uni) If you don't have iso-8859-1
schilling@fokus.fraunhofer.de (work) chars I am J"org Schilling
URL: http://www.fokus.fraunhofer.de/usr/schilling ftp://ftp.berlios.de/pub/schily
|
|
|
0
|
|
|
|
Reply
|
js
|
12/18/2004 3:42:27 PM
|
|
Richard L. Hamilton poked his little head through the XP firewall and said:
>> Why not? Is the SPARC optimized to run Sun operating systems?
>
> Not so much that as that gcc doesn't generate particularly well optimized
> code for SPARC; that's particularly important on a RISC processor, where
> much of the point of the design is to trade off a more difficult
> compile-time job for less run-time overhead.
Yeah, that does make sense. RISC puts great demands on the compiler.
> (It's also a pain that there's no C++ ABI; i.e. straight C will link
> between gcc and Sun's (or some other) compiler, but C++ typically won't.)
Why can't we have a name-mangling standard??? <grin>
> One could also just as easily say that since x86 is the primary Linux
> platform, unless some computer vendor that uses a different processor
> family has put great effort into optimizing the Linux port for that family,
> Linux is likely to be more optimized to run on x86 than on that other
> family.
That might be true. But I suspect that the optimization areas depend on the
chip being supported. I also suspect that features of the x86 (limited
numbers of registers) hurt it. I say "suspect" because this is an area I
know little about anymore.
--
Linux, because a CPU is a terrible thing to waste.
|
|
|
0
|
|
|
|
Reply
|
iso
|
12/18/2004 3:51:54 PM
|
|
Casper H.S Dik poked his little head through the XP firewall and said:
> Richard.L.Hamilton@mindwarp.smart.net (Richard L. Hamilton) writes:
>
>>(It's also a pain that there's no C++ ABI; i.e. straight C will link
>>between gcc and Sun's (or some other) compiler, but C++ typically won't.)
>
> C++ is a pain. Period. Avoid it. (And there another problem disappaers).
I must love pain, then. C++ and it's accompanying libraries (STL), along
with supplements such as blitz and boost, make it a pretty powerful way to
write code.
Having said that, I've lately adopted the habit of writing some very basic C
code and global functions taking pointers to structures that do things to
those structures, to support those who prefer a strict C interface. Then I
wrap them in a C++ class where it makes sense.
> What's worse, there's no stable C++ ABI between gcc versions and
> Sun C++ major versions (though there's a backward compatibility switch)
I know you'll blanch at this solution, but what the heck -- autoconf tools.
That's one reason I've downloaded Solaris 10 and am now trying to figure how
to get some hardware to load it onto. So I can make sure my code builds and
runs on it. What the heck, I'm an old guy in programmer's years, but I can
still learn.
--
Linux, because a CPU is a terrible thing to waste.
|
|
|
0
|
|
|
|
Reply
|
iso
|
12/18/2004 3:58:37 PM
|
|
Joerg Schilling poked his little head through the XP firewall and said:
>>One could also just as easily say that since x86 is the primary Linux
>>platform, unless some computer vendor that uses a different processor
>>family has put great effort into optimizing the Linux port for that family,
>>Linux is likely to be more optimized to run on x86 than on that other
>>family.
>
> Then tell me why gcc in most cases creates less performant x86 code than
> Sun's C-compiler.
glibc?
--
Linux, because a CPU is a terrible thing to waste.
|
|
|
0
|
|
|
|
Reply
|
iso
|
12/18/2004 4:01:11 PM
|
|
Joerg Schilling poked his little head through the XP firewall and said:
> In article <_vednWnKCsBOPl7cRVn-jw@comcast.com>,
>=?iso-8859-1?Q?Lin=F8nut?= <=?iso-8859-1?Q?lin=F8nut@bone.com?=> wrote:
>>Rich Teer poked his little head through the XP firewall and said:
>
>>> But I'm game: where's your proof that SPARC is not an open standard,
>>> and that x86 is not proprietory?
>>
>>SPARC: http://www.sparc.com/v8.shtml
>
> This proves that Sparc _is_ open Standard.
>
>>x86: http://www.x86.org (site not responding, though)
>
> This does not prove anything except when you are able to prove that
> if this site announces new features, Intel will implement them.....
Not necessary. As long as most features of the chip are well known, then
AMD and others can implement work-alike chips, and the architecture is de
facto open, and, if Intel drops the line, it will carry on.
Anyway, please don't think I'm a big fan of x86. It works well enough, I
can get the machines cheap (in fact, I got one for free from a guy who found
his computer unusable with Windows, and now it makes a nice Linux platform
for my old SCSI-lite scanner), and it has OS's that are well-supported.
That's all I ask.
--
Linux, because a CPU is a terrible thing to waste.
|
|
|
0
|
|
|
|
Reply
|
iso
|
12/18/2004 4:06:39 PM
|
|
Toomas Soome poked his little head through the XP firewall and said:
>> Code is important. Why? Because, you if have code, you can then build it
>> and run it.
>
> I have seen *lots* of GPL'ed code, yes I can read it and test it, but
> despite it, sometimes I can't use it. why? because it does not follow some
> standards. and because of this I can't compile the code and it will take
> *too* much of time to fix the code. so, there is useless code witch called
> to be open, but it's really proprietary. *GNU* proprietary.
That's because programmers usually glom onto "solutions" too quickly.
Don't use GNU extensions. Don't use Microsoft APIs. If you do use them,
write wrapper functions and use automake/autoconf, or at least some macros,
to adjust them to your local architecture.
> If the code is there, it will *help* to make life simpler, but open standards
> are only what counts. if I don't have a documented interface to build my
> program against, there is no point to write program at all.
What an extreme attitude. An attitude that, I suspect, is belied by a fair
number of open-source projects out there. How else do you explain cdoe
forking?
I'm not saying code is all. Documents make it easy to grasp a design, by
far. But, having worked through a number of design documents myself, what
do I find? I find that the developers (myself included) go through much
argument, discussion, and discovery as we write the code. Then we have to
go back to the design documents and spend some time "in transition" updating
the design documents to reflect the reality that was hammered out in the
lab.
--
Linux, because a CPU is a terrible thing to waste.
|
|
|
0
|
|
|
|
Reply
|
iso
|
12/18/2004 4:14:18 PM
|
|
Joerg Schilling poked his little head through the XP firewall and said:
> In article <T82dnZ4DPtn2OF7cRVn-1w@comcast.com>,
>=?iso-8859-1?Q?Lin=F8nut?= <=?iso-8859-1?Q?lin=F8nut@bone.com?=> wrote:
>>Richard L. Hamilton poked his little head through the XP firewall and said:
>
>>If you have an open standard, now you must write code based on that
>>standard. Many people cannot write code. Therefore, the open standard does
>>them no good, unless someone writes the code for them. If there's a charge
>
> Open Source does not help those people either.
Sure it does. It provides a much larger body of people willing to modify or
write the code.
> And charging is completely unrelated....
No it isn't. It often makes the difference between a solution being
available or unavailable.
You think I'd have all these servers (ssh, samba, apache, vnc, ntp) running
on my machine if I had to pay even $50 for each one)?
>>Also, standards are not really easy to test. Code is much easier to test.
>
> It is just the other way!
>
> Fact is that code, not based on a standard, in many cases is not even well
> defined. The most important issue with a written standard is that it
> defines how the implementation is expected to work. This makes it
> definitely easier to test as you know what to test.
I agree, in some cases <grin>.
> Let me give an example:
>
> GNU tar did start claiming it is able to do incremental backups and restores
> in 1992. This has been done using a proprietary (non standard) implementation.
>
> Even the latest version of GNU tar fails miserably with simple test cases
> for backups & restores. It is still completely unproven if is possible to
> implement working backups & restores using the GNU tar archive format because
> this format is widely undocumented. So we know that it does not work and we
> don't know whether the bugs may be fixed at all.
Interesting. I'd like to look into that at some point.
So "tar" is an app that grew, undesigned. Sounds like the great majority of
UNIX command-line commands, to my limited view.
Thanks for the discussion!
--
Linux, because a CPU is a terrible thing to waste.
|
|
|
0
|
|
|
|
Reply
|
iso
|
12/18/2004 4:22:27 PM
|
|
Casper H.S Dik poked his little head through the XP firewall and said:
> If you compare the UI/Solaris threads with the POSIX threads you will
> notice interesting differences such as all kinds of *_init functions in
> pthreads; once you've been doing software engineering for a while
> you'll understand that difference and why pthreads are superior even
> though UI threads are much easier to use.
I've gotten into POSIX threads lately. Cool! Much better than Windows
threads.
Of course, I've had to wrapper them so that my code will work on both
platforms.
--
Linux, because a CPU is a terrible thing to waste.
|
|
|
0
|
|
|
|
Reply
|
iso
|
12/18/2004 4:24:01 PM
|
|
In article <wLydnXdU57fayVncRVn-vw@comcast.com>,
=?iso-8859-1?Q?Lin=F8nut?= <=?iso-8859-1?Q?lin=F8nut@bone.com?=> wrote:
>Joerg Schilling poked his little head through the XP firewall and said:
>
>>>One could also just as easily say that since x86 is the primary Linux
>>>platform, unless some computer vendor that uses a different processor
>>>family has put great effort into optimizing the Linux port for that family,
>>>Linux is likely to be more optimized to run on x86 than on that other
>>>family.
>>
>> Then tell me why gcc in most cases creates less performant x86 code than
>> Sun's C-compiler.
>
>glibc?
No, if I check performance with my Reed Solomon coder, there is no libc code
involved. Sun's cc gives 30% more performance (even on x86) then gcc compiled
code does.
--
EMail:joerg@schily.isdn.cs.tu-berlin.de (home) J�rg Schilling D-13353 Berlin
js@cs.tu-berlin.de (uni) If you don't have iso-8859-1
schilling@fokus.fraunhofer.de (work) chars I am J"org Schilling
URL: http://www.fokus.fraunhofer.de/usr/schilling ftp://ftp.berlios.de/pub/schily
|
|
|
0
|
|
|
|
Reply
|
js
|
12/18/2004 5:11:54 PM
|
|
-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1
On Fri, 17 Dec 2004 20:31:44 GMT,
Rich Teer <rich.teer@rite-group.com> wrote:
> On Fri, 17 Dec 2004, Casper H.S. Dik wrote:
>
>> Proprietary has *no* meaning in the context of source code and the
>> concept of proprietary (meaning one of a kind, vendor specific like
>> Windows, VMS, AOS/VS) was used to set apart the "Open Standard"
>> based (read: Unix) based implementations.
>>
>> Proprietary was only recently hijacked by the open source proponents
>> *when they went commercial*.
>
> What I find most amusing in these rants by GPL worhipers, who openly
> despise anything they deem as "proprietory", is the fact that most of
> them run Linux on proprietory x86-based machines yet don't give Sun
> credit for their SPARC CPU, which is an open standard. Essentially,
> they're hypocrites. (I hasten to add that those running Linux on SPARC
> are NOT hypocrites, even though I personally don't see the point of doing
> so.)
>
Is SPARC an open source CPU? or is it licenced on a per manufacturer
basis? Can I go into fab on an UltraSPARC III without talking to SUN
about it? I know that the CPUs are available from a number of companies
(Fujitsu among them) but I thought it was simply widely licenced, like
the ARM series.
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.2.5 (GNU/Linux)
iD8DBQFBxHkad90bcYOAWPYRAtmBAKCgGomPF3itGe1Se/3dBsCNGNZc0ACeInHI
hQiKAjxTIoNO0IPR0CxjyMw=
=2sL8
-----END PGP SIGNATURE-----
--
Jim Richardson http://www.eskimo.com/~warlock
It's a very *__UN*lucky week in which to be took dead.
-- Churchy La Femme
|
|
|
0
|
|
|
|
Reply
|
Jim
|
12/18/2004 7:10:41 PM
|
|
John Bailo wrote:
> Having used Solaris.
>
> I doubt it.
APPARENTLY you haven't used it MUCH.
|
|
|
0
|
|
|
|
Reply
|
UNIX
|
12/18/2004 7:42:10 PM
|
|
On Sat, 18 Dec 2004, Jim Richardson wrote:
> Is SPARC an open source CPU? or is it licenced on a per manufacturer
In short, both. For a longer (better) answer, please see:
http://www.sparc.org/faq.html
> basis? Can I go into fab on an UltraSPARC III without talking to SUN
Dunno about that. The UltraSPARC III is Sun's implementation of the
SPARC v9 standard, so I suspect that they might have something to say
about that! But, there's nothing preventing you from taking the SPARC v9
standard, and desiging your own, compatible CPU.
--
Rich Teer, SCNA, SCSA, author of "Solaris Systems Programming"
. * * . * .* .
. * . .*
President, * . . /\ ( . . *
Rite Online Inc. . . / .\ . * .
.*. / * \ . .
. /* o \ .
Voice: +1 (250) 979-1638 * '''||''' .
URL: http://www.rite-online.net ******************
|
|
|
0
|
|
|
|
Reply
|
Rich
|
12/18/2004 9:21:30 PM
|
|
Jim Richardson <warlock@eskimo.com> writes:
>Is SPARC an open source CPU? or is it licenced on a per manufacturer
>basis? Can I go into fab on an UltraSPARC III without talking to SUN
>about it? I know that the CPUs are available from a number of companies
>(Fujitsu among them) but I thought it was simply widely licenced, like
>the ARM series.
You can get a SPARC cpu design essentially for free from ESA.
You can design/build your own and have it certified.
But Sun owns the specific implementation known as US-III so you'd need
to pay them for the rights to use that.
(Fujitsu builds their own SPARC cpus)
Casper
--
Expressed in this posting are my opinions. They are in no way related
to opinions held by my employer, Sun Microsystems.
Statements on Sun products included here are not gospel and may
be fiction rather than truth.
|
|
|
0
|
|
|
|
Reply
|
Casper
|
12/18/2004 11:26:22 PM
|
|
begin Casper H.S. Dik wrote:
> Jim Richardson <warlock@eskimo.com> writes:
>
>>Is SPARC an open source CPU? or is it licenced on a per manufacturer
>>basis? Can I go into fab on an UltraSPARC III without talking to SUN
>>about it? I know that the CPUs are available from a number of companies
>>(Fujitsu among them) but I thought it was simply widely licenced, like
>>the ARM series.
>
> You can get a SPARC cpu design essentially for free from ESA.
>
> You can design/build your own and have it certified.
>
> But Sun owns the specific implementation known as US-III so you'd need
> to pay them for the rights to use that.
>
In other words, there goes the "open SPARC" argument
--
The probability of someone watching you is proportional to the
stupidity of your action.
|
|
|
0
|
|
|
|
Reply
|
Peter
|
12/19/2004 1:31:56 AM
|
|
In article <cq1273$7ak$1@news.cs.tu-berlin.de> (Sat, 18 Dec 2004 10:53:23
+0000), Joerg Schilling wrote:
> The fact that star is able to read and write a subset of the proprietray
> GNU tar format does not make the GNU tar archive format open.....
It certainly doesn't. The fact that GNU tar source code is freely
available under the GPL is sufficient by itself to make the GNU tar
archive format open.
> in most cases it is even esier to reverse engineer the format than to
> look into the source!
This has nothing to do with open!
> Well this is a result of the fact that there is not
> much struture in the GNU tar source.
This also has nothing to do with open.
> The fact that somtheing like the TransMeta CPU exists is not a fact of x86 being
> open but just because it is possible to reverse engineer.
I'd agree there, too.
--
"The European Commission has concluded, after a five-year investigation,
that Microsoft Corporation broke European Union competition law ..."
-- Text of the European Commission ruling. 24 March 2004.
|
|
|
0
|
|
|
|
Reply
|
Hamilcar
|
12/19/2004 3:55:13 AM
|
|
In article <41c41cca$0$90761$e4fe514c@news.xs4all.nl> (Sat, 18 Dec 2004
12:04:26 +0000), Casper H. S. Dik wrote:
> There have been lawsuits that proof that assertion: Intel trying
> to stop the competition. It's also the sole reason of the Itanic's
> existance; make something so weird and so encumbered that it is impossible
> to clone.
The IA-64's saving grace is that it's so weird and so encumbered that it
is impossible to make a profit on it.
--
"The way things are structured today, from a licensing perspective,
in the Linux world nobody will ever commercialize Linux the way that
Sun commercialized FreeBSD."
-- Steve Ballmer. http://news.com.com/2008-1082-998297.html. 25 Apr 2003
|
|
|
0
|
|
|
|
Reply
|
Hamilcar
|
12/19/2004 3:56:24 AM
|
|
In article <41c3ebcc$0$85990$e4fe514c@news.xs4all.nl> (Sat, 18 Dec 2004
08:35:24 +0000), Casper H. S. Dik wrote:
> Hamilcar Barca <hamilcar@tld.always.invalid> writes:
>
>>Perhaps it was. A similar ending was found in the story of Kodak v. Sun.
>>No issues of any kind were decided in either of these situations, except
>>that of corporations doing business.
>
> But the fact remains that the GPL restricts people from using "free"
> software; you cannot dispute that because I have proven that.
The disproof is incontrovertible. The GPL prevents no one from using any
software; it places restrictions only on distribution.
> It does not need a completed lawsuit, the mere threat of one is sometimes
> sufficient.
The threat of or occurrence of a lawsuit do not by themselves validate or
negate licenses.
--
"The European Commission has concluded, after a five-year investigation,
that Microsoft Corporation broke European Union competition law ..."
-- Text of the European Commission ruling. 24 March 2004.
|
|
|
0
|
|
|
|
Reply
|
Hamilcar
|
12/19/2004 3:58:25 AM
|
|
In article <cq191i$icm$1@news.cs.tu-berlin.de> (Sat, 18 Dec 2004 12:49:54
+0000), Joerg Schilling wrote:
> When trying to analyse the postings from GPL zelots,
You should read the GPL before making any additional false claims.
> it seems that RMS did do a really good marketing job with redefining
> the meaning of the word "free" and "proprietary".
It might seem that way, when you haven't read the GPL.
> It seems that most GPL zelots do ot even remember the
> official meaning anymore :-(
It might seem that way, when you haven't read the GPL.
--
"When we speak of Free Software, we are referring to freedom, not price."
-- Richard M. Stallman
|
|
|
0
|
|
|
|
Reply
|
Hamilcar
|
12/19/2004 4:00:55 AM
|
|
-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1
On 18 Dec 2004 13:14:11 GMT,
Casper H.S Dik <Casper.Dik@Sun.COM> wrote:
> js@cs.tu-berlin.de (Joerg Schilling) writes:
>
>>Here in Europe many people have problems to take his OSS activism for serious
>>because he is a gun fetishist. OSS and rifles do not fit well together....
>
> Nah, all Americans are gun fetishist[1]. You've just upset all
> the gun toting OSSers; better not fly their for a bit :-)
>
>
> Casper
>
> [1] Fortunately, the Americans likely to be upset by being tarred with this
> brush are not likely to own guns.
You'd be surprised.
<http://www.eskimo.com/~warlock/firearms/shooting.html>
But then, I don't listen to the gun fetish crowd (Oh! A gun! it's eeevil!)
I just treat them like any other tool. :)
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.2.5 (GNU/Linux)
iD8DBQFBxPo2d90bcYOAWPYRAkT+AJ96HCPW+05bAwNiKb3iuYlc9NijPACgpAM1
J9kiX5I/U8KFBqhRUfKft/U=
=B+6Z
-----END PGP SIGNATURE-----
--
Jim Richardson http://www.eskimo.com/~warlock
Soap and education are not as sudden as a massacre, but they are more
deadly in the long run.
-- Mark Twain
|
|
|
0
|
|
|
|
Reply
|
Jim
|
12/19/2004 4:10:40 AM
|
|
-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1
On 18 Dec 2004 15:42:27 GMT,
Joerg Schilling <js@cs.tu-berlin.de> wrote:
> In article <10s8ijle8pu15d0@corp.supernews.com>,
> Richard L. Hamilton <Richard.L.Hamilton@mindwarp.smart.net> wrote:
>
>>>>>Certainly you don't think that any European takes Eric Raymond seriously
>>>>>as he completely fails to understand how cathedrals were build?
>>>>
>>>>Actually, I don't take him seriously because he's madder than a box of frogs.
>>>
>>> Here in Europe many people have problems to take his OSS activism for serious
>>> because he is a gun fetishist. OSS and rifles do not fit well together....
>>
>>Why? Is there some technical reason that a computer-controlled rifle
>>couldn't run a stripped down version of Linux? :-)
>
> No, the answer is much simpler:
>
> OSS gives you freedom while weapons (in the hand of others) take your freedom.
>
> So what does ESR like?
>
> - fight for your freedom
>
> - threaten your freedom with his guns?
>
> BTW: even weapons in your hand have nothing in common with freedom.
> They just give you force about others. If you don't like to threaten
> the freedom of others you don't need weapons. What kind of threat
> weapons (specially in the USA) press on others may easily be proved by
> comparing fatalities caused by weapons in the USA to the numbers in
> the rest of the world.
>
You have a very poor understanding of sociology. Free clue, in the UK,
violent crime is on the rise, and has been since they started
instituting their gun laws. Prior to 1921, anyone could buy just about
anything. Upto and including machineguns. Crime was low, and murder,
almos unheard of.
Ooops, there goes one more gun fetishist theory down the drain...
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.2.5 (GNU/Linux)
iD8DBQFBxPrzd90bcYOAWPYRAnIPAKDwSOZA3tvoacOqM33nf01RJZas6gCfWcUa
NBOtCkWeJ0fW/pj/qENkrJI=
=xACj
-----END PGP SIGNATURE-----
--
Jim Richardson http://www.eskimo.com/~warlock
Q: What's a WASP's idea of open-mindedness?
A: Dating a Canadian.
|
|
|
0
|
|
|
|
Reply
|
Jim
|
12/19/2004 4:10:43 AM
|
|
-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1
On Sat, 18 Dec 2004 21:21:30 GMT,
Rich Teer <rich.teer@rite-group.com> wrote:
> On Sat, 18 Dec 2004, Jim Richardson wrote:
>
>> Is SPARC an open source CPU? or is it licenced on a per manufacturer
>
> In short, both. For a longer (better) answer, please see:
>
> http://www.sparc.org/faq.html
>
>> basis? Can I go into fab on an UltraSPARC III without talking to SUN
>
> Dunno about that. The UltraSPARC III is Sun's implementation of the
> SPARC v9 standard, so I suspect that they might have something to say
> about that! But, there's nothing preventing you from taking the SPARC
> v9 standard, and desiging your own, compatible CPU.
>
The FAQ says that three things make SPARC open, anyone can licence it,
the instruction set is published as an ieee standard, and specs are in
control of a commitee open to anyone. Now, anyone can licence an x86
(just costs money) and while I don't know if the inst set is published
as ieee, it's certainly open enough for there to be numerous x86 clones
that have nothing to do with Intel.
Now, I am not going to claim that x86 is a better architecture, as I
don't happen to think that. But it sure seems open to me.
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.2.5 (GNU/Linux)
iD8DBQFBxPwrd90bcYOAWPYRAjaHAJ9ej5B5Pt4yrY+qXwXBBPrZGr18GQCfZV24
aiv5wrYqKuRgxgSvrp8Bbpw=
=I92t
-----END PGP SIGNATURE-----
--
Jim Richardson http://www.eskimo.com/~warlock
You may my glories and my state dispose,
But not my griefs; still am I king of those.
-- William Shakespeare, "Richard II"
|
|
|
0
|
|
|
|
Reply
|
Jim
|
12/19/2004 4:10:44 AM
|
|
In article <41c3ea5b$0$86949$e4fe514c@news.xs4all.nl> (Sat, 18 Dec 2004
08:29:15 +0000), Casper H. S. Dik wrote:
> Hamilcar Barca <hamilcar@tld.always.invalid> writes:
>
>>In article <Pine.SOL.4.58.0412171225480.19783@zaphod> (Fri, 17 Dec 2004
>>20:31:44 +0000), Rich Teer wrote:
>
>>> is the fact that most of them run Linux on proprietory x86-based
>>> machines
>
>>It is? There are CPUs from Intel, AMD, Cyrix (National Semiconductor?),
>>VIA, and TransMeta. Are these proprietary.
>
> You seem to have missed the wars about this; there are several
> reasons why others can make x86 CPUs; one being the near monopoly
> Intel has as well as some licensing deals Intel did in the past.
I didn't miss the wars, but I also didn't know if all the others had
"clean-room" implementations.
> Certainly Intel felt they owned x86; they did sue AMD.
Intel lost their trademark suit and the suit to prevent AMD from using
licensed materials. A presence of a lawsuit is never evidence of the
merit of its claims.
>>> yet don't give Sun credit for their SPARC CPU, which is an
>>> open standard.
>
>>Who's the accrediting body?
>
> SPARC international.
It seems that "SPARC international" is part of Sun Microsystems.
> There are several different SPARC implementation (Sun, Fujitsu)
I thought we had established the fact that "several different ...
implementation[s]" was irrelevant to being an "open standard".
--
"Today is really the formalization of our going down the path of broad-
ening our case to go beyond just contracts to include copyrights."
-- Darl McBride. CEO, The SCO Group. InternetNews, 7 July 2003
|
|
|
0
|
|
|
|
Reply
|
Hamilcar
|
12/19/2004 4:12:35 AM
|
|
In article <cq2lj9$h2j$02$6@news.t-online.com>,
Peter =?UTF-8?B?S8O2aGxtYW5u?= <Peter.Koehlmann@t-online.de> writes:
> begin Casper H.S. Dik wrote:
>
>> Jim Richardson <warlock@eskimo.com> writes:
>>
>>>Is SPARC an open source CPU? or is it licenced on a per manufacturer
>>>basis? Can I go into fab on an UltraSPARC III without talking to SUN
>>>about it? I know that the CPUs are available from a number of companies
>>>(Fujitsu among them) but I thought it was simply widely licenced, like
>>>the ARM series.
>>
>> You can get a SPARC cpu design essentially for free from ESA.
>>
>> You can design/build your own and have it certified.
>>
>> But Sun owns the specific implementation known as US-III so you'd need
>> to pay them for the rights to use that.
>>
>
> In other words, there goes the "open SPARC" argument
How do you get that? The "open" is that anyone can implement the
specification; and only incidentally that the design of one implementation
is available for free.
If you think "open" has to mean that _all_ info is free, there's really no
point in even talking to you. Original parts of this posting are
copyright(C) me, and I don't promise not to revoke the license to use it
someday. How do you like them apples? Have I really just subverted Usenet
by stating something that's implicitly true anyway?
Copyrights and patents (although I agree that software algorithm patents
are problematic) are perfectly legitimate. The problem is with the
extremely aggressive behavior of some ("DOS ain't done 'til Lotus won't
run","embrace and extend", ...), along with the constant extensions of
the copyright term in the US (said to be such that Mickey Mouse never goes
out of copyright). At least in the US, the whole point is Article I,
section 8:
The Congress shall have power
[other powers omitted]
To promote the progress of science and useful arts, by securing for
limited times to authors and inventors the exclusive right to their
respective writings and discoveries;
Unfortunately, the courts chose to say that "limited" means whatever
Congress says it does, even if the limits go up constantly so that the
effective result is just about unlimited, and more than that, that
Congress is trivially influenced by lobbyists, at least on those issues
that don't have an immediately obvious impact on people's lives. At
least in the US, a few million people writing their respective Congressmen
and Senators _could_ turn that around, so it's ultimately a question of
whether or not enough people care enough to spend half an hour and less
than a dollar.
--
mailto:rlhamil@smart.net http://www.smart.net/~rlhamil
Lasik/PRK theme music:
"In the Hall of the Mountain King", from "Peer Gynt"
|
|
|
0
|
|
|
|
Reply
|
Richard
|
12/19/2004 6:07:38 AM
|
|
In article <cq1j53$rpl$1@news.cs.tu-berlin.de>,
js@cs.tu-berlin.de (Joerg Schilling) writes:
> In article <10s8ijle8pu15d0@corp.supernews.com>,
> Richard L. Hamilton <Richard.L.Hamilton@mindwarp.smart.net> wrote:
>
>>>>>Certainly you don't think that any European takes Eric Raymond seriously
>>>>>as he completely fails to understand how cathedrals were build?
>>>>
>>>>Actually, I don't take him seriously because he's madder than a box of frogs.
>>>
>>> Here in Europe many people have problems to take his OSS activism for serious
>>> because he is a gun fetishist. OSS and rifles do not fit well together....
>>
>>Why? Is there some technical reason that a computer-controlled rifle
>>couldn't run a stripped down version of Linux? :-)
>
> No, the answer is much simpler:
>
> OSS gives you freedom while weapons (in the hand of others) take your freedom.
>
> So what does ESR like?
>
> - fight for your freedom
>
> - threaten your freedom with his guns?
>
> BTW: even weapons in your hand have nothing in common with freedom. They just
> give you force about others. If you don't like to threaten the freedom of others
> you don't need weapons. What kind of threat weapons (specially in the USA) press
> on others may easily be proved by comparing fatalities caused by weapons in the
> USA to the numbers in the rest of the world.
>
Yeah, in Japan they've banned swords and considered regulating baseball
bats. In England (and maybe Australia), burglaries and assaults went up
after severe firearms restrictions went into place.
Violence in the US doesn't depend on guns; it's just a different
culture; freedom is more important than safety, that's all. (Ben Franklin
had some words on that, I think)
Let's suppose I have a Really Scary rifle or pistol. If I never shoot
anything except paper targets at a safely designed range, or perhaps
occasionally non-endangered species under reasonable regulation, how am I
threatening anyone else's freedom in a way that I couldn't do with any
number of improvised materials that simply can't all be banned? Crossbows
can be made out of leaf springs and wood or scrap metal; Molotov cocktails
are what, a bottle, some gas, and a rag? If I wanted to harm someone, I
could do so a thousand ways without a gun, even if they took a bit more
planning or preparation. What a gun would do (in a state that didn't
regulate them as heavily as the one I live in) would make me the full
equal of anyone with any other weapon (including advanced martial arts
skills). I just heard someone say that they realized that in Los Angeles,
people were all suspicious, but in Texas, they were all polite. Why? In
LA, you don't know who is out to get you and who isn't. In Texas, you
just assume _everyone_ is packing, so you're polite, and so are they -
they know they wouldn't get far if they started something.
Read just about any book by L. Neil Smith, you'll get the picture. :-)
It's not wrong, it's just _different_. Whether or not you agree doesn't
really change that.
--
mailto:rlhamil@smart.net http://www.smart.net/~rlhamil
Lasik/PRK theme music:
"In the Hall of the Mountain King", from "Peer Gynt"
|
|
|
0
|
|
|
|
Reply
|
Richard
|
12/19/2004 6:32:31 AM
|
|
Richard L. Hamilton, <Richard.L.Hamilton@mindwarp.smart.net>, the dwarfish,
insensible garden shrub, and old woman who goes about the streets gathering
dog droppings to be used for tanning leather, extravagated:
> Violence in the US doesn't depend on guns; it's just a different
> culture; freedom is more important than safety, that's all.
Oh, my favourite topic. A fuckwit gun-nut raving on about freedom and safety
in the same single phrase. Ok, retard. It's logic-jousting time. Here's a
pencil sharpener... sharpen your wit.
Given the context of the statement you made, safety is taken to mean
personal safety.
If safety is a lesser ideal than freedom, do you really have freedom when
you're not safe to walk down the street?
--
Registered Linux User: #344402
Akhenaten: Registered Linux Machine: #235500
2.6.8.1-12-i686-smp Dual i686 Intel(R) Xeon(TM) CPU 2.80GHz GNU/Linux
|
|
|
0
|
|
|
|
Reply
|
Kordia
|
12/19/2004 6:59:10 AM
|
|
Mike Cox wrote:
> That is why they post so much FUD on sites such as slashdot.org. The
> reason Linux became popular was that they were not as conservative as BSD,
> and
> would add features constantly. That meant if you wanted some enterprise
> functionality for free, you'd have to go with linux (SMP, etc.).
>
> The downside of that linux philosophy was that backwards compatiblitiy
> went
> down the drain. glibc interfaces and kernel calls would change
> significantly between versions. Sometimes the "enterprise" feature would
> be botched, like the scheduler in the 2.4 kernel.
>
> Now that Solaris 10 is going to be open sourced, Linux advocates rightly
> fear that Linux will be abandoned in favor of Solaris 10. Similar to how
> BSD was left in droves for Linux. That's right, SUN's brilliant strategy
> of making Solaris open sourced is using the linux advocates' techniques
> against them.
Actually, I'm worried that the OpenSolaris license will prevent me from
fooling around with in on Xen.
>
> The reason people will move to Solaris in droves is the same that caused
> BSD's user flight. It supports more enterprise features. But the
> difference this time is that Solaris is cheaper than Redhat, is completely
> backwards compatible, and scales much better on many more CPUs.
>
> Now if only SUN would release its compiler under a BSD license....
|
|
|
0
|
|
|
|
Reply
|
Tux
|
12/19/2004 7:48:17 AM
|
|
thad01@dexter.glaci.com wrote:
> In comp.os.linux.advocacy Rich Teer <rich.teer@rite-group.com> wrote:
>>> most of the Solaris development burden. If they go Linux, donate
>>> some Sun big iron to a few universities and prominent kernel
>>> developers, they begin to reap immediate benefits.
>>
>> What's to stop them from doing that for Open Solaris?
>
> There is nothing to stop them doing it, and if they are serious
> about open Solaris they will. My argument is that their return on
> investment would be better in trying to tap into the existing Linux
> developer pool rather than trying to build a new community around
> Solaris. They seem to be clinging to the idea of using Solaris to
> differentiate themselves from the other hardware vendors. That
> made sense in the old days of the closed source unix wars, but is
> much harder to pull off in today's fast paced open source world.
>
> Of course I'm doing some major crystal ball gazing here. If Sun
> plays it right and engages the community with the right kind of
> challenges, they can certainly build a thriving developer pool
> around Solaris. Given some of their past missteps, however, I'm
> not so optimistic. Sun has actually done a lot of great stuff
> for the open source community, yet their upper managment continues
> to throw away goodwill by the truckloads with some of their
> boneheaded statements and other ill-thought actions. I'm not
> convinced they have the 'community building' knack. Time will
They managed it with OpenOffice.org, by getting out of the way of the
community.
> tell, and I wish them well. Sun has done some great things over
> the years and continues to churn out some really great tech.
>
> Later,
|
|
|
0
|
|
|
|
Reply
|
Tux
|
12/19/2004 8:06:08 AM
|
|
In article <846b85f26222495a9f4b558927d54723@you.orwellian-goat-herded-febrifuge.net>,
Kordia Zanoeta Acidocaldarius <kordia.zanoeta.acidocaldarius@you.tin-canned-cloying-boil-on-the-arse-of-humanity.org> writes:
> Richard L. Hamilton, <Richard.L.Hamilton@mindwarp.smart.net>, the dwarfish,
> insensible garden shrub, and old woman who goes about the streets gathering
> dog droppings to be used for tanning leather, extravagated:
>
>
>> Violence in the US doesn't depend on guns; it's just a different
>> culture; freedom is more important than safety, that's all.
>
> Oh, my favourite topic. A fuckwit gun-nut raving on about freedom and safety
> in the same single phrase. Ok, retard. It's logic-jousting time. Here's a
> pencil sharpener... sharpen your wit.
>
> Given the context of the statement you made, safety is taken to mean
> personal safety.
>
> If safety is a lesser ideal than freedom, do you really have freedom when
> you're not safe to walk down the street?
>
By your polite logic, you have perfect freedom: to walk down the street, or
not to.
If you have freedom, you can alway go get safety (neighborhood patrols, if
need be). If you don't have freedom, you won't have safety either anyway.
Life is violent; people die, whether at the hands of another, or from their
own stupidity. People who are old enough and should know better, should
not be protected from themselves or from protecting themselves. As far as
I'm concerned, people shouldn't even be rescued from the consequences of
their own stupidity except by strictly private charity (of which there
should be more than there is).
--
mailto:rlhamil@smart.net http://www.smart.net/~rlhamil
Lasik/PRK theme music:
"In the Hall of the Mountain King", from "Peer Gynt"
|
|
|
0
|
|
|
|
Reply
|
Richard
|
12/19/2004 10:34:58 AM
|
|
Richard L. Hamilton, <Richard.L.Hamilton@mindwarp.smart.net>, the
arm-swinging, cacophonous dope, and hoarder of the domestic refuse of
others, bleated:
> In article
>
<846b85f26222495a9f4b558927d54723@you.orwellian-goat-herded-febrifuge.net>,
> Kordia Zanoeta Acidocaldarius
>
<kordia.zanoeta.acidocaldarius@you.tin-canned-cloying-boil-on-the-arse-of-hu
manity.org>
> writes:
>> Richard L. Hamilton, <Richard.L.Hamilton@mindwarp.smart.net>, the
>> dwarfish, insensible garden shrub, and old woman who goes about the
>> streets gathering dog droppings to be used for tanning leather,
>> extravagated:
>>
>>
>>> Violence in the US doesn't depend on guns; it's just a different
>>> culture; freedom is more important than safety, that's all.
>>
>> Oh, my favourite topic. A fuckwit gun-nut raving on about freedom
>> and safety in the same single phrase. Ok, retard. It's
>> logic-jousting time. Here's a pencil sharpener... sharpen your wit.
>>
>> Given the context of the statement you made, safety is taken to mean
>> personal safety.
>>
>> If safety is a lesser ideal than freedom, do you really have freedom
>> when you're not safe to walk down the street?
>>
>
> By your polite logic, you have perfect freedom: to walk down the
> street, or
> not to.
That's your fucked up logic, you dumb cunt.
> If you have freedom, you can alway go get safety (neighborhood
> patrols, if
> need be). If you don't have freedom, you won't have safety either
> anyway.
BWAHAHAHAAHAHAAHAHAHAA!!!
> Life is violent; people die,
No fucking shit?
--
Registered Linux User: #344402
Akhenaten: Registered Linux Machine: #235500
2.6.8.1-12-i686-smp Dual i686 Intel(R) Xeon(TM) CPU 2.80GHz GNU/Linux
|
|
|
0
|
|
|
|
Reply
|
Chadwyk
|
12/19/2004 10:59:41 AM
|
|
In article <cq2lj9$h2j$02$6@news.t-online.com>,
Peter =?UTF-8?B?S8O2aGxtYW5u?= <Peter.Koehlmann@t-online.de> wrote:
>begin Casper H.S. Dik wrote:
>
>> You can get a SPARC cpu design essentially for free from ESA.
>>
>> You can design/build your own and have it certified.
>>
>> But Sun owns the specific implementation known as US-III so you'd need
>> to pay them for the rights to use that.
>>
>
>In other words, there goes the "open SPARC" argument
With the same logic, Linux is not open because you cannot use Linux
driver code for FreeBSD.
So thank you for pointing out that you believe Linux is not open....
--
EMail:joerg@schily.isdn.cs.tu-berlin.de (home) J�rg Schilling D-13353 Berlin
js@cs.tu-berlin.de (uni) If you don't have iso-8859-1
schilling@fokus.fraunhofer.de (work) chars I am J"org Schilling
URL: http://www.fokus.fraunhofer.de/usr/schilling ftp://ftp.berlios.de/pub/schily
|
|
|
0
|
|
|
|
Reply
|
js
|
12/19/2004 11:47:36 AM
|
|
In article <20041218225446.747$ta@news.newsreader.com>,
Hamilcar Barca <hamilcar@tld.always.invalid> wrote:
>In article <cq1273$7ak$1@news.cs.tu-berlin.de> (Sat, 18 Dec 2004 10:53:23
>+0000), Joerg Schilling wrote:
>
>> The fact that star is able to read and write a subset of the proprietray
>> GNU tar format does not make the GNU tar archive format open.....
>
>It certainly doesn't. The fact that GNU tar source code is freely
>available under the GPL is sufficient by itself to make the GNU tar
>archive format open.
No, it does not even make the GNU tar format open to itself :-(
Note that there have been various maintainers for GNU tar over the
years. As the GNU tar archive format is not sufficiently documented,
new GNU tar maintainers have a hard job.
The fact that the format is not ducumented makes it impossible to decide
what a "compatible" implementation is. You may write something that is
compatible today but that isn't anymore tomorow.
--
EMail:joerg@schily.isdn.cs.tu-berlin.de (home) J�rg Schilling D-13353 Berlin
js@cs.tu-berlin.de (uni) If you don't have iso-8859-1
schilling@fokus.fraunhofer.de (work) chars I am J"org Schilling
URL: http://www.fokus.fraunhofer.de/usr/schilling ftp://ftp.berlios.de/pub/schily
|
|
|
0
|
|
|
|
Reply
|
js
|
12/19/2004 12:05:46 PM
|
|
In article <20041218225758.976$1q@news.newsreader.com>,
Hamilcar Barca <hamilcar@tld.always.invalid> wrote:
>In article <41c3ebcc$0$85990$e4fe514c@news.xs4all.nl> (Sat, 18 Dec 2004
>08:35:24 +0000), Casper H. S. Dik wrote:
>
>> Hamilcar Barca <hamilcar@tld.always.invalid> writes:
>>
>>>Perhaps it was. A similar ending was found in the story of Kodak v. Sun.
>>>No issues of any kind were decided in either of these situations, except
>>>that of corporations doing business.
>>
>> But the fact remains that the GPL restricts people from using "free"
>> software; you cannot dispute that because I have proven that.
>
>The disproof is incontrovertible. The GPL prevents no one from using any
>software; it places restrictions only on distribution.
So one could call the GPL schizophrenic. The GPL states that it's primary
intention is to require and grant redistribution of the source but in the
case where I e.g. like to add a small piece of GPL code to a larger work
under BSD license it forbids redistribution.
This is because the GPL tries to enforce terms of redistribution that would
limit the usability of the whole work which is under BSD. The only way to be
compliant with the GPL would be to change the license of the whole work from
BSD to GPL.
--
EMail:joerg@schily.isdn.cs.tu-berlin.de (home) J�rg Schilling D-13353 Berlin
js@cs.tu-berlin.de (uni) If you don't have iso-8859-1
schilling@fokus.fraunhofer.de (work) chars I am J"org Schilling
URL: http://www.fokus.fraunhofer.de/usr/schilling ftp://ftp.berlios.de/pub/schily
|
|
|
0
|
|
|
|
Reply
|
js
|
12/19/2004 12:17:45 PM
|
|
In article <20041218230029.092$Zo@news.newsreader.com>,
Hamilcar Barca <hamilcar@tld.always.invalid> wrote:
>In article <cq191i$icm$1@news.cs.tu-berlin.de> (Sat, 18 Dec 2004 12:49:54
>+0000), Joerg Schilling wrote:
>
>> When trying to analyse the postings from GPL zelots,
>
>You should read the GPL before making any additional false claims.
>
>> it seems that RMS did do a really good marketing job with redefining
>> the meaning of the word "free" and "proprietary".
>
>It might seem that way, when you haven't read the GPL.
A poor but often used "argument" of people who did not understand the GPL.
If you like to take part on this discussion, you better should try to
find arguments instead of writing uselss phrases.
If you believe that I did not read the GPL, send us an evidence....
--
EMail:joerg@schily.isdn.cs.tu-berlin.de (home) J�rg Schilling D-13353 Berlin
js@cs.tu-berlin.de (uni) If you don't have iso-8859-1
schilling@fokus.fraunhofer.de (work) chars I am J"org Schilling
URL: http://www.fokus.fraunhofer.de/usr/schilling ftp://ftp.berlios.de/pub/schily
|
|
|
0
|
|
|
|
Reply
|
js
|
12/19/2004 12:21:02 PM
|
|
In article <jgaf92-f9i.ln1@grendel.myth>,
Jim Richardson <warlock@eskimo.com> wrote:
>> OSS gives you freedom while weapons (in the hand of others) take your freedom.
>>
>> So what does ESR like?
>>
>> - fight for your freedom
>>
>> - threaten your freedom with his guns?
>>
>> BTW: even weapons in your hand have nothing in common with freedom.
>> They just give you force about others. If you don't like to threaten
>> the freedom of others you don't need weapons. What kind of threat
>> weapons (specially in the USA) press on others may easily be proved by
>> comparing fatalities caused by weapons in the USA to the numbers in
>> the rest of the world.
>>
>
>You have a very poor understanding of sociology. Free clue, in the UK,
>violent crime is on the rise, and has been since they started
Looks like you did not understand that the topic is not violence but freedom.
Weapons only give freedom from the short sighted view of the people who carry
weapons but not from the global view.
--
EMail:joerg@schily.isdn.cs.tu-berlin.de (home) J�rg Schilling D-13353 Berlin
js@cs.tu-berlin.de (uni) If you don't have iso-8859-1
schilling@fokus.fraunhofer.de (work) chars I am J"org Schilling
URL: http://www.fokus.fraunhofer.de/usr/schilling ftp://ftp.berlios.de/pub/schily
|
|
|
0
|
|
|
|
Reply
|
js
|
12/19/2004 12:27:09 PM
|
|
begin Joerg Schilling wrote:
> In article <20041218225758.976$1q@news.newsreader.com>,
> Hamilcar Barca <hamilcar@tld.always.invalid> wrote:
>>In article <41c3ebcc$0$85990$e4fe514c@news.xs4all.nl> (Sat, 18 Dec 2004
>>08:35:24 +0000), Casper H. S. Dik wrote:
>>
>>> Hamilcar Barca <hamilcar@tld.always.invalid> writes:
>>>
>>>>Perhaps it was. A similar ending was found in the story of Kodak v.
>>>>Sun. No issues of any kind were decided in either of these situations,
>>>>except that of corporations doing business.
>>>
>>> But the fact remains that the GPL restricts people from using "free"
>>> software; you cannot dispute that because I have proven that.
>>
>>The disproof is incontrovertible. The GPL prevents no one from using any
>>software; it places restrictions only on distribution.
>
> So one could call the GPL schizophrenic. The GPL states that it's primary
> intention is to require and grant redistribution of the source but in the
> case where I e.g. like to add a small piece of GPL code to a larger work
> under BSD license it forbids redistribution.
>
> This is because the GPL tries to enforce terms of redistribution that
> would limit the usability of the whole work which is under BSD. The only
> way to be compliant with the GPL would be to change the license of the
> whole work from BSD to GPL.
>
>
Oh, thats simple: Since it is "just a small piece of GPL code" you recreate
it yourself. Should not take too long, since you are free to study how it
is done in the GPLed code
If it is not "a small piece of code" though, why do you want to act against
the express wishes of the programmer who wrote it? It is *his* code, and
*he* is the one who decides howq to deal with it. It seems rather strange
that you have so much problems with the GPL. Is it because it makes it
difficult to steal other peoples code?
Yet you support a company (Sun) which makes it possible in the first place
that sewer-rats like SCO try to steal all of linux. A fine bunch of
hypocrates and thiefs are you "solaris-advocates"
--
You're genuinely bogus.
|
|
|
0
|
|
|
|
Reply
|
Peter
|
12/19/2004 12:29:20 PM
|
|
In article <bqaf92-f9i.ln1@grendel.myth>,
Jim Richardson <warlock@eskimo.com> wrote:
>Now, I am not going to claim that x86 is a better architecture, as I
>don't happen to think that. But it sure seems open to me.
It isn't open because it is owned by Intel.
In theory, Intel could use a new instruction introduced by AMD
for a different purpose.
Sun could not do similar things with Sparc as Sparc is not owned
by Sun but by Sparc International.
--
EMail:joerg@schily.isdn.cs.tu-berlin.de (home) J�rg Schilling D-13353 Berlin
js@cs.tu-berlin.de (uni) If you don't have iso-8859-1
schilling@fokus.fraunhofer.de (work) chars I am J"org Schilling
URL: http://www.fokus.fraunhofer.de/usr/schilling ftp://ftp.berlios.de/pub/schily
|
|
|
0
|
|
|
|
Reply
|
js
|
12/19/2004 12:30:06 PM
|
|
In article <4Ymdnb7RbajHylncRVn-tA@comcast.com>,
=?iso-8859-1?Q?Lin=F8nut?= <=?iso-8859-1?Q?lin=F8nut@bone.com?=> wrote:
>Toomas Soome poked his little head through the XP firewall and said:
>
>I'm not saying code is all. Documents make it easy to grasp a design, by
>far. But, having worked through a number of design documents myself, what
>do I find? I find that the developers (myself included) go through much
>argument, discussion, and discovery as we write the code. Then we have to
>go back to the design documents and spend some time "in transition" updating
>the design documents to reflect the reality that was hammered out in the
>lab.
So you can't tell the difference between standard documents (agreed to
by an offical body, hopefully representing a lot of interessents)
and a design document, essentially a crutch, because we can't write
a program reliably without it?
I give up.
>
>--
>Linux, because a CPU is a terrible thing to waste.
Groetjes Albert
--
--
Albert van der Horst,Oranjestr 8,3511 RA UTRECHT,THE NETHERLANDS
One man-hour to invent,
One man-week to implement,
One lawyer-year to patent.
|
|
|
0
|
|
|
|
Reply
|
Albert
|
12/19/2004 3:08:48 PM
|
|
Gary L Burnore poked his little head through the XP firewall and said:
>>Maybe that reputation is now better. So, where can I get Solaris to
>>download and try? Ah, found it. Damn, now I can't rag on Sun until I
>>actually try this out. Hmmmm, I only have one box with 512Mb on it, and
>>that's this one. Still, maybe it's worth trying.
>
> I'm running it on an x86 with 128mb ram and it works just fine.
Thanks for that info. How much hard-drive space will I need to install
Solaris and have enough room for some software and a /home? I've had a
couple of drives crap out the last few months and am not sure what I have
left.
--
Linux, because a CPU is a terrible thing to waste.
|
|
|
0
|
|
|
|
Reply
|
iso
|
12/19/2004 3:20:11 PM
|
|
Joerg Schilling poked his little head through the XP firewall and said:
>>For some meanings of "better". I'm disinclined, for example, to download
>>Solaris and install it on one of my x86 boxes, because of its poor
>>reputation on that hardware.
>
> Of do you are just too lazy to test a recent version of Solaris?
> It is interesting to see that because it Linux zelots did act the same way
> on Linux, Linux currently would have no more than 1000 users.
I'm not inclined to base a discussion on terms such as "zealot".
All I'm saying is that I've heard Solaris called "Slowlaris", and
also some anecdotes about Linux kicking its ass performance-wise.
> It is many years ago, that I really had problems with hardware
> support for X86 from Solaris.
> Solaris even works on boxes with much less memory.
>
>>Damn, what an ordeal just to get to the license!
>
> What is your problem?
Sun's web site is a disorganized mess! At least it downloads fast!
> As it is a matter of fact that > 90% of the Linux users also run Win32,
> most people do not even have problems with mush more restrictive licenses.
Most people use their boxes like they do a microwave oven. If it craps out
they buy another.
> Note that this is for the _current_ _betas_, not for the OpenSolaris
> you will see soon.
Cool, thanks!
--
Linux, because a CPU is a terrible thing to waste.
|
|
|
0
|
|
|
|
Reply
|
iso
|
12/19/2004 3:24:38 PM
|
|
In article <nKmdnXIKl9e2AVjcRVn-3A@comcast.com>,
=?iso-8859-1?Q?Lin=F8nut?= <=?iso-8859-1?Q?lin=F8nut@bone.com?=> wrote:
>> I'm running it on an x86 with 128mb ram and it works just fine.
>
>Thanks for that info. How much hard-drive space will I need to install
>Solaris and have enough room for some software and a /home? I've had a
>couple of drives crap out the last few months and am not sure what I have
>left.
Dateisystem kByte belegt verf�gbar Kapazit�t Eingeh�ngt auf
/dev/dsk/c1t0d0s0 10319148 5240381 4975576 52% /
/devices 0 0 0 0% /devices
ctfs 0 0 0 0% /system/contract
proc 0 0 0 0% /proc
mnttab 0 0 0 0% /etc/mnttab
swap 12631964 296 12631668 1% /etc/svc/volatile
objfs 0 0 0 0% /system/object
/dev/dsk/c1t0d0p0:boot
10698 1500 9198 15% /boot
Depends on what you install....
If you install everything from the S 10 express CDs + /opt/sfw
from the Sun FreeWare CD + Sun's version of GCC that supports Opteron
+ Sun Studio 10, you need 5.1 GB.
Subtract
~ 700 MB for Sun Studio 10
~ 170 MB for GCC
~ 1.6 GB for /opt/sfw/
--
EMail:joerg@schily.isdn.cs.tu-berlin.de (home) J�rg Schilling D-13353 Berlin
js@cs.tu-berlin.de (uni) If you don't have iso-8859-1
schilling@fokus.fraunhofer.de (work) chars I am J"org Schilling
URL: http://www.fokus.fraunhofer.de/usr/schilling ftp://ftp.berlios.de/pub/schily
|
|
|
0
|
|
|
|
Reply
|
js
|
12/19/2004 3:29:51 PM
|
|
In article <nKmdnW0Kl9erAFjcRVn-3A@comcast.com>,
=?iso-8859-1?Q?Lin=F8nut?= <=?iso-8859-1?Q?lin=F8nut@bone.com?=> wrote:
>Joerg Schilling poked his little head through the XP firewall and said:
>
>>>For some meanings of "better". I'm disinclined, for example, to download
>>>Solaris and install it on one of my x86 boxes, because of its poor
>>>reputation on that hardware.
>>
>> Of do you are just too lazy to test a recent version of Solaris?
>> It is interesting to see that because it Linux zelots did act the same way
>> on Linux, Linux currently would have no more than 1000 users.
>
>I'm not inclined to base a discussion on terms such as "zealot".
>All I'm saying is that I've heard Solaris called "Slowlaris", and
>also some anecdotes about Linux kicking its ass performance-wise.
Well the term "Slowlaris" is a creation of the Linux zelots :-(
What you quote is nothing but typical advertising lies from the Linux camp.
For me and my usage scenario, Solaris did always give me better performance
than Linux.
--
EMail:joerg@schily.isdn.cs.tu-berlin.de (home) J�rg Schilling D-13353 Berlin
js@cs.tu-berlin.de (uni) If you don't have iso-8859-1
schilling@fokus.fraunhofer.de (work) chars I am J"org Schilling
URL: http://www.fokus.fraunhofer.de/usr/schilling ftp://ftp.berlios.de/pub/schily
|
|
|
0
|
|
|
|
Reply
|
js
|
12/19/2004 3:33:18 PM
|
|
Joerg Schilling poked his little head through the XP firewall and said:
>>> Then tell me why gcc in most cases creates less performant x86 code than
>>> Sun's C-compiler.
>>
>>glibc?
>
> No, if I check performance with my Reed Solomon coder, there is no libc code
> involved. Sun's cc gives 30% more performance (even on x86) then gcc compiled
> code does.
I thought Intel's compiler yielded the best code, and that it was 25% faster
code than yielded by gcc. You sure you're not giving numbers based on SPARC
benchmarks?
But, this post (search for "Jim Balson") seems to partly back you up:
http://www.osnews.com/comment.php?news_id=5830&limit=no
--
Linux, because a CPU is a terrible thing to waste.
|
|
|
0
|
|
|
|
Reply
|
iso
|
12/19/2004 3:33:39 PM
|
|
In article <L46dnYp9A4DOAljcRVn-jA@comcast.com>,
=?iso-8859-1?Q?Lin=F8nut?= <=?iso-8859-1?Q?lin=F8nut@bone.com?=> wrote:
>Joerg Schilling poked his little head through the XP firewall and said:
>> No, if I check performance with my Reed Solomon coder, there is no libc code
>> involved. Sun's cc gives 30% more performance (even on x86) then gcc compiled
>> code does.
>
>I thought Intel's compiler yielded the best code, and that it was 25% faster
>code than yielded by gcc. You sure you're not giving numbers based on SPARC
>benchmarks?
As I use a PC at home, important benchmarks are done on x86.
I did not yet do full tests with all my software.
There are a few applictions where GCC did create better code than
Sun Studio 8 but with Studio 9 this seems to be at least compare equal.
My Reed Solomon coder runs 35% faster when compiled with Suns compiler.
dcraw (to convert Nikon RAW images to PPM) was 40% faster when
compiled with Sun's cc. With the latest GCC, this did reduce to 5% but
Sun's compiler still creates better code.
--
EMail:joerg@schily.isdn.cs.tu-berlin.de (home) J�rg Schilling D-13353 Berlin
js@cs.tu-berlin.de (uni) If you don't have iso-8859-1
schilling@fokus.fraunhofer.de (work) chars I am J"org Schilling
URL: http://www.fokus.fraunhofer.de/usr/schilling ftp://ftp.berlios.de/pub/schily
|
|
|
0
|
|
|
|
Reply
|
js
|
12/19/2004 3:47:21 PM
|
|
Lin�nut wrote:
>Gary L Burnore poked his little head through the XP firewall and said:
>
>
>
>>>Maybe that reputation is now better. So, where can I get Solaris to
>>>download and try? Ah, found it. Damn, now I can't rag on Sun until I
>>>actually try this out. Hmmmm, I only have one box with 512Mb on it, and
>>>that's this one. Still, maybe it's worth trying.
>>>
>>>
>>I'm running it on an x86 with 128mb ram and it works just fine.
>>
>>
>
>Thanks for that info. How much hard-drive space will I need to install
>Solaris and have enough room for some software and a /home? I've had a
>couple of drives crap out the last few months and am not sure what I have
>left.
>
>
>
10Gb is enough. 20Gb is better. Solaris can be downloaded from Sun.com
or you can buy a media kit (CDROMS and Manuals) for around $100 US.
Manuals can also be downloaded as PDF and you supply the paper and
binders. All in alll, I think the media kit is a better deal. You
can also buy media kits on e-Bay for the Sparc platform, I haven't
looked for X86.
|
|
|
0
|
|
|
|
Reply
|
Richard
|
12/19/2004 4:26:21 PM
|
|
Peter =?UTF-8?B?S8O2aGxtYW5u?= <Peter.Koehlmann@t-online.de> writes:
>begin Casper H.S. Dik wrote:
>> Jim Richardson <warlock@eskimo.com> writes:
>>
>>>Is SPARC an open source CPU? or is it licenced on a per manufacturer
>>>basis? Can I go into fab on an UltraSPARC III without talking to SUN
>>>about it? I know that the CPUs are available from a number of companies
>>>(Fujitsu among them) but I thought it was simply widely licenced, like
>>>the ARM series.
>>
>> You can get a SPARC cpu design essentially for free from ESA.
>>
>> You can design/build your own and have it certified.
>>
>> But Sun owns the specific implementation known as US-III so you'd need
>> to pay them for the rights to use that.
>>
>In other words, there goes the "open SPARC" argument
Reading skills are a prerequisite for this forum.
You don't seem to grasp the difference between a proprietary *implementation*
and a *open standard*.
Even though you can buy many different x86 CPUs, you cannot get Intel
or AMD to give you the CPU designs and help in manufacturing them.
Casper
--
Expressed in this posting are my opinions. They are in no way related
to opinions held by my employer, Sun Microsystems.
Statements on Sun products included here are not gospel and may
be fiction rather than truth.
|
|
|
0
|
|
|
|
Reply
|
Casper
|
12/19/2004 4:34:07 PM
|
|
Hamilcar Barca <hamilcar@tld.always.invalid> writes:
>It seems that "SPARC international" is part of Sun Microsystems.
It isn't.
>> There are several different SPARC implementation (Sun, Fujitsu)
>I thought we had established the fact that "several different ...
>implementation[s]" was irrelevant to being an "open standard".
Sure; but an open standard with but one implementation isn't
very interesting. (OpenPower)
Casper
--
Expressed in this posting are my opinions. They are in no way related
to opinions held by my employer, Sun Microsystems.
Statements on Sun products included here are not gospel and may
be fiction rather than truth.
|
|
|
0
|
|
|
|
Reply
|
Casper
|
12/19/2004 4:36:44 PM
|
|
js@cs.tu-berlin.de (Joerg Schilling) writes:
>In article <nKmdnW0Kl9erAFjcRVn-3A@comcast.com>,
>=?iso-8859-1?Q?Lin=F8nut?= <=?iso-8859-1?Q?lin=F8nut@bone.com?=> wrote:
>>
>>I'm not inclined to base a discussion on terms such as "zealot".
>>All I'm saying is that I've heard Solaris called "Slowlaris", and
>>also some anecdotes about Linux kicking its ass performance-wise.
>Well the term "Slowlaris" is a creation of the Linux zelots :-(
I think the term may pre-date Linux, certainly Linux as a usable
OS and it was to contrast it to SunOS 4.x and not something else
(SunOS 4.x certainly was snappier in some aspects; suntools, e.g.,
could be rather quick).
Such things stick.
Casper
--
Expressed in this posting are my opinions. They are in no way related
to opinions held by my employer, Sun Microsystems.
Statements on Sun products included here are not gospel and may
be fiction rather than truth.
|
|
|
0
|
|
|
|
Reply
|
Casper
|
12/19/2004 4:41:53 PM
|
|
=?iso-8859-1?Q?Lin=F8nut?= <=?iso-8859-1?Q?lin=F8nut?=@bone.com> writes:
>I thought Intel's compiler yielded the best code, and that it was 25% faster
>code than yielded by gcc. You sure you're not giving numbers based on SPARC
>benchmarks?
No he isn't. With Sun's renewed emphasis on Solaris on x86 such as
finally adding SSE/SSe2 support in the OS, we also needed to
make the compilers competitive again.
Gcc was an easy target to surpass; I think Sun's compilers are now
within spitting distance of the Intel compilers.
Casper
--
Expressed in this posting are my opinions. They are in no way related
to opinions held by my employer, Sun Microsystems.
Statements on Sun products included here are not gospel and may
be fiction rather than truth.
|
|
|
0
|
|
|
|
Reply
|
Casper
|
12/19/2004 4:44:07 PM
|
|
Joerg Schilling wrote:
> In article <nKmdnW0Kl9erAFjcRVn-3A@comcast.com>,
> =?iso-8859-1?Q?Lin=F8nut?= <=?iso-8859-1?Q?lin=F8nut@bone.com?=> wrote:
>
>>Joerg Schilling poked his little head through the XP firewall and said:
>>
>>
>>>>For some meanings of "better". I'm disinclined, for example, to download
>>>>Solaris and install it on one of my x86 boxes, because of its poor
>>>>reputation on that hardware.
>>>
Tried solaris for x86 when it first came out too. Was not impressed.
>>>Of do you are just too lazy to test a recent version of Solaris?
For me, not lazy, just no incentive. The licencing terms didn't help
any. "Another CPU in the box" eh? That might a good incentive to try it
agaon.
>>>It is interesting to see that because it Linux zelots did act the same way
>>>on Linux, Linux currently would have no more than 1000 users.
>>
Did you mean GPL zealots, by any chance? ;-)
>>I'm not inclined to base a discussion on terms such as "zealot".
>>All I'm saying is that I've heard Solaris called "Slowlaris", and
>>also some anecdotes about Linux kicking its ass performance-wise.
>
IIRC, the original benchmarks in the late 90's did put Linux ahead of
Solaris on sparc H/W.
>
> Well the term "Slowlaris" is a creation of the Linux zelots :-(
>
Eh, Joerg, we know of your differences with the Linux kernel folk but
that's no reason to distort history.
The term was coined way back when Solaris first came out, because it was
slower than SunOS, not because it was slower than Linux.
> What you quote is nothing but typical advertising lies from the Linux camp.
>
In the early days, loading RH (4.x? that was tooo long ago) onto my
sparcstation *was* faster, but I mainly used it as a nice 19" xterminal
for my faster x86 linux box (with 14" screen).
Now, upgrading and spares was another matter, but we are discussing OS
software anyway, so I won't bring that in.
> For me and my usage scenario, Solaris did always give me better performance
> than Linux.
>
That's nice. I wish you all the best.
Of course, the original x-posting to CUS was /intended/ to troll/fan
flames, but that's not your fault.
Cheers,
WS
--
linux user #61399
change to leews to mail
|
|
|
0
|
|
|
|
Reply
|
Lee
|
12/19/2004 5:21:11 PM
|
|
Joerg Schilling wrote:
> Depends on what you install....
>
> If you install everything from the S 10 express CDs + /opt/sfw
> from the Sun FreeWare CD + Sun's version of GCC that supports Opteron
> + Sun Studio 10, you need 5.1 GB.
>
> Subtract
>
> ~ 700 MB for Sun Studio 10
Joerg,
Where can one get the Sun Studio 10 betas? I can only find Sun Studio 9.
Thanks,
--
Coy Hile
hile@cse.psu.edu
|
|
|
0
|
|
|
|
Reply
|
Coy
|
12/19/2004 5:50:16 PM
|
|
In comp.unix.solaris Coy Hile <hile@cse.psu.edu> wrote:
> Joerg Schilling wrote:
>> Depends on what you install....
>>
>> If you install everything from the S 10 express CDs + /opt/sfw
>> from the Sun FreeWare CD + Sun's version of GCC that supports Opteron
>> + Sun Studio 10, you need 5.1 GB.
>>
>> Subtract
>>
>> ~ 700 MB for Sun Studio 10
>
> Joerg,
>
> Where can one get the Sun Studio 10 betas? I can only find Sun Studio 9.
>
registration for preview release is in solaris express site, perhaps someday
will be in studio page as well...
toomas
--
Critic, n.:
A person who boasts himself hard to please because nobody tries
to please him.
-- Ambrose Bierce, "The Devil's Dictionary"
|
|
|
0
|
|
|
|
Reply
|
Toomas
|
12/19/2004 6:51:23 PM
|
|
-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1
On 19 Dec 2004 12:27:09 GMT,
Joerg Schilling <js@cs.tu-berlin.de> wrote:
> In article <jgaf92-f9i.ln1@grendel.myth>,
> Jim Richardson <warlock@eskimo.com> wrote:
>
>>> OSS gives you freedom while weapons (in the hand of others) take your freedom.
>>>
>>> So what does ESR like?
>>>
>>> - fight for your freedom
>>>
>>> - threaten your freedom with his guns?
>>>
>>> BTW: even weapons in your hand have nothing in common with freedom.
>>> They just give you force about others. If you don't like to threaten
>>> the freedom of others you don't need weapons. What kind of threat
>>> weapons (specially in the USA) press on others may easily be proved by
>>> comparing fatalities caused by weapons in the USA to the numbers in
>>> the rest of the world.
>>>
>>
>>You have a very poor understanding of sociology. Free clue, in the UK,
>>violent crime is on the rise, and has been since they started
>
> Looks like you did not understand that the topic is not violence but freedom.
>
> Weapons only give freedom from the short sighted view of the people who carry
> weapons but not from the global view.
>
You talk about freedom, then shift to fatalities, you brought up
violence, now you wish to ignore it. Figures.
As for weapons, as tools of freedom. You are free to voice your opinion,
oddly enough, that freedom, came after men with weapons, "discussed" it.
Weapons, firearms included, are just tools. Like any other tool, they
can be abused. They are neither good, nor bad, they are simply tools. To
imbue them with mythical powers as you have, is fetishism.
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.2.5 (GNU/Linux)
iD8DBQFBxefGd90bcYOAWPYRAnqvAKCZVdds3BwhHsNtFROrdVAL6s8FmACfTGT2
iqOWiFNuU4sU+yVM6mXv4CI=
=Vxur
-----END PGP SIGNATURE-----
--
Jim Richardson http://www.eskimo.com/~warlock
Don't relax! It's only your tension that's holding you together.
|
|
|
0
|
|
|
|
Reply
|
Jim
|
12/19/2004 9:10:43 PM
|
|
-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1
On 19 Dec 2004 15:47:21 GMT,
Joerg Schilling <js@cs.tu-berlin.de> wrote:
> In article <L46dnYp9A4DOAljcRVn-jA@comcast.com>,
>=?iso-8859-1?Q?Lin=F8nut?= <=?iso-8859-1?Q?lin=F8nut@bone.com?=> wrote:
>>Joerg Schilling poked his little head through the XP firewall and said:
>
>>> No, if I check performance with my Reed Solomon coder, there is no libc code
>>> involved. Sun's cc gives 30% more performance (even on x86) then gcc compiled
>>> code does.
>>
>>I thought Intel's compiler yielded the best code, and that it was 25% faster
>>code than yielded by gcc. You sure you're not giving numbers based on SPARC
>>benchmarks?
>
> As I use a PC at home, important benchmarks are done on x86.
>
> I did not yet do full tests with all my software.
>
> There are a few applictions where GCC did create better code than
> Sun Studio 8 but with Studio 9 this seems to be at least compare equal.
>
> My Reed Solomon coder runs 35% faster when compiled with Suns compiler.
>
> dcraw (to convert Nikon RAW images to PPM) was 40% faster when
> compiled with Sun's cc. With the latest GCC, this did reduce to 5% but
> Sun's compiler still creates better code.
>
The GCC crowd has long put portability of the toolchain higher than
performance. Which is why it's available over so many platforms and
architectures. It's nice to see that they are also addressing some of
the performance issues as you mention.
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.2.5 (GNU/Linux)
iD8DBQFBxejwd90bcYOAWPYRAri5AJ9bI6cw6Wt9jKDelj+7iGfJrdjo/wCfV23u
qmpX9vvN/inPsa65a7ZCHfY=
=cy6s
-----END PGP SIGNATURE-----
--
Jim Richardson http://www.eskimo.com/~warlock
You're not my type. For that matter, you're not even my species!!!
|
|
|
0
|
|
|
|
Reply
|
Jim
|
12/19/2004 9:10:44 PM
|
|
In article <41c5af51$0$80189$e4fe514c@news.xs4all.nl>,
Casper H.S. Dik <Casper.Dik@Sun.COM> wrote:
>js@cs.tu-berlin.de (Joerg Schilling) writes:
>>Well the term "Slowlaris" is a creation of the Linux zelots :-(
>
>I think the term may pre-date Linux, certainly Linux as a usable
>OS and it was to contrast it to SunOS 4.x and not something else
>(SunOS 4.x certainly was snappier in some aspects; suntools, e.g.,
>could be rather quick).
>
>Such things stick.
You are right, I now remember again.....
Solaris 2.1...2.3 really have been slower then SunOS-4.x and another
reason was that Solaris was very slow in becoming ready for the market.
--
EMail:joerg@schily.isdn.cs.tu-berlin.de (home) J�rg Schilling D-13353 Berlin
js@cs.tu-berlin.de (uni) If you don't have iso-8859-1
schilling@fokus.fraunhofer.de (work) chars I am J"org Schilling
URL: http://www.fokus.fraunhofer.de/usr/schilling ftp://ftp.berlios.de/pub/schily
|
|
|
0
|
|
|
|
Reply
|
js
|
12/19/2004 10:55:13 PM
|
|
In article <cq4ath$19j$1@reader01.singnet.com.sg>,
Lee Wei Shun <see.sig@pacific.net.sg> wrote:
>> Well the term "Slowlaris" is a creation of the Linux zelots :-(
>>
>
>Eh, Joerg, we know of your differences with the Linux kernel folk but
>that's no reason to distort history.
>
>The term was coined way back when Solaris first came out, because it was
>slower than SunOS, not because it was slower than Linux.
I remember again, but this is 12 years ago. People who use this term now
do it without reason.
--
EMail:joerg@schily.isdn.cs.tu-berlin.de (home) J�rg Schilling D-13353 Berlin
js@cs.tu-berlin.de (uni) If you don't have iso-8859-1
schilling@fokus.fraunhofer.de (work) chars I am J"org Schilling
URL: http://www.fokus.fraunhofer.de/usr/schilling ftp://ftp.berlios.de/pub/schily
|
|
|
0
|
|
|
|
Reply
|
js
|
12/19/2004 10:57:44 PM
|
|
In article <cq4f0o$b6i$1@neuromancer.cse.psu.edu>,
Coy Hile <hile@cse.psu.edu> wrote:
>Joerg Schilling wrote:
>> Depends on what you install....
>>
>> If you install everything from the S 10 express CDs + /opt/sfw
>> from the Sun FreeWare CD + Sun's version of GCC that supports Opteron
>> + Sun Studio 10, you need 5.1 GB.
>>
>> Subtract
>>
>> ~ 700 MB for Sun Studio 10
>
>Joerg,
>
>Where can one get the Sun Studio 10 betas? I can only find Sun Studio 9.
Ask the Sun dealer of your trust.
The beta program has been closed AFAIK.
--
EMail:joerg@schily.isdn.cs.tu-berlin.de (home) J�rg Schilling D-13353 Berlin
js@cs.tu-berlin.de (uni) If you don't have iso-8859-1
schilling@fokus.fraunhofer.de (work) chars I am J"org Schilling
URL: http://www.fokus.fraunhofer.de/usr/schilling ftp://ftp.berlios.de/pub/schily
|
|
|
0
|
|
|
|
Reply
|
js
|
12/19/2004 10:59:55 PM
|
|
In article <cq3rh9$4eg$1@news.cs.tu-berlin.de> (Sun, 19 Dec 2004 12:17:45
+0000), Joerg Schilling wrote:
> So one could call the GPL schizophrenic.
Certainly, if one wanted to be known as a dissembler.
> The GPL states that it's primary intention is to require and grant
> redistribution of the source but in the case where I e.g. like to
> add a small piece of GPL code to a larger work under BSD license
> it forbids redistribution.
Yes, it's quite clear that the GPL forbids relicensing or the combining of
GPL code with non-GPL code.
> This is because the GPL tries to enforce terms of redistribution that would
> limit the usability of the whole work which is under BSD.
The GPL doesn't enforce anything. It explicitly forbids limiting the
usability of code licensed under the GPL by combining it with other
licenses.
> The only way to be compliant with the GPL would be to change the license
> of the whole work from BSD to GPL.
If the rights are yours, you may use any license that pleases you. If the
rights aren't yours, it's immaterial.
--
Q: Speaking of security, Internet Explorer has had well-publicized holes.
Gates: Understand those are cases where you are downloading third-party
software.
-- Bill Gates. Chairman, Microsoft. USA Today interview. 13 Oct 2004.
|
|
|
0
|
|
|
|
Reply
|
Hamilcar
|
12/20/2004 1:19:19 AM
|
|
Here in comp.unix.solaris,
js@cs.tu-berlin.de (Joerg Schilling) spake unto us, saying:
>With the same logic, Linux is not open because you cannot use Linux
>driver code for FreeBSD.
But you can. All you have to do is re-release the BSD code under the
GPL, something which is probably permitted under the BSDL.
--
-Rich Steiner >>>---> http://www.visi.com/~rsteiner >>>---> Smyrna, GA USA
OS/2 + eCS + Linux + Win95 + DOS + PC/GEOS + Executor = PC Hobbyist Heaven!
WARNING: I've seen FIELDATA FORTRAN V and I know how to use it!
The Theorem Theorem: If If, Then Then.
|
|
|
0
|
|
|
|
Reply
|
rsteiner
|
12/20/2004 3:31:18 AM
|
|
In article <GekxBpHpvmWR092yn@visi.com>,
Richard Steiner <rsteiner@visi.com> wrote:
>Here in comp.unix.solaris,
>js@cs.tu-berlin.de (Joerg Schilling) spake unto us, saying:
>
>>With the same logic, Linux is not open because you cannot use Linux
>>driver code for FreeBSD.
>
>But you can. All you have to do is re-release the BSD code under the
>GPL, something which is probably permitted under the BSDL.
Do you _really_ like to make the proposal that the whole FreeBSD kernel
should be relicensed under GPL?
--
EMail:joerg@schily.isdn.cs.tu-berlin.de (home) J�rg Schilling D-13353 Berlin
js@cs.tu-berlin.de (uni) If you don't have iso-8859-1
schilling@fokus.fraunhofer.de (work) chars I am J"org Schilling
URL: http://www.fokus.fraunhofer.de/usr/schilling ftp://ftp.berlios.de/pub/schily
|
|
|
0
|
|
|
|
Reply
|
js
|
12/20/2004 10:15:46 AM
|
|
Joerg Schilling poked his little head through the XP firewall and said:
>>I think the term may pre-date Linux, certainly Linux as a usable
>>OS and it was to contrast it to SunOS 4.x and not something else
>>(SunOS 4.x certainly was snappier in some aspects; suntools, e.g.,
>>could be rather quick).
>>
>>Such things stick.
>
> Solaris 2.1...2.3 really have been slower then SunOS-4.x and another
> reason was that Solaris was very slow in becoming ready for the market.
I remember SunOS with fondness (more than a decade ago). And also remember
being able to capture someone else's screen, and multimedia, just when
Windows was about ready to become as useable (version 3.0) as DOS.
--
Linux, because a CPU is a terrible thing to waste.
|
|
|
0
|
|
|
|
Reply
|
iso
|
12/20/2004 12:25:33 PM
|
|
Albert van der Horst poked his little head through the XP firewall and said:
> In article <4Ymdnb7RbajHylncRVn-tA@comcast.com>,
>=?iso-8859-1?Q?Lin=F8nut?= <=?iso-8859-1?Q?lin=F8nut@bone.com?=> wrote:
>>Toomas Soome poked his little head through the XP firewall and said:
>>
>>I'm not saying code is all. Documents make it easy to grasp a design, by
>>far. But, having worked through a number of design documents myself, what
>>do I find? I find that the developers (myself included) go through much
>>argument, discussion, and discovery as we write the code. Then we have to
>>go back to the design documents and spend some time "in transition" updating
>>the design documents to reflect the reality that was hammered out in the
>>lab.
>
> So you can't tell the difference between standard documents (agreed to
> by an offical body, hopefully representing a lot of interessents)
> and a design document, essentially a crutch, because we can't write
> a program reliably without it?
Sure I can. Standards documents are even more useless to the developer than
a design document.
> I give up.
You give up too easily. But I can understand, we all have a lot do and must
select our causes carefully <grin>.
--
Linux, because a CPU is a terrible thing to waste.
|
|
|
0
|
|
|
|
Reply
|
iso
|
12/20/2004 12:29:13 PM
|
|
In article <cq47q9$gqh$1@news.cs.tu-berlin.de>,
js@cs.tu-berlin.de (Joerg Schilling) writes:
[...]
> dcraw (to convert Nikon RAW images to PPM) was 40% faster when
> compiled with Sun's cc. With the latest GCC, this did reduce to 5% but
> Sun's compiler still creates better code.
Do you have a dcraw binary compiled with Sun's latest, suitable for
Solaris 9 (SPARC)? Compiled with gcc, it's rather slow on my Sun Blade
100 (although it's existence was one of the reasons I recently got a Nikon
D70, and have been fairly pleased with dcraw+gimp so far).
--
mailto:rlhamil@smart.net http://www.smart.net/~rlhamil
Lasik/PRK theme music:
"In the Hall of the Mountain King", from "Peer Gynt"
|
|
|
0
|
|
|
|
Reply
|
Richard
|
12/20/2004 12:40:38 PM
|
|
In article <7c5d4d7bdf1848bfbd4801814fc416ea@you.transsexual-humourless-flunkey.org>,
Chadwyk Ebbledock-Sesslehotch <chadwyk.ebbledock-sesslehotch@you.swaggering-pupillary-freakface.com> writes:
[...]
Wow. No more than the 2nd or 3rd time I ever used a killfile!
--
mailto:rlhamil@smart.net http://www.smart.net/~rlhamil
Lasik/PRK theme music:
"In the Hall of the Mountain King", from "Peer Gynt"
|
|
|
0
|
|
|
|
Reply
|
Richard
|
12/20/2004 12:52:40 PM
|
|
Richard L. Hamilton, <Richard.L.Hamilton@mindwarp.smart.net>, the
meaningless, self-seeded giglet, and retailer of stockings and nightcaps,
hawked:
> In article
> <7c5d4d7bdf1848bfbd4801814fc416ea@you.transsexual-humourless-flunkey.org>,
> Chadwyk Ebbledock-Sesslehotch
> <chadwyk.ebbledock-sesslehotch@you.swaggering-pupillary-freakface.com>
> writes:
> [...]
>
> Wow. No more than the 2nd or 3rd time I ever used a killfile!
BWAHAHAHAHAAHAHAH!! You fucking idjit.
--
Registered Linux User: #344402
Akhenaten: Registered Linux Machine: #235500
2.6.8.1-12-i686-smp Dual i686 Intel(R) Xeon(TM) CPU 2.80GHz GNU/Linux
|
|
|
0
|
|
|
|
Reply
|
Mendosicutes
|
12/20/2004 1:22:29 PM
|
|
begin Richard L. Hamilton wrote:
> In article
> <7c5d4d7bdf1848bfbd4801814fc416ea@you.transsexual-humourless-flunkey.org>,
> Chadwyk Ebbledock-Sesslehotch
> <chadwyk.ebbledock-sesslehotch@you.swaggering-pupillary-freakface.com>
> writes:
> [...]
>
> Wow. No more than the 2nd or 3rd time I ever used a killfile!
>
Well, you've met Kadaitcha Man now. Although K-Toddler would be more
appropriate, since he is mainly occupied flinging the stuff he hauls out of
his diapers. It seems he also oozes that smellish brown goo out of his head
--
Programmer: A biological system designed to convert coffee and cheesies into
code
|
|
|
0
|
|
|
|
Reply
|
Peter
|
12/20/2004 2:46:52 PM
|
|
Richard L. Hamilton wrote:
> In article
> <7c5d4d7bdf1848bfbd4801814fc416ea@you.transsexual-humourless-flunkey.org>,
> Chadwyk Ebbledock-Sesslehotch
> <chadwyk.ebbledock-sesslehotch@you.swaggering-pupillary-freakface.com>
> writes: [...]
>
> Wow. No more than the 2nd or 3rd time I ever used a killfile!
Finally figured out how to do it, huh. "linux makes you stupid."
|
|
|
0
|
|
|
|
Reply
|
Damian
|
12/20/2004 3:12:20 PM
|
|
Lin�nut <lin�nut@bone.com> wrote:
....
>> (It's also a pain that there's no C++ ABI; i.e. straight C will link
>> between gcc and Sun's (or some other) compiler, but C++ typically won't.)
>
> Why can't we have a name-mangling standard??? <grin>
Just in case this misleads some other misinformed people
to believe gcc and Sun are quibbling with each other
over a petty "name mangling" stuff
(not that name mangling is a petty issue),
there's much more than name-mangling in C++ ABI -
in particular, object model, exception handling and RTTI.
And those are core implementation details and can not easily be
made (or changed) to be compatible.
PS. Followup-to set to NG I read.
--
#pragma ident "Seongbae Park, compiler, http://blogs.sun.com/seongbae/"
|
|
|
0
|
|
|
|
Reply
|
Seongbae
|
12/20/2004 7:17:08 PM
|
|
Freeride wrote:
> Bullshit! Sun's CDDL conflicts with the GPL and no Open Source developer is
> going to touch it with a 20 foot stick. And I am sure that Sun would love
> to have some of its code show up in Linux so it can throw a big giant
> monkey wrench into the development and adoption of Linux.
We say "10 foot pole"...
Anyways, GOOD! I don't want to see Solaris's ADVANCED stuff in a P.O.S.
renegade OS codenamed "Linux". I want Solaris to kick Linux's butt,
because Linux *IS* a P.O.S. Right now Linux has the upper hand because
of wider driver and fanatical support from PC-bucket kids. However, if
Sun doesn't screw this up, the P.O.S. Linux OS will suck it.
And yes, I openly hate Linux, because I have to work with it every day
for a living and because it always causes problems. And yes, I know how
to properly configure & admin a Linux system, thank you. It's just that
it's crap compared to Solaris, not to mention compared to IRIX.
Oh, and BTW: whoever doesn't like corporate UNIX can shove Linux where
the Sun don't shine. I think I pretty much said everything with that
last sentence.
|
|
|
0
|
|
|
|
Reply
|
UNIX
|
12/20/2004 7:44:38 PM
|
|
Richard L. Hamilton wrote:
> In article <7c5d4d7bdf1848bfbd4801814fc416ea@you.transsexual-humourless-flunkey.org>,
> Chadwyk Ebbledock-Sesslehotch <chadwyk.ebbledock-sesslehotch@you.swaggering-pupillary-freakface.com> writes:
> [...]
>
> Wow. No more than the 2nd or 3rd time I ever used a killfile!
Yup. Richard, you appear to have attracted the attention of
a buttplugger. Bad luck.
|
|
|
0
|
|
|
|
Reply
|
Beardy
|
12/20/2004 11:58:35 PM
|
|
Beardy wrote:
> Richard L. Hamilton wrote:
>> In article
>> <7c5d4d7bdf1848bfbd4801814fc416ea@you.transsexual-humourless-flunkey.org>,
>> Chadwyk Ebbledock-Sesslehotch
>> <chadwyk.ebbledock-sesslehotch@you.swaggering-pupillary-freakface.com>
>> writes: [...] Wow. No more than the 2nd or 3rd time I ever used a
>> killfile!
>
> Yup. Richard, you appear to have attracted the attention of
> a buttplugger. Bad luck.
You?
|
|
|
0
|
|
|
|
Reply
|
Diogenes
|
12/21/2004 12:08:22 AM
|
|
begin In <cq10hh$or8$6@anubis.demon.co.uk>, on 12/18/2004
at 10:24 AM, huge@ukmisc.org.uk (Huge) said:
>And you're posting non-ASCII characrters in Usenet headers.
What you quoted was ASCII. MIME encoded characters are legitimate in
headers; what is not legitimate is raw non-ASCII. Which, of course,
doesn't change any of the other issues.
--
Shmuel (Seymour J.) Metz, SysProg and JOAT <http://patriot.net/~shmuel>
Unsolicited bulk E-mail subject to legal action. I reserve the
right to publicly post or ridicule any abusive E-mail. Reply to
domain Patriot dot net user shmuel+news to contact me. Do not
reply to spamtrap@library.lspace.org
|
|
|
0
|
|
|
|
Reply
|
Shmuel
|
12/21/2004 12:24:33 AM
|
|
UNIX admin wrote:
> Freeride wrote:
>
>> Bullshit! Sun's CDDL conflicts with the GPL and no Open Source
>> developer is
>> going to touch it with a 20 foot stick. And I am sure that Sun would love
>> to have some of its code show up in Linux so it can throw a big giant
>> monkey wrench into the development and adoption of Linux.
>
>
> We say "10 foot pole"...
>
> Anyways, GOOD! I don't want to see Solaris's ADVANCED stuff in a P.O.S.
> renegade OS codenamed "Linux". I want Solaris to kick Linux's butt,
> because Linux *IS* a P.O.S. Right now Linux has the upper hand because
> of wider driver and fanatical support from PC-bucket kids. However, if
> Sun doesn't screw this up, the P.O.S. Linux OS will suck it.
>
> And yes, I openly hate Linux, because I have to work with it every day
> for a living and because it always causes problems. And yes, I know how
> to properly configure & admin a Linux system, thank you. It's just that
> it's crap compared to Solaris, not to mention compared to IRIX.
>
> Oh, and BTW: whoever doesn't like corporate UNIX can shove Linux where
> the Sun don't shine. I think I pretty much said everything with that
> last sentence.
Jeez... Can you bottle some of that wisdom and put it in a
global crop-duster? We need it. Not (never) enuf people are
playing the table-banging advocate of Solaris currently.
Great to see that you are. Thanks.
|
|
|
0
|
|
|
|
Reply
|
Beardy
|
12/21/2004 12:54:55 AM
|
|
Diogenes wrote:
> Beardy wrote:
>
>>Richard L. Hamilton wrote:
>>
>>>In article
>>><7c5d4d7bdf1848bfbd4801814fc416ea@you.transsexual-humourless-flunkey.org>,
>>>Chadwyk Ebbledock-Sesslehotch
>>><chadwyk.ebbledock-sesslehotch@you.swaggering-pupillary-freakface.com>
>>>writes: [...] Wow. No more than the 2nd or 3rd time I ever used a
>>>killfile!
>>
>>Yup. Richard, you appear to have attracted the attention of
>>a buttplugger. Bad luck.
>
>
> You?
:-D
|
|
|
0
|
|
|
|
Reply
|
Beardy
|
12/21/2004 1:06:32 AM
|
|
Beardy wrote:
> Diogenes wrote:
>
>> Beardy wrote:
>>
>>> Richard L. Hamilton wrote:
>>>
>>>> In article
>>>> <7c5d4d7bdf1848bfbd4801814fc416ea@you.transsexual-humourless-flunkey.org>,
>>>>
>>>> Chadwyk Ebbledock-Sesslehotch
>>>> <chadwyk.ebbledock-sesslehotch@you.swaggering-pupillary-freakface.com>
>>>> writes: [...] Wow. No more than the 2nd or 3rd time I ever used a
>>>> killfile!
>>>
>>>
>>> Yup. Richard, you appear to have attracted the attention of
>>> a buttplugger. Bad luck.
>>
>>
>>
>> You?
>
>
> :-D
That was a laugh in the face, rather than admission. Natch.
|
|
|
0
|
|
|
|
Reply
|
Beardy
|
12/21/2004 1:09:07 AM
|
|
Beardy wrote:
> Beardy wrote:
>> Diogenes wrote:
>>
>>> Beardy wrote:
>>>
>>>> Richard L. Hamilton wrote:
>>>>
>>>>> In article
>>>>> <7c5d4d7bdf1848bfbd4801814fc416ea@you.transsexual-humourless-flunkey.org>,
>>>>>
>>>>> Chadwyk Ebbledock-Sesslehotch
>>>>> <chadwyk.ebbledock-sesslehotch@you.swaggering-pupillary-freakface.com>
>>>>> writes: [...] Wow. No more than the 2nd or 3rd time I ever used a
>>>>> killfile!
>>>>
>>>>
>>>> Yup. Richard, you appear to have attracted the attention of
>>>> a buttplugger. Bad luck.
>>>
>>>
>>>
>>> You?
>>
>>
>> :-D
>
> That was a laugh in the face, rather than admission. Natch.
Natch. (Tacit admission noted.)
|
|
|
0
|
|
|
|
Reply
|
Diogenes
|
12/21/2004 1:15:44 AM
|
|
Diogenes, <diogenes@sinope.gr>, the stochastic, mellowing fleck, and circus
clown, uttered:
> Beardy wrote:
>> Beardy wrote:
>>> Diogenes wrote:
>>>
>>>> Beardy wrote:
>>>>
>>>>> Richard L. Hamilton wrote:
>>>>>
>>>>>> In article
>>>>>>
<7c5d4d7bdf1848bfbd4801814fc416ea@you.transsexual-humourless-flunkey.org>,
>>>>>>
>>>>>> Chadwyk Ebbledock-Sesslehotch
>>>>>>
<chadwyk.ebbledock-sesslehotch@you.swaggering-pupillary-freakface.com>
>>>>>> writes: [...] Wow. No more than the 2nd or 3rd time I ever used
>>>>>> a killfile!
>>>>>
>>>>>
>>>>> Yup. Richard, you appear to have attracted the attention of
>>>>> a buttplugger. Bad luck.
>>>>
>>>>
>>>>
>>>> You?
>>>
>>>
>>> :-D
>>
>> That was a laugh in the face, rather than admission. Natch.
>
> Natch. (Tacit admission noted.)
And of course, most of us here always deny 20-20 hindsight a look see, eh.
--
Registered Linux User: #344402
Akhenaten: Registered Linux Machine: #235500
2.6.8.1-12-i686-smp Dual i686 Intel(R) Xeon(TM) CPU 2.80GHz GNU/Linux
|
|
|
0
|
|
|
|
Reply
|
Desulfurococcales
|
12/21/2004 1:20:12 AM
|
|
Mike Cox wrote:
> Now that Solaris 10 is going to be open sourced, Linux advocates rightly
> fear that Linux will be abandoned in favor of Solaris 10. Similar to how
> BSD was left in droves for Linux. That's right, SUN's brilliant strategy of
> making Solaris open sourced is using the linux advocates' techniques
> against them.
Bla, bla, bla. Nonsenses.
Who fears that Sun have to Open Source his operating system because
everybody it's switching from Solaris to Linux? Not me. You can be
sure. It's Sun who it's trying to survive. Not Linux. Linux don't
have to do marketing or empresarial decisions because they are
losing market. It's Sun that it's doing that. And that's all folks.
> The reason people will move to Solaris in droves is the same that caused
> BSD's user flight. It supports more enterprise features. But the
> difference this time is that Solaris is cheaper than Redhat, is completely
> backwards compatible, and scales much better on many more CPUs.
More enterprise features? Cheaper than Redhat?
Do you know how much it costs Solaris consultory? Or Solaris support?
A Solaris system will still be being extremely expensive even if they
open source the main system. Just think about all the software you
need to buy when you have a Solaris system, software that you can have
for free with Linux. Solaris will a TCO much greater than Linux.
> Now if only SUN would release its compiler under a BSD license....
Why you want their compiler? Everybody uses GCC under Solaris.
--
Jose Maria Lopez Hernandez
Director Tecnico de bgSEC
jkerouac@bgsec.com
bgSEC Seguridad y Consultoria de Sistemas Informaticos
http://www.bgsec.com
ESPA�A
The only people for me are the mad ones -- the ones who are mad to live,
mad to talk, mad to be saved, desirous of everything at the same time,
the ones who never yawn or say a commonplace thing, but burn, burn, burn
like fabulous yellow Roman candles.
-- Jack Kerouac, "On the Road"
|
|
|
0
|
|
|
|
Reply
|
Jose
|
12/21/2004 1:44:29 AM
|
|
Rich Teer wrote:
> Only by people who are as out of touch as you are. As documented
> on this page: http://www.sun.com/2004-1012/feature/ Apache 1.3.29
> on Solaris 10 can handle substantially more connections per second
> than RHEL 3.0.
Let's be serious. It's a page from the Sun site. Ask people who
have to use Solaris everyday and you will have the truth about
the story.
--
Jose Maria Lopez Hernandez
Director Tecnico de bgSEC
jkerouac@bgsec.com
bgSEC Seguridad y Consultoria de Sistemas Informaticos
http://www.bgsec.com
ESPA�A
The only people for me are the mad ones -- the ones who are mad to live,
mad to talk, mad to be saved, desirous of everything at the same time,
the ones who never yawn or say a commonplace thing, but burn, burn, burn
like fabulous yellow Roman candles.
-- Jack Kerouac, "On the Road"
|
|
|
0
|
|
|
|
Reply
|
Jose
|
12/21/2004 1:46:17 AM
|
|
Edwin wrote:
> If Solaris is better than Linux, and it really does become open source,
> why would current Linux users have any problem switching to it?
Because it won't be really open source as the GPL.
> What Eltee wrote is more likely, that Linux will adopt all the best
> parts of Solaris, and become an even better OS.
The same problem. What Sun really wants it's people working in their
sources but controlling everything, and impeding other operating
systems to use their sources. Being open source doesn't mean they
let you use the code. Microsoft have done something similar with
some of their code.
--
Jose Maria Lopez Hernandez
Director Tecnico de bgSEC
jkerouac@bgsec.com
bgSEC Seguridad y Consultoria de Sistemas Informaticos
http://www.bgsec.com
ESPA�A
The only people for me are the mad ones -- the ones who are mad to live,
mad to talk, mad to be saved, desirous of everything at the same time,
the ones who never yawn or say a commonplace thing, but burn, burn, burn
like fabulous yellow Roman candles.
-- Jack Kerouac, "On the Road"
|
|
|
0
|
|
|
|
Reply
|
Jose
|
12/21/2004 1:50:08 AM
|
|
Drazen Kacar wrote:
> Joerg Schilling wrote:
>
>
>> The result of the changes on cdrtools made by e.g. SuSE has _not_ been
>> better software but rather _broken_ software that is no longer useful
>> for even the basic intention of this software :-(
>
>
> You're in Germany, SuSE is based in Germany. I was under impression that
> the German copyright law has integrity rights under the author's moral
> rights. If that's the case you're perfectly capable of stopping the
> distribution of such modifications.
>
I don't think so. The cdrtools were licensed under the GPL, so anyone
can modify them with the only caution of releasing the changes under the
GPL. Like Linus says anyone can take the Linux sources and do whatever
they want with them. If the result it's better then the project will
be successful, if it's worse Linux will still be headed by Linus. It's
the GPL. If the SuSE version it's worse than the original version
customers will end up using the original version and SuSE will be
obligated to pack the original version.
Even a fork could be possible anytime. I have written GPL software
and I accept that, this are the rules of the free software.
>
>> As a result, a lot of SuSE customers did send bug reports via mail to _me_
>> and have been angry because of the bugs introduced not my me but by SuSE.
>
>
> So you can even prove the damage to the reputation.
I don't think so. I agree it's an unfriendly behaviour not adverting
their customers that the cdrtools SuSE were shipping were a modified
version and that any faults or bugs should be sended to SuSE and not
to Joerg Schilling, and I can understand why he got so angry.
--
Jose Maria Lopez Hernandez
Director Tecnico de bgSEC
jkerouac@bgsec.com
bgSEC Seguridad y Consultoria de Sistemas Informaticos
http://www.bgsec.com
ESPA�A
The only people for me are the mad ones -- the ones who are mad to live,
mad to talk, mad to be saved, desirous of everything at the same time,
the ones who never yawn or say a commonplace thing, but burn, burn, burn
like fabulous yellow Roman candles.
-- Jack Kerouac, "On the Road"
|
|
|
0
|
|
|
|
Reply
|
Jose
|
12/21/2004 2:02:41 AM
|
|
Casper H.S. Dik wrote:
> But the fact remains that the GPL restricts people from using "free"
> software; you cannot dispute that because I have proven that.
The GPL restricts people from stealing the intellectual property of
programmers. Everyone can use and distribute free software, but they
can't use it in closed source products because that implicates stealing
the IP.
--
Jose Maria Lopez Hernandez
Director Tecnico de bgSEC
jkerouac@bgsec.com
bgSEC Seguridad y Consultoria de Sistemas Informaticos
http://www.bgsec.com
ESPA�A
The only people for me are the mad ones -- the ones who are mad to live,
mad to talk, mad to be saved, desirous of everything at the same time,
the ones who never yawn or say a commonplace thing, but burn, burn, burn
like fabulous yellow Roman candles.
-- Jack Kerouac, "On the Road"
|
|
|
0
|
|
|
|
Reply
|
Jose
|
12/21/2004 2:10:48 AM
|
|
Joerg Schilling wrote:
> BTW: even weapons in your hand have nothing in common with freedom. They just
> give you force about others. If you don't like to threaten the freedom of others
> you don't need weapons. What kind of threat weapons (specially in the USA) press
> on others may easily be proved by comparing fatalities caused by weapons in the
> USA to the numbers in the rest of the world.
Almost every European will agree with that. You have explained it
perfectly. Why does American people doesn't understand it?
--
Jose Maria Lopez Hernandez
Director Tecnico de bgSEC
jkerouac@bgsec.com
bgSEC Seguridad y Consultoria de Sistemas Informaticos
http://www.bgsec.com
ESPA�A
The only people for me are the mad ones -- the ones who are mad to live,
mad to talk, mad to be saved, desirous of everything at the same time,
the ones who never yawn or say a commonplace thing, but burn, burn, burn
like fabulous yellow Roman candles.
-- Jack Kerouac, "On the Road"
|
|
|
0
|
|
|
|
Reply
|
Jose
|
12/21/2004 2:15:00 AM
|
|
Kordia Zanoeta Acidocaldarius wrote:
> If safety is a lesser ideal than freedom, do you really have freedom when
> you're not safe to walk down the street?
Talking about freedom in a country where you can walk down the
street loaded with guns but if the police gets you with a blunt
of pot you go to jail it's amazing. Americans are mad.
--
Jose Maria Lopez Hernandez
Director Tecnico de bgSEC
jkerouac@bgsec.com
bgSEC Seguridad y Consultoria de Sistemas Informaticos
http://www.bgsec.com
ESPA�A
The only people for me are the mad ones -- the ones who are mad to live,
mad to talk, mad to be saved, desirous of everything at the same time,
the ones who never yawn or say a commonplace thing, but burn, burn, burn
like fabulous yellow Roman candles.
-- Jack Kerouac, "On the Road"
|
|
|
0
|
|
|
|
Reply
|
Jose
|
12/21/2004 2:17:50 AM
|
|
Richard L. Hamilton wrote:
> In article <846b85f26222495a9f4b558927d54723@you.orwellian-goat-herded-febrifuge.net>,
> Life is violent; people die, whether at the hands of another, or from their
> own stupidity. People who are old enough and should know better, should
> not be protected from themselves or from protecting themselves. As far as
> I'm concerned, people shouldn't even be rescued from the consequences of
> their own stupidity except by strictly private charity (of which there
> should be more than there is).
So you think ALL the drugs should be legal? Don't you?
--
Jose Maria Lopez Hernandez
Director Tecnico de bgSEC
jkerouac@bgsec.com
bgSEC Seguridad y Consultoria de Sistemas Informaticos
http://www.bgsec.com
ESPA�A
The only people for me are the mad ones -- the ones who are mad to live,
mad to talk, mad to be saved, desirous of everything at the same time,
the ones who never yawn or say a commonplace thing, but burn, burn, burn
like fabulous yellow Roman candles.
-- Jack Kerouac, "On the Road"
|
|
|
0
|
|
|
|
Reply
|
Jose
|
12/21/2004 2:20:28 AM
|
|
Joerg Schilling wrote:
> Of do you are just too lazy to test a recent version of Solaris?
> It is interesting to see that because it Linux zelots did act the same way
> on Linux, Linux currently would have no more than 1000 users.
Many Linux zealots like me have to use Solaris at work. And I don't
like it. I find it inferior to Linux in so many ways I couldn't make
a single list. So why should I try OpenSolaris or whatever invent
Sun it's trying to impose us? I just don't want it. I like Linux and
I will be using Linux even if OpenSolaris it's released under the
GPL license.
--
Jose Maria Lopez Hernandez
Director Tecnico de bgSEC
jkerouac@bgsec.com
bgSEC Seguridad y Consultoria de Sistemas Informaticos
http://www.bgsec.com
ESPA�A
The only people for me are the mad ones -- the ones who are mad to live,
mad to talk, mad to be saved, desirous of everything at the same time,
the ones who never yawn or say a commonplace thing, but burn, burn, burn
like fabulous yellow Roman candles.
-- Jack Kerouac, "On the Road"
|
|
|
0
|
|
|
|
Reply
|
Jose
|
12/21/2004 2:28:04 AM
|
|
Joerg Schilling wrote:
> So one could call the GPL schizophrenic. The GPL states that it's primary
> intention is to require and grant redistribution of the source but in the
> case where I e.g. like to add a small piece of GPL code to a larger work
> under BSD license it forbids redistribution.
>
> This is because the GPL tries to enforce terms of redistribution that would
> limit the usability of the whole work which is under BSD. The only way to be
> compliant with the GPL would be to change the license of the whole work from
> BSD to GPL.
That's the idea of the GPL license. If you want to use GPL code then you
must license ALL your code as GPL. Nobody will obligate you to use their
GPL code, you can write your own code, buy it or use some BSD code.
Some people said that the GPL was a "viral license". Of course it is. It
was thought that way...
--
Jose Maria Lopez Hernandez
Director Tecnico de bgSEC
jkerouac@bgsec.com
bgSEC Seguridad y Consultoria de Sistemas Informaticos
http://www.bgsec.com
ESPA�A
The only people for me are the mad ones -- the ones who are mad to live,
mad to talk, mad to be saved, desirous of everything at the same time,
the ones who never yawn or say a commonplace thing, but burn, burn, burn
like fabulous yellow Roman candles.
-- Jack Kerouac, "On the Road"
|
|
|
0
|
|
|
|
Reply
|
Jose
|
12/21/2004 2:32:54 AM
|
|
In article <Ik0yd.185090$r4.10978510@news-reader.eresmas.com>,
Jose Maria Lopez Hernandez <jkerouac@bgsec.com> wrote:
>Edwin wrote:
>The same problem. What Sun really wants it's people working in their
>sources but controlling everything, and impeding other operating
>systems to use their sources. Being open source doesn't mean they
This is what Linus Torvalds wants... dont try to apply the same
assumptions on Solaris.
This is why Linux curently is already on a declining branch.
The fact that there are still people that move towards Linux does not
matter, these people just don't know enough about Linux.
A lot of people I know did already move apart from Linux.
Unfortumately not to Solaris but to MacOS X. The reason
for these people is that the Linux kernel people do no longer take
the Linux users for serious. Look what happened to CD/DVD recording
in the Linux kernel this year. A lot of bugs and unneeded incompatibilities
have been introduced and nobody cares about fixing even reported bugs.
--
EMail:joerg@schily.isdn.cs.tu-berlin.de (home) J�rg Schilling D-13353 Berlin
js@cs.tu-berlin.de (uni) If you don't have iso-8859-1
schilling@fokus.fraunhofer.de (work) chars I am J"org Schilling
URL: http://www.fokus.fraunhofer.de/usr/schilling ftp://ftp.berlios.de/pub/schily
|
|
|
0
|
|
|
|
Reply
|
js
|
12/21/2004 9:21:30 PM
|
|
On Sun, 19 Dec 2004 04:10:43 GMT, Jim Richardson <warlock@eskimo.com> wrote:
>
> You have a very poor understanding of sociology. Free clue, in the UK,
> violent crime is on the rise, and has been since they started
> instituting their gun laws. Prior to 1921, anyone could buy just about
> anything. Upto and including machineguns. Crime was low, and murder,
> almos unheard of.
>
> Ooops, there goes one more gun fetishist theory down the drain...
Well, I'd wait a bit longer before leaping to any conclusions (though I
have little doubt that you would be most happy to have your prejudices
comforted by such a conclusion). Handguns were only banned a few years
ago, and murder by firearms is at such a low level (around 1 or 2 per million
inhabitants per year) that it's difficult to say that the changes are
significant. Having said that, I'm rather pessimistic, as I imagine that
firearms are entering the UK illegally from the Balkans and Eastern
Europe. Unless of course you believe that firearms offences somehow
increase even when there are none of them.
A bientot
Paul
--
Paul Floyd http://paulf.free.fr (for what it's worth)
Surgery: ennobled Gerald.
|
|
|
0
|
|
|
|
Reply
|
Paul
|
12/21/2004 9:54:58 PM
|
|
On Mon, 20 Dec 2004 19:17:08 +0000 (UTC),
Seongbae Park <Seongbae.Park@Sun.COM> wrote:
> Lin�nut <lin�nut@bone.com> wrote:
> ...
>>> (It's also a pain that there's no C++ ABI; i.e. straight C will link
>>> between gcc and Sun's (or some other) compiler, but C++ typically won't.)
>>
>> Why can't we have a name-mangling standard??? <grin>
>
> Just in case this misleads some other misinformed people
> to believe gcc and Sun are quibbling with each other
> over a petty "name mangling" stuff
> (not that name mangling is a petty issue),
> there's much more than name-mangling in C++ ABI -
> in particular, object model, exception handling and RTTI.
> And those are core implementation details and can not easily be
> made (or changed) to be compatible.
Isn't the ANSI/ISO committee doing work on a C++ ABI? I thought I read
something about it once.
A bientot
Paul
--
Paul Floyd http://paulf.free.fr (for what it's worth)
Surgery: ennobled Gerald.
|
|
|
0
|
|
|
|
Reply
|
Paul
|
12/21/2004 9:58:13 PM
|
|
Jose Maria Lopez Hernandez wrote:
> Joerg Schilling wrote:
>
>> Of do you are just too lazy to test a recent version of Solaris?
>> It is interesting to see that because it Linux zelots did act the same
>> way
>> on Linux, Linux currently would have no more than 1000 users.
>
>
> Many Linux zealots like me have to use Solaris at work. And I don't
> like it. I find it inferior to Linux in so many ways I couldn't make
> a single list. So why should I try OpenSolaris or whatever invent
> Sun it's trying to impose us? I just don't want it. I like Linux and
> I will be using Linux even if OpenSolaris it's released under the
> GPL license.
>
The difference is really when you have a problem with an
operating environment: if you have a trusted, stable,
reliable, scalable, yet forward-facing environment like
Solaris (and especially Solaris 10 to be), you can rely
upon a depth of technical excellence to sort out any problem
that you may encounter (OK, it may cost a bit, but Usenet is
free); whereas with Linux, if you have a problem with the
operating environment, you have to depend on a
malhygenically-endowed coder to scrape itself off the
tarmac, get home, reheat the mug of caffeine (no bad
thing), fight the dog for the last piece of half-gnawed
pizza from the floor, put on Eminem track at lethally high
volume (in earphones), and then work on the problem. The fix
in this scenario could be approximately 3 minutes, or
failing that, 3 months (with an astronomical beer-bill (and
hospital charges) on top of the charge for the support
contract).
|
|
|
0
|
|
|
|
Reply
|
Beardy
|
12/21/2004 10:36:58 PM
|
|
Jose Maria Lopez Hernandez wrote:
> Richard L. Hamilton wrote:
>
>> In article
>> <846b85f26222495a9f4b558927d54723@you.orwellian-goat-herded-febrifuge.net>,
>>
>> Life is violent; people die, whether at the hands of another, or from
>> their
>> own stupidity. People who are old enough and should know better, should
>> not be protected from themselves or from protecting themselves. As
>> far as
>> I'm concerned, people shouldn't even be rescued from the consequences of
>> their own stupidity except by strictly private charity (of which there
>> should be more than there is).
>
>
> So you think ALL the drugs should be legal? Don't you?
>
<alleged_opinions>
<2_cents>
If you want to do drugs/whatever that has a percentage
chance of killing you, then it *is* your incontrovertible
choice. You can claim stupidty, but you're possibly dead
already. It should be considered as Euthanasia; and
Euthanasia should be legalised (speaking for the UK) - we
should all be allowed to choose the moment/method of our
exit. ECHR should get beaten up over this.
Use of drugs by under-18's? - no. Not under any
circumstances (except medical).
Use of drugs/drug-induced effects to facilitate the purchase
of further drugs? - no. Not under any circumstances.
Use of drugs to have a great f*cking laugh, and not care
about pissing anyone else off? Yes.
</2_cents>
</alleged_opinions>
|
|
|
0
|
|
|
|
Reply
|
Beardy
|
12/21/2004 10:52:28 PM
|
|
On Tue, 21 Dec 2004, Jose Maria Lopez Hernandez wrote:
> More enterprise features? Cheaper than Redhat?
Yes and yes.
> Do you know how much it costs Solaris consultory? Or Solaris support?
Last time I checked, access to Usenet was free.
> A Solaris system will still be being extremely expensive even if they
> open source the main system. Just think about all the software you
> need to buy when you have a Solaris system, software that you can have
> for free with Linux. Solaris will a TCO much greater than Linux.
Err, wake up call: that free software you get for Linux can (usually)
be compiled and run on Solaris, also free. Oh, and Solaris runs on
inexpensive, low end systems, just like Linux.
> Why you want their compiler? Everybody uses GCC under Solaris.
Well make up your mind: in the previous paragraph you talk about
all the free software that's available on Linux that must be bought
for Solaris, and now you're saying (correctly) that it is available
for free on Solaris too. I guess you've inadervtaly proved my point.
OBTW, I don't use gcc on Solaris. I prefer to use Sun's compiler,
because it produces better code.
--
Rich Teer, SCNA, SCSA, author of "Solaris Systems Programming"
. * * . * .* .
. * . .*
President, * . . /\ ( . . *
Rite Online Inc. . . / .\ . * .
.*. / * \ . .
. /* o \ .
Voice: +1 (250) 979-1638 * '''||''' .
URL: http://www.rite-online.net ******************
|
|
|
0
|
|
|
|
Reply
|
Rich
|
12/21/2004 11:17:45 PM
|
|
In article <3l0yd.185095$r4.10978275@news-reader.eresmas.com>,
Jose Maria Lopez Hernandez <jkerouac@bgsec.com> writes:
> Richard L. Hamilton wrote:
>> In article <846b85f26222495a9f4b558927d54723@you.orwellian-goat-herded-febrifuge.net>,
>> Life is violent; people die, whether at the hands of another, or from their
>> own stupidity. People who are old enough and should know better, should
>> not be protected from themselves or from protecting themselves. As far as
>> I'm concerned, people shouldn't even be rescued from the consequences of
>> their own stupidity except by strictly private charity (of which there
>> should be more than there is).
>
> So you think ALL the drugs should be legal? Don't you?
For anyone 21 or older, maybe - provided that it were possible to convict
someone who kills another person while impaired of first degree murder.
But anyone convicted of selling to minors should be executed. Heck,
people convicted of a whole lot of things should be executed...
--
mailto:rlhamil@smart.net http://www.smart.net/~rlhamil
Lasik/PRK theme music:
"In the Hall of the Mountain King", from "Peer Gynt"
|
|
|
0
|
|
|
|
Reply
|
Richard
|
12/22/2004 1:09:27 AM
|
|
On Tue, 21 Dec 2004 03:20:28 +0100, Jose Maria Lopez Hernandez wrote:
> So you think ALL the drugs should be legal? Don't you?
Yes, but people should be prosecuted for the consequences (like injuring
people while high, killing people while high, causing property damage
while high, and any other such actions that directly cause harm to
people).
Why is it that here in the USA (and certain other parts of the world) "my
body, my choice" only applies to women, and only then to reproduction?
|
|
|
0
|
|
|
|
Reply
|
Liam
|
12/22/2004 2:43:52 AM
|
|
In article <cqa44q$ga3$1@news.cs.tu-berlin.de> (Tue, 21 Dec 2004 21:21:30
+0000), Joerg Schilling wrote:
> In article <Ik0yd.185090$r4.10978510@news-reader.eresmas.com>,
> Jose Maria Lopez Hernandez <jkerouac@bgsec.com> wrote:
>
>>The same problem. What Sun really wants it's people working in their
>>sources but controlling everything, and impeding other operating
>>systems to use their sources. Being open source doesn't mean they
>
> This is what Linus Torvalds wants... dont try to apply the same
> assumptions on Solaris.
It's an assumption supported fully by the evidence: Sun's floating of its
new C[sic]DDL is evidence they intend to claim to support "open source"
[sic] while retaining full control.
> This is why Linux curently is already on a declining branch.
This is a result of Sun's continuing decline and its new marriage to
Microsoft. While its customers leave and potential new customers are
driven away by Sun's publicdissembling, Sun has been forced, as has been
Microsoft, to use marketing as its main product.
> The fact that there are still people that move towards Linux does not
> matter, these people just don't know enough about Linux.
The fact that there are many people moving away from Solaris matters;
these people know about Solaris and more importantly about Sun
Microsystems.
> A lot of people I know did already move apart from Linux.
Of course they did. They've been misled by you on every detail of the GPL.
--
"It was a trojan now it was identified that it was that machine the
trojan was not identified but it was identified as being that machine."
-- Copyright 2003 by Jamie. Used with permission.
<3sJnb.487317B1.20557@news04.bloor.is.net.cable.rogers.com>
|
|
|
0
|
|
|
|
Reply
|
Hamilcar
|
12/22/2004 3:05:28 AM
|
|
Here in comp.unix.solaris,
js@cs.tu-berlin.de (Joerg Schilling) spake unto us, saying:
>In article <GekxBpHpvmWR092yn@visi.com>,
>Richard Steiner <rsteiner@visi.com> wrote:
>
>>But you can. All you have to do is re-release the BSD code under the
>>GPL, something which is probably permitted under the BSDL.
>
>Do you _really_ like to make the proposal that the whole FreeBSD kernel
>should be relicensed under GPL?
Why not? Release it under both licenses. That way, GPL'd code can be
added legitimately to the public version of the FreeBSD kernel, but the
BSDL portions can still be used in proprietary ventures.
I think it's an amusing idea, anyway. :-)
More to the point: since the BSDL doesn't really defend the FreeBSD
code again such an action (by my layman's reading; IANAL), I'm actually
a little bit surpised that a pro-GPL group hasn't done it themselves.
--
-Rich Steiner >>>---> http://www.visi.com/~rsteiner >>>---> Smyrna, GA USA
OS/2 + eCS + Linux + Win95 + DOS + PC/GEOS + Executor = PC Hobbyist Heaven!
WARNING: I've seen FIELDATA FORTRAN V and I know how to use it!
The Theorem Theorem: If If, Then Then.
|
|
|
0
|
|
|
|
Reply
|
rsteiner
|
12/22/2004 3:37:43 AM
|
|
Lin�nut <lin�nut@bone.com> writes:
> Casper H.S Dik poked his little head through the XP firewall and said:
>
> > If you compare the UI/Solaris threads with the POSIX threads you will
> > notice interesting differences such as all kinds of *_init functions in
> > pthreads; once you've been doing software engineering for a while
> > you'll understand that difference and why pthreads are superior even
> > though UI threads are much easier to use.
>
> I've gotten into POSIX threads lately. Cool! Much better than Windows
> threads.
That's good. Now when you start needing enterprise functionality, you'll
go with Solaris and true pthreads threading.
> Of course, I've had to wrapper them so that my code will work on both
> platforms.
>
> --
> Linux, because a CPU is a terrible thing to waste.
|
|
|
0
|
|
|
|
Reply
|
Mike
|
12/22/2004 4:21:27 AM
|
|
Rich Teer <rich.teer@rite-group.com> writes:
> On Tue, 21 Dec 2004, Jose Maria Lopez Hernandez wrote:
>
> > More enterprise features? Cheaper than Redhat?
>
> Yes and yes.
>
> > Do you know how much it costs Solaris consultory? Or Solaris support?
>
> Last time I checked, access to Usenet was free.
EXACTLY! And if you visit comp.unix.solaris, you even have someone who
wrote a huge volume on Solaris programming!
> > A Solaris system will still be being extremely expensive even if they
> > open source the main system. Just think about all the software you
> > need to buy when you have a Solaris system, software that you can have
> > for free with Linux. Solaris will a TCO much greater than Linux.
>
> Err, wake up call: that free software you get for Linux can (usually)
> be compiled and run on Solaris, also free. Oh, and Solaris runs on
> inexpensive, low end systems, just like Linux.
>
> > Why you want their compiler? Everybody uses GCC under Solaris.
And how much slower is GCC on just about every platform? I've read
stats saying up to 40% on x86, SPARC when compared to native Intel and
SUN compilers.
> Well make up your mind: in the previous paragraph you talk about
> all the free software that's available on Linux that must be bought
> for Solaris, and now you're saying (correctly) that it is available
> for free on Solaris too. I guess you've inadervtaly proved my point.
>
> OBTW, I don't use gcc on Solaris. I prefer to use Sun's compiler,
> because it produces better code.
>
> --
> Rich Teer, SCNA, SCSA, author of "Solaris Systems Programming"
>
> . * * . * .* .
> . * . .*
> President, * . . /\ ( . . *
> Rite Online Inc. . . / .\ . * .
> .*. / * \ . .
> . /* o \ .
> Voice: +1 (250) 979-1638 * '''||''' .
> URL: http://www.rite-online.net ******************
|
|
|
0
|
|
|
|
Reply
|
Mike
|
12/22/2004 5:10:47 AM
|
|
js@cs.tu-berlin.de (Joerg Schilling) writes:
> In article <Ik0yd.185090$r4.10978510@news-reader.eresmas.com>,
> Jose Maria Lopez Hernandez <jkerouac@bgsec.com> wrote:
> >Edwin wrote:
>
> >The same problem. What Sun really wants it's people working in their
> >sources but controlling everything, and impeding other operating
> >systems to use their sources. Being open source doesn't mean they
>
>
> This is what Linus Torvalds wants... dont try to apply the same
> assumptions on Solaris.
>
> This is why Linux curently is already on a declining branch.
> The fact that there are still people that move towards Linux does not
> matter, these people just don't know enough about Linux.
Joerg, have you considered forking the Linux project? Instead of one person
deciding what should be in the kernel, there will be experts in each branch which
will only include the best patches. I read about your problems with Linus rejecting
your great patches and instead including patches from others that were in your opinion
much worse.
A fork would preserve some of the good things about the linux kernel all the while
enabling people who understand their specific domain, such as cd writing to be the final
say in what goes in the kernel.
|
|
|
0
|
|
|
|
Reply
|
Mike
|
12/22/2004 5:16:20 AM
|
|
On Wed, 22 Dec 2004, Richard L. Hamilton wrote:
> But anyone convicted of selling to minors should be executed. Heck,
> people convicted of a whole lot of things should be executed...
[Not directed at anyone in particular] Oh, if only stupidity was
a capital offence...
--
Rich Teer, SCNA, SCSA, author of "Solaris Systems Programming"
. * * . * .* .
. * . .*
President, * . . /\ ( . . *
Rite Online Inc. . . / .\ . * .
.*. / * \ . .
. /* o \ .
Voice: +1 (250) 979-1638 * '''||''' .
URL: http://www.rite-online.net ******************
|
|
|
0
|
|
|
|
Reply
|
Rich
|
12/22/2004 5:34:56 AM
|
|
On Tue, 21 Dec 2004, Liam Slider wrote:
> Why is it that here in the USA (and certain other parts of the world) "my
> body, my choice" only applies to women, and only then to reproduction?
I don't drink, I've never smoked, and I've never taken drugs. But I
have often wondered why so many governments are hypocritical enough to
make drugs like marijuana (sp?) illegal, yet they continue to allow the
use of harmful addictive substances, like tobacco (nicotine is one of the
most addictive substances known). Well, apart from all the cash they get
in the way of taxes, of course...
I say either ban tobacco and its related drugs completely, or legalize
all of them, slapping a hefty tax on them, and having VERY severe penalties
for their abuse or crimes commited while under their influence.
But now we're wayyyy of topic!
--
Rich Teer, SCNA, SCSA, author of "Solaris Systems Programming"
. * * . * .* .
. * . .*
President, * . . /\ ( . . *
Rite Online Inc. . . / .\ . * .
.*. / * \ . .
. /* o \ .
Voice: +1 (250) 979-1638 * '''||''' .
URL: http://www.rite-online.net ******************
|
|
|
0
|
|
|
|
Reply
|
Rich
|
12/22/2004 5:48:37 AM
|
|
In article <20041221220441.862$O8@news.newsreader.com>,
Hamilcar Barca <hamilcar@tld.always.invalid> wrote:
>In article <cqa44q$ga3$1@news.cs.tu-berlin.de> (Tue, 21 Dec 2004 21:21:30
>+0000), Joerg Schilling wrote:
>>>The same problem. What Sun really wants it's people working in their
>>>sources but controlling everything, and impeding other operating
>>>systems to use their sources. Being open source doesn't mean they
>>
>> This is what Linus Torvalds wants... dont try to apply the same
>> assumptions on Solaris.
>
>It's an assumption supported fully by the evidence: Sun's floating of its
>new C[sic]DDL is evidence they intend to claim to support "open source"
>[sic] while retaining full control.
You should give yourself a kick and read both GPL and CDDL. Once you did more
that a coarse scan which means you definitely need more than three hours, you
will find that CDDL is a better suited liscense for building a community and
for creating a vital project than the GPL.
It may even be that GPL-3 will become compatible to CDDL, so what is your point?
--
EMail:joerg@schily.isdn.cs.tu-berlin.de (home) J�rg Schilling D-13353 Berlin
js@cs.tu-berlin.de (uni) If you don't have iso-8859-1
schilling@fokus.fraunhofer.de (work) chars I am J"org Schilling
URL: http://www.fokus.fraunhofer.de/usr/schilling ftp://ftp.berlios.de/pub/schily
|
|
|
0
|
|
|
|
Reply
|
js
|
12/22/2004 8:26:38 AM
|
|
In article <HwOyBpHpvmSR092yn@visi.com>,
Richard Steiner <rsteiner@visi.com> wrote:
>Here in comp.unix.solaris,
>js@cs.tu-berlin.de (Joerg Schilling) spake unto us, saying:
>
>>In article <GekxBpHpvmWR092yn@visi.com>,
>>Richard Steiner <rsteiner@visi.com> wrote:
>>
>>>But you can. All you have to do is re-release the BSD code under the
>>>GPL, something which is probably permitted under the BSDL.
>>
>>Do you _really_ like to make the proposal that the whole FreeBSD kernel
>>should be relicensed under GPL?
>
>Why not? Release it under both licenses. That way, GPL'd code can be
>added legitimately to the public version of the FreeBSD kernel, but the
>BSDL portions can still be used in proprietary ventures.
You cannot because you cannot do things that affect the rights of a third
party.
--
EMail:joerg@schily.isdn.cs.tu-berlin.de (home) J�rg Schilling D-13353 Berlin
js@cs.tu-berlin.de (uni) If you don't have iso-8859-1
schilling@fokus.fraunhofer.de (work) chars I am J"org Schilling
URL: http://www.fokus.fraunhofer.de/usr/schilling ftp://ftp.berlios.de/pub/schily
|
|
|
0
|
|
|
|
Reply
|
js
|
12/22/2004 8:30:42 AM
|
|
In article <m3y8fq7wa0.fsf@linux.local>,
Mike Cox <mikecoxlinux@yahoo.com> wrote:
>Rich Teer <rich.teer@rite-group.com> writes:
>> Err, wake up call: that free software you get for Linux can (usually)
>> be compiled and run on Solaris, also free. Oh, and Solaris runs on
>> inexpensive, low end systems, just like Linux.
>>
>> > Why you want their compiler? Everybody uses GCC under Solaris.
>
>And how much slower is GCC on just about every platform? I've read
>stats saying up to 40% on x86, SPARC when compared to native Intel and
>SUN compilers.
Intels compiler is a "native" compiler, but Sun's cmpiler is a multi
target compiler similar to GCC.
--
EMail:joerg@schily.isdn.cs.tu-berlin.de (home) J�rg Schilling D-13353 Berlin
js@cs.tu-berlin.de (uni) If you don't have iso-8859-1
schilling@fokus.fraunhofer.de (work) chars I am J"org Schilling
URL: http://www.fokus.fraunhofer.de/usr/schilling ftp://ftp.berlios.de/pub/schily
|
|
|
0
|
|
|
|
Reply
|
js
|
12/22/2004 8:32:47 AM
|
|
In article <m3u0qe7w0r.fsf@linux.local>,
Mike Cox <mikecoxlinux@yahoo.com> wrote:
>js@cs.tu-berlin.de (Joerg Schilling) writes:
>> This is what Linus Torvalds wants... dont try to apply the same
>> assumptions on Solaris.
>>
>> This is why Linux curently is already on a declining branch.
>> The fact that there are still people that move towards Linux does not
>> matter, these people just don't know enough about Linux.
>
>Joerg, have you considered forking the Linux project? Instead of one person
>deciding what should be in the kernel, there will be experts in each branch which
>will only include the best patches. I read about your problems with Linus rejecting
>your great patches and instead including patches from others that were in your opinion
>much worse.
I believe that my time is better used when I work for OpenSolaris instead.
OpenSolaris has a kernel that already has a much better structure than the Linux
kernel and it is much easier to talk to _all_ relevant Authors of the Solaris
kernel than trying to do similar things with Linux.
--
EMail:joerg@schily.isdn.cs.tu-berlin.de (home) J�rg Schilling D-13353 Berlin
js@cs.tu-berlin.de (uni) If you don't have iso-8859-1
schilling@fokus.fraunhofer.de (work) chars I am J"org Schilling
URL: http://www.fokus.fraunhofer.de/usr/schilling ftp://ftp.berlios.de/pub/schily
|
|
|
0
|
|
|
|
Reply
|
js
|
12/22/2004 8:37:56 AM
|
|
On Wed, 22 Dec 2004 05:34:56 GMT
Rich Teer <rich.teer@rite-group.com> wrote:
> Oh, if only stupidity was a capital offence...
Alas, one man's stupidity is another man's genius.
--
Stefaan
--
As complexity rises, precise statements lose meaning,
and meaningful statements lose precision. -- Lotfi Zadeh
|
|
|
0
|
|
|
|
Reply
|
Stefaan
|
12/22/2004 9:21:51 AM
|
|
On Tue, 21 Dec 2004 03:17:50 +0100
Jose Maria Lopez Hernandez <jkerouac@bgsec.com> wrote:
> Talking about freedom in a country where you can walk down the
> street loaded with guns but if the police gets you with a blunt
> of pot you go to jail it's amazing.
What you say makes no sense.
> Americans are mad.
Yes, but aren't the only people the mad ones? ;-)
--
Stefaan
--
As complexity rises, precise statements lose meaning,
and meaningful statements lose precision. -- Lotfi Zadeh
|
|
|
0
|
|
|
|
Reply
|
Stefaan
|
12/22/2004 9:25:56 AM
|
|
Joerg Schilling wrote:
> In article <m3u0qe7w0r.fsf@linux.local>,
> Mike Cox <mikecoxlinux@yahoo.com> wrote:
>
>>js@cs.tu-berlin.de (Joerg Schilling) writes:
>
>
>>>This is what Linus Torvalds wants... dont try to apply the same
>>>assumptions on Solaris.
>>>
>>>This is why Linux curently is already on a declining branch.
>>>The fact that there are still people that move towards Linux does not
>>>matter, these people just don't know enough about Linux.
>>
>>Joerg, have you considered forking the Linux project? Instead of one person
>>deciding what should be in the kernel, there will be experts in each branch which
>>will only include the best patches. I read about your problems with Linus rejecting
>>your great patches and instead including patches from others that were in your opinion
>>much worse.
>
>
> I believe that my time is better used when I work for OpenSolaris instead.
>
> OpenSolaris has a kernel that already has a much better structure than the Linux
> kernel and it is much easier to talk to _all_ relevant Authors of the Solaris
> kernel than trying to do similar things with Linux.
>
You mean like kludge your non-existent SCSI triplets onto ATAPI devices?
Welcome to the year 2004, almost 2005, dip switched devices are on the
way out.
So what turned poor little J�rg into a Solaris troll?
Did all those nasty distro's shipping cdrecord with DVD support kill
your sales of your own commercial product that charged for the same
functionality?
Pity for you others were willing to continue, where you wanted to stop
and make a buck.
|
|
|
0
|
|
|
|
Reply
|
Philip
|
12/22/2004 10:06:28 AM
|
|
In article <EGbyd.566120$%k.286788@pd7tw2no>,
Philip Callan <callanca@shaw.ca> wrote:
>> I believe that my time is better used when I work for OpenSolaris instead.
>>
>> OpenSolaris has a kernel that already has a much better structure than the Linux
>> kernel and it is much easier to talk to _all_ relevant Authors of the Solaris
>> kernel than trying to do similar things with Linux.
>>
>
>You mean like kludge your non-existent SCSI triplets onto ATAPI devices?
They did exist on Solaris from the beginning, I don't need to do anything
here. You better should try to educate yourself and have a look at other
OS instead of only looking at you poor Linux centric world.
--
EMail:joerg@schily.isdn.cs.tu-berlin.de (home) J�rg Schilling D-13353 Berlin
js@cs.tu-berlin.de (uni) If you don't have iso-8859-1
schilling@fokus.fraunhofer.de (work) chars I am J"org Schilling
URL: http://www.fokus.fraunhofer.de/usr/schilling ftp://ftp.berlios.de/pub/schily
|
|
|
0
|
|
|
|
Reply
|
js
|
12/22/2004 10:16:31 AM
|
|
Joerg Schilling wrote:
> In article <EGbyd.566120$%k.286788@pd7tw2no>,
> Philip Callan <callanca@shaw.ca> wrote:
>
>
>>>I believe that my time is better used when I work for OpenSolaris instead.
>>>
>>>OpenSolaris has a kernel that already has a much better structure than the Linux
>>>kernel and it is much easier to talk to _all_ relevant Authors of the Solaris
>>>kernel than trying to do similar things with Linux.
>>>
>>
>>You mean like kludge your non-existent SCSI triplets onto ATAPI devices?
>
>
> They did exist on Solaris from the beginning, I don't need to do anything
> here. You better should try to educate yourself and have a look at other
> OS instead of only looking at you poor Linux centric world.
>
You don't need to do anything there?
Good, stay there.
You can have fun in your little solaris niche, and while you sing it's
praises, Linux will keep spreading onto phones, pda's, desktops,
servers, kiosks and in 1001 embedded spaces. Why should I look at
another OS, when the one I use is capable of so much, from handheld to
supercomputing clusters....
So go ahead and jump ship, judging from the attitude you show on LKML
and elsewhere, to you it's all about 'open' and not at all about 'free'
Look at all your Solaris source code, and start coding applications for
the business that actually will use it, or maybe universities, you know,
the places Sun has been stuck for 20 years.
Maybe in another 5 years, you'll have 1/1000th the installed base of
users that Linux has given 'cdrecord' already, of course by then the
Linux community will have already fixed cdrecord for HD-DVD and Blu-ray
support, while your Solaris customers are grumbling.....
|
|
|
0
|
|
|
|
Reply
|
Philip
|
12/22/2004 10:44:57 AM
|
|
In article <Jecyd.561779$nl.450079@pd7tw3no>,
Philip Callan <callanca@shaw.ca> wrote:
>Maybe in another 5 years, you'll have 1/1000th the installed base of
>users that Linux has given 'cdrecord' already, of course by then the
>Linux community will have already fixed cdrecord for HD-DVD and Blu-ray
>support, while your Solaris customers are grumbling.....
You seem to have problems with reality :-(
SunOS did have more than a million users before Linux exists...
Solaris users did write DVD using cdrecord since March 1998 long before
Linux users could afford DVD writers.
The Linux failed to add anything useful to cdrecord in the past 5 years
and they will not be able to add HD-DVD and Blu-ray. The official cdrecord
of copourse will support it when it is available.
--
EMail:joerg@schily.isdn.cs.tu-berlin.de (home) J�rg Schilling D-13353 Berlin
js@cs.tu-berlin.de (uni) If you don't have iso-8859-1
schilling@fokus.fraunhofer.de (work) chars I am J"org Schilling
URL: http://www.fokus.fraunhofer.de/usr/schilling ftp://ftp.berlios.de/pub/schily
|
|
|
0
|
|
|
|
Reply
|
js
|
12/22/2004 10:52:05 AM
|
|
In article <Pine.SOL.4.58.0412212133080.27833@zaphod>,
Rich Teer <rich.teer@rite-group.com> writes:
> On Wed, 22 Dec 2004, Richard L. Hamilton wrote:
>
>> But anyone convicted of selling to minors should be executed. Heck,
>> people convicted of a whole lot of things should be executed...
>
> [Not directed at anyone in particular] Oh, if only stupidity was
> a capital offence...
>
So true, except...which of us doesn't have an off day every now and again?
--
mailto:rlhamil@smart.net http://www.smart.net/~rlhamil
Lasik/PRK theme music:
"In the Hall of the Mountain King", from "Peer Gynt"
|
|
|
0
|
|
|
|
Reply
|
Richard
|
12/22/2004 12:42:45 PM
|
|
Jose Maria Lopez Hernandez wrote:
> Drazen Kacar wrote:
> > You're in Germany, SuSE is based in Germany. I was under impression that
> > the German copyright law has integrity rights under the author's moral
> > rights. If that's the case you're perfectly capable of stopping the
> > distribution of such modifications.
>
> I don't think so. The cdrtools were licensed under the GPL, so anyone
> can modify them with the only caution of releasing the changes under the
> GPL. [...] It's the GPL.
Releasing is always possible. The question is whether the author has any
say if he wants to stop the distribution of a derived work. GPL draws it's
power from the copyright law. If the copyright law grants certain rights
to some people, GPL can't change that.
> If the SuSE version it's worse than the original version
> customers will end up using the original version and SuSE will be
> obligated to pack the original version.
Just being "worse" is not enough. The derived version would have to be
much worse than "worse". The language is an obstacle for this kind of
discussion (I'm not capable of translating legal texts to English), so
I'll take Irish copyright law as an example.
http://www.bailii.org/ie/legis/num_act/2000/2000-28.html
109.-#(1) Subject to the exceptions and qualifications specified in
sections 110 and 111, the author of a work shall have the right to object
to any distortion, mutilation or other modification of, or other
derogatory action in relation to, the work which would prejudice his or
her reputation and that right shall also apply in relation to an
adaptation of the work.
112.-A person infringes the integrity right where he or she-
[...]
(d) makes available to the public,
a work or a copy of a work or an adaptation thereof which has, and
which he or she knows or has reason to believe has, been subjected to
any distortion, mutilation or other modification or other derogatory
action within the meaning of section 109.
I haven't studied the complete law, so it's possible that there's an
exception somewhere for this particular situation. I only wanted to
illustrate the principle. The author would have to claim that the derived
work is a "mutilation" of his work (and prove it if need be). In that case
distribution of such derived work is a copyright infringement.
Under Irish law the integrity right is not transferable (it's "incapable of
assignment or alienation"), but the author can vaiwe it off in writing
(and I don't think that the GPL qualifies as such a vaiwer). OTOH, my
copyright law grants the same right, but doesn't allow a vaiwer, so no
matter what the authors say or sign, the integrity right stays with them.
> > So you can even prove the damage to the reputation.
>
> I don't think so.
I think it depends on the jurisdiction.
> I agree it's an unfriendly behaviour not adverting
> their customers that the cdrtools SuSE were shipping were a modified
> version
Hmm. Irish law again:
113.-#(1) A person has the right not to have a work falsely attributed to
him or her as author.
--
.-. .-. Yes, I am an agent of Satan, but my duties are largely
(_ \ / _) ceremonial.
|
| dave@fly.srk.fer.hr
|
|
|
0
|
|
|
|
Reply
|
Drazen
|
12/22/2004 1:03:01 PM
|
|
Mike Cox <mikecoxlinux@yahoo.com> writes:
> That is why they post so much FUD on sites such as slashdot.org.
Let's see, cross-posted to both comp.unix.solaris and
comp.os.linux.advocacy... Hmm...
+-------------------+ .:\:\:/:/:.
| PLEASE DO NOT | :.:\:\:/:/:.:
| FEED THE TROLLS | :=.' - - '.=:
| | '=(\ 9 9 /)='
| Thank you, | ( (_) )
| Management | /`-vvv-'\
+-------------------+ / \
| | @@@ / /|,,,,,|\ \
| | @@@ /_// /^\ \\_\
@x@@x@ | | |/ WW( ( ) )WW
\||||/ | | \| __\,,\ /,,/__
\||/ | | | jgs (______Y______)
/\/\/\/\/\/\/\/\//\/\\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\
=====================================================================
Source:
http://it.geocities.com/roccopapaleo/faq/italiano/asciidontfeed.html
--
David Magda <dmagda at ee.ryerson.ca>, http://www.magda.ca/
Because the innovator has for enemies all those who have done well under
the old conditions, and lukewarm defenders in those who may do well
under the new. -- Niccolo Machiavelli, _The Prince_, Chapter VI
|
|
|
0
|
|
|
|
Reply
|
David
|
12/22/2004 2:14:33 PM
|
|
Philip Callan <callanca@shaw.ca> writes:
> [...] will keep spreading onto phones, pda's, desktops, servers,
> kiosks and in 1001 embedded spaces. Why should I look at another
> OS, when the one I use is capable of so much, from handheld to
> supercomputing clusters....
NetBSD?
(Sorry, couldn't resist. :)
--
David Magda <dmagda at ee.ryerson.ca>, http://www.magda.ca/
Because the innovator has for enemies all those who have done well under
the old conditions, and lukewarm defenders in those who may do well
under the new. -- Niccolo Machiavelli, _The Prince_, Chapter VI
|
|
|
0
|
|
|
|
Reply
|
David
|
12/22/2004 2:20:29 PM
|
|
In article <cqbbp4$3n4$1@news.cs.tu-berlin.de> (Wed, 22 Dec 2004 08:37:56
+0000), Joerg Schilling wrote:
> I believe that my time is better used when I work for OpenSolaris instead.
Great! Please limit your postings to the Open[sic]Solaris newsgroups.
> OpenSolaris has a kernel that already has a much better structure than the Linux
> kernel and it is much easier to talk to _all_ relevant Authors of the Solaris
> kernel than trying to do similar things with Linux.
If you think the "relevant Authors of the Solaris kernel" are going to be
any more receptive to your attempts to dictate development, you're going
to be disappointed again.
--
"[P]eople who use MSN's search engine probably...expose Microstupidware
to the Internet and so try to flood my mailbox with virus and worm
spoors. They probably also send HTML encrypted mail."
-- Vernon Schryver, in <c7ocrs@calcite.rhyolite.com>
|
|
|
0
|
|
|
|
Reply
|
Hamilcar
|
12/22/2004 5:01:53 PM
|
|
In article <cqbjkl$ma3$1@news.cs.tu-berlin.de> (Wed, 22 Dec 2004 10:52:05
+0000), Joerg Schilling wrote:
> SunOS did have more than a million users before Linux exists...
So did IBM's OS/360.
> Solaris users did write DVD using cdrecord since March 1998 long before
> Linux users could afford DVD writers.
Your unhappiness at being rejected leads you to inane personal insult.
> The Linux failed to add anything useful to cdrecord in the past 5 years
> and they will not be able to add HD-DVD and Blu-ray.
Of course they won't. Without your "sufficient knowledge", no operating
system development will occur anywhere.
> The official cdrecord of copourse will support it when it is available.
Of course it will. See you.
--
ITB: How worried are you about the number of attacks on Microsoft software?
BG: Actually in a sense it's very good to have this maturity ...
-- IT Business interview with Bill Gates (Chairman, Microsoft). 10/29/2003
http://www.itbusiness.ca/index.asp?theaction=61&sid=53897
|
|
|
0
|
|
|
|
Reply
|
Hamilcar
|
12/22/2004 5:04:37 PM
|
|
Rich Teer wrote:
> On Wed, 22 Dec 2004, Richard L. Hamilton wrote:
>
>> But anyone convicted of selling to minors should be executed. Heck,
>> people convicted of a whole lot of things should be executed...
>
> [Not directed at anyone in particular] Oh, if only stupidity was
> a capital offence...
You wouldn't be here.
|
|
|
0
|
|
|
|
Reply
|
Damian
|
12/22/2004 5:12:48 PM
|
|
Joerg Schilling wrote:
> In article <Jecyd.561779$nl.450079@pd7tw3no>,
> Philip Callan <callanca@shaw.ca> wrote:
>
>
>>Maybe in another 5 years, you'll have 1/1000th the installed base of
>>users that Linux has given 'cdrecord' already, of course by then the
>>Linux community will have already fixed cdrecord for HD-DVD and Blu-ray
>>support, while your Solaris customers are grumbling.....
>
>
> You seem to have problems with reality :-(
>
> SunOS did have more than a million users before Linux exists...
Yeah, I know, they also had a half million people calling it 'Slowaris'
and security advisories like nuts, I used Sun workstations, and SGI, and
IBM AIX, and HP-UX, like you, I have experience with a lot of operating
systems, but unlike you I haven't coded for them, I've /used/ them, and
although Solaris 10 may be a great leap for Sun, it doesn't hold a
candle to the thousands of programs and support Home/Average users will
want, and will get, with Linux and repositories of GPL software.
>
> Solaris users did write DVD using cdrecord since March 1998 long before
> Linux users could afford DVD writers.
Because SCSI drives were exorbitantly priced, as the IDE/ATAPI drives
came on the market, PC/Linux users started buying them a lot more, and
there is no doubt in my mind that there are /far/ more PC users who burn
DVD's than there are Sun users.
Just in case that wasn't clear, more people own DVD-R/RW drives on their
x86/x86_64 based PC's than there are Sun users, flat out. I recall
seeing in one of the online reports that adpotion rate of DVD burners
was just as fast, and in some countries, faster than CD burners.
> The Linux failed to add anything useful to cdrecord in the past 5 years
> and they will not be able to add HD-DVD and Blu-ray. The official cdrecord
> of copourse will support it when it is available.
Excuse me?
DVD support wasnt 'anything useful' ?
Some /YOU/ left out, because /YOU/ were trying to sell a commercial
product to add that functionality.
Like I said before, all that matters to you is 'open' not 'free'
So you'll be in good company with the rest of the Sun crowd. None of
them seem to make the distinction between 'open source' and 'free
software' either.
|
|
|
0
|
|
|
|
Reply
|
Philip
|
12/22/2004 5:33:43 PM
|
|
In article <EGbyd.566120$%k.286788@pd7tw2no> (Wed, 22 Dec 2004 10:06:28
+0000), Philip Callan wrote:
> So what turned poor little J�rg into a Solaris troll?
Poor Joerg barged into the Linux Kernel Mailing List (LKML), like a bull
in a china shop, making demands for kernel changes. After a complete
rejection, he left with bitter resentment and now harbors personal
animosity toward those who failed to do as they were instructed.
> Did all those nasty distro's shipping cdrecord with DVD support kill
> your sales of your own commercial product that charged for the same
> functionality?
He made the irrevocable decision to license his software under the GPL,
obviously without realizing such a license denied him complete personal
control. I don't think he'll make that mistake again.
--
"[T]he Boies team is really stepping up with the key generals there running
the case now as it relates to the litigation. They're firmly driving that."
-- Darl McBride. CEO, The SCO Group. 15 June 2004.
|
|
|
0
|
|
|
|
Reply
|
 hamilcar2 (2631)
|
12/22/2004 6:32:10 PM
|
|
In article <Xdiyd.567352$%k.409308@pd7tw2no>,
Philip Callan <callanca@shaw.ca> writes:
> Joerg Schilling wrote:
>> In article <Jecyd.561779$nl.450079@pd7tw3no>,
>> Philip Callan <callanca@shaw.ca> wrote:
>>
>>
>>>Maybe in another 5 years, you'll have 1/1000th the installed base of
>>>users that Linux has given 'cdrecord' already, of course by then the
>>>Linux community will have already fixed cdrecord for HD-DVD and Blu-ray
>>>support, while your Solaris customers are grumbling.....
>>
>>
>> You seem to have problems with reality :-(
>>
>> SunOS did have more than a million users before Linux exists...
>
> Yeah, I know, they also had a half million people calling it 'Slowaris'
> and security advisories like nuts, I used Sun workstations, and SGI, and
> IBM AIX, and HP-UX, like you, I have experience with a lot of operating
> systems, but unlike you I haven't coded for them, I've /used/ them, and
> although Solaris 10 may be a great leap for Sun, it doesn't hold a
> candle to the thousands of programs and support Home/Average users will
> want, and will get, with Linux and repositories of GPL software.
blastwave? sunfreeware.com? (and one or two smaller ones)
As for usability, I've seen some decent reviews of the Sun "Java" Desktop
on Linux; presumably the Solaris version is the same, so that should be
ok (modulo your definition of ok; I'm just hoping that CDE stays around
until well after that bloated crap gets a whole lot faster and more reliable).
[...]
--
mailto:rlhamil@smart.net http://www.smart.net/~rlhamil
Lasik/PRK theme music:
"In the Hall of the Mountain King", from "Peer Gynt"
|
|
|
0
|
|
|
|
Reply
|
Richard
|
12/22/2004 6:38:20 PM
|
|
Hamilcar Barca wrote:
> He made the irrevocable decision to license his software under the GPL,
> obviously without realizing such a license denied him complete personal
> control. I don't think he'll make that mistake again.
>
And thats the best part, he doesn't need to!
As DVD-burning support has shown, others are more than willing, the
source code is already freely out there under the GPL, to be modified
and extended later for blu-ray and hd-dvd, with or without him, as he
has nothing to say on the matter anymore.
|
|
|
0
|
|
|
|
Reply
|
 callanca (1267)
|
12/22/2004 7:06:00 PM
|
|
begin Philip Callan wrote:
> Hamilcar Barca wrote:
>> He made the irrevocable decision to license his software under the GPL,
>> obviously without realizing such a license denied him complete personal
>> control. I don't think he'll make that mistake again.
>>
>
> And thats the best part, he doesn't need to!
>
> As DVD-burning support has shown, others are more than willing, the
> source code is already freely out there under the GPL, to be modified
> and extended later for blu-ray and hd-dvd, with or without him, as he
> has nothing to say on the matter anymore.
Thats not completely true, as he is free to fork off from his code under a
different licence. If that version proves to be way better, I am sure
people would buy it.
As it is, though, the mods made by SuSE and others work. They were
problematic for a few people, but I am sure even these rare problems will
be sorted out in due time. I certainly have never had any real problems
with the SuSE-changes
--
Who the fuck is General Failure, and why is he reading my harddisk?
|
|
|
0
|
|
|
|
Reply
|
Peter.Koehlmann (13202)
|
12/22/2004 7:16:31 PM
|
|
Peter K�hlmann wrote:
> begin Philip Callan wrote:
>
>
>>Hamilcar Barca wrote:
>>
>>>He made the irrevocable decision to license his software under the GPL,
>>>obviously without realizing such a license denied him complete personal
>>>control. I don't think he'll make that mistake again.
>>>
>>
>>And thats the best part, he doesn't need to!
>>
>>As DVD-burning support has shown, others are more than willing, the
>>source code is already freely out there under the GPL, to be modified
>>and extended later for blu-ray and hd-dvd, with or without him, as he
>>has nothing to say on the matter anymore.
>
>
> Thats not completely true, as he is free to fork off from his code under a
> different licence. If that version proves to be way better, I am sure
> people would buy it.
Of course, and that would end up in the 'competing on merits' category,
something I have never had a problem with. I love Kino, and I love k3b,
now if someone had the ingenuity to package together a program along
with cdrecord-dvd, the vcd tools, transcode etc, and make a
'totalvideosolution.rpm' that would allow me to drag/drop MiniDV
footage, or AVI or any of the 'supported' codecs under linux, and make a
menu system etc, with a GUI, I would buy it, hell I would pay 100-150$
Canadian for a linux product that did that.
As it is I have to do the capture, conversion, constructing menus,
screwing up something small a few times for good measure, burn, cross
fingers and play on consumer player, if it doesnt work, rinse/repeat, at
least from the menu stage.
> As it is, though, the mods made by SuSE and others work. They were
> problematic for a few people, but I am sure even these rare problems will
> be sorted out in due time. I certainly have never had any real problems
> with the SuSE-changes
Neither have I with the MDK changes.
I think from a 'code' perspective Joerg could have a point, talking to
the devices with their triplets is more portable, but from a 'use'
perspective, especially for those with multiple burners, using whatever
special file name people assign /dev/cdburner /dev/dvdburner etc, is a
lot easier for /users/, especially in the day of USB external burners,
which may or may not grab the same triplet if hotplugged.
|
|
|
0
|
|
|
|
Reply
|
callanca (1267)
|
12/22/2004 7:58:36 PM
|
|
In article <Xdiyd.567352$%k.409308@pd7tw2no>,
Philip Callan <callanca@shaw.ca> wrote:
>Yeah, I know, they also had a half million people calling it 'Slowaris'
>and security advisories like nuts, I used Sun workstations, and SGI, and
>IBM AIX, and HP-UX, like you, I have experience with a lot of operating
>systems, but unlike you I haven't coded for them, I've /used/ them, and
>although Solaris 10 may be a great leap for Sun, it doesn't hold a
>candle to the thousands of programs and support Home/Average users will
>want, and will get, with Linux and repositories of GPL software.
It seems that even you are one of the people who forget that on Linux
you have _only_ the GPLd software while on Solaris you have both, the
UNIX tools and the other free software for your choice......
--
EMail:joerg@schily.isdn.cs.tu-berlin.de (home) J�rg Schilling D-13353 Berlin
js@cs.tu-berlin.de (uni) If you don't have iso-8859-1
schilling@fokus.fraunhofer.de (work) chars I am J"org Schilling
URL: http://www.fokus.fraunhofer.de/usr/schilling ftp://ftp.berlios.de/pub/schily
|
|
|
0
|
|
|
|
Reply
|
js
|
12/23/2004 12:29:27 AM
|
|
begin Joerg Schilling wrote:
> In article <Xdiyd.567352$%k.409308@pd7tw2no>,
> Philip Callan <callanca@shaw.ca> wrote:
>
>>Yeah, I know, they also had a half million people calling it 'Slowaris'
>>and security advisories like nuts, I used Sun workstations, and SGI, and
>>IBM AIX, and HP-UX, like you, I have experience with a lot of operating
>>systems, but unlike you I haven't coded for them, I've /used/ them, and
>>although Solaris 10 may be a great leap for Sun, it doesn't hold a
>>candle to the thousands of programs and support Home/Average users will
>>want, and will get, with Linux and repositories of GPL software.
>
> It seems that even you are one of the people who forget that on Linux
> you have _only_ the GPLd software while on Solaris you have both, the
> UNIX tools and the other free software for your choice......
>
Interesting. What happened to X? To Apache? To other software not under GPL
license? If they vanished from my linux install, I for sure missed that
fact
--
"I don't want to belong to any club that would have me as a member."
-- Groucho Marx
|
|
|
0
|
|
|
|
Reply
|
Peter
|
12/23/2004 12:44:47 AM
|
|
Joerg Schilling wrote:
> It seems that even you are one of the people who forget that on Linux
> you have _only_ the GPLd software while on Solaris you have both, the
> UNIX tools and the other free software for your choice......
>
Okay, please list any 'UNIX tools' that Linux lacks, but Solaris has.
Not /SOLARIS SPECIFIC/ tools mind you, like dtrace etc, but a full out
'UNIX' command, that could be found on any of the Unix98 or higher spec
systems (AIX, Tru64, Solaris), but there is no equivelant in Linux.
I doubt you'll find many.
You seem to be one of those people that DONT CARE, about the freedoms
you sign away in exchange for the 'right' to run someone else's OS.
Unlike Linux, which is coded by Linus et al, but as an end-user,
belongs to you, to do with as you wish, as it places ZERO restrictions
or requirements on USAGE.
|
|
|
0
|
|
|
|
Reply
|
Philip
|
12/23/2004 12:57:24 AM
|
|
In article <UJoyd.568418$%k.434086@pd7tw2no>,
Philip Callan <callanca@shaw.ca> wrote:
>Joerg Schilling wrote:
>> It seems that even you are one of the people who forget that on Linux
>> you have _only_ the GPLd software while on Solaris you have both, the
>> UNIX tools and the other free software for your choice......
>>
>
>Okay, please list any 'UNIX tools' that Linux lacks, but Solaris has.
There are many.
tar, awk, lex, ....
On Linux you have only gtar, gawk, flex, ....
--
EMail:joerg@schily.isdn.cs.tu-berlin.de (home) J�rg Schilling D-13353 Berlin
js@cs.tu-berlin.de (uni) If you don't have iso-8859-1
schilling@fokus.fraunhofer.de (work) chars I am J"org Schilling
URL: http://www.fokus.fraunhofer.de/usr/schilling ftp://ftp.berlios.de/pub/schily
|
|
|
0
|
|
|
|
Reply
|
js
|
12/23/2004 1:09:00 AM
|
|
In article <cqd3h7$rjl$1@news.cs.tu-berlin.de> (Thu, 23 Dec 2004 00:29:27
+0000), Joerg Schilling wrote:
> It seems that even you are one of the people who forget that on Linux
> you have _only_ the GPLd software
This is a lie, Joerg.
--
"Any time [Sun Microsystems tries] to use my work to justify retaining
proprietary control or argue that Linux is somehow less open than
Java, that's either culpable stupidity or dishonesty [...]"
-- Eric S. Raymond
|
|
|
0
|
|
|
|
Reply
|
Hamilcar
|
12/23/2004 1:23:11 AM
|
|
In article <cqd5rc$15k$1@news.cs.tu-berlin.de> (Thu, 23 Dec 2004 01:09:00
+0000), Joerg Schilling wrote:
>>Okay, please list any 'UNIX tools' that Linux lacks, but Solaris has.
>
> There are many.
>
> tar, awk, lex, ....
>
> On Linux you have only gtar, gawk, flex, ....
You're stretching a tiny truth extremely thin to make your unimaginative
point, Joerg. You've obviously failed to notice that the GNU tools are
the Open Standard.
Why not give up and move on to your new masterpiece, OpenSolaris?
--
"Our claims aren't trivial."
-- Darl McBride. CEO, The SCO Group. SLTribune, 19 November 2003
|
|
|
0
|
|
|
|
Reply
|
Hamilcar
|
12/23/2004 1:29:11 AM
|
|
In article <20041222202905.123$ZB@news.newsreader.com>,
Hamilcar Barca <hamilcar@tld.always.invalid> wrote:
>In article <cqd5rc$15k$1@news.cs.tu-berlin.de> (Thu, 23 Dec 2004 01:09:00
>+0000), Joerg Schilling wrote:
>
>>>Okay, please list any 'UNIX tools' that Linux lacks, but Solaris has.
>>
>> There are many.
>>
>> tar, awk, lex, ....
>>
>> On Linux you have only gtar, gawk, flex, ....
>
>You're stretching a tiny truth extremely thin to make your unimaginative
>point, Joerg. You've obviously failed to notice that the GNU tools are
>the Open Standard.
You fail to understand that the GNU tools in many cases do not follow the
the standard while the UNIX tools do.
--
EMail:joerg@schily.isdn.cs.tu-berlin.de (home) J�rg Schilling D-13353 Berlin
js@cs.tu-berlin.de (uni) If you don't have iso-8859-1
schilling@fokus.fraunhofer.de (work) chars I am J"org Schilling
URL: http://www.fokus.fraunhofer.de/usr/schilling ftp://ftp.berlios.de/pub/schily
|
|
|
0
|
|
|
|
Reply
|
js
|
12/23/2004 1:30:56 AM
|
|
On Thu, 23 Dec 2004, Peter [UTF-8] Köhlmann wrote:
> Interesting. What happened to X? To Apache? To other software not under GPL
I'm amazed you allow that non-GPL filth onto your system.
--
Rich Teer, SCNA, SCSA, author of "Solaris Systems Programming"
. * * . * .* .
. * . .*
President, * . . /\ ( . . *
Rite Online Inc. . . / .\ . * .
.*. / * \ . .
. /* o \ .
Voice: +1 (250) 979-1638 * '''||''' .
URL: http://www.rite-online.net ******************
|
|
|
0
|
|
|
|
Reply
|
Rich
|
12/23/2004 1:38:08 AM
|
|
Rich Teer wrote:
> On Thu, 23 Dec 2004, Peter [UTF-8] Köhlmann wrote:
>
>
>>Interesting. What happened to X? To Apache? To other software not under GPL
>
>
> I'm amazed you allow that non-GPL filth onto your system.
>
But it's GPL compatible, which is what matters.
|
|
|
0
|
|
|
|
Reply
|
Philip
|
12/23/2004 2:02:01 AM
|
|
In article <Pine.SOL.4.58.0412221736160.698@zaphod> (Thu, 23 Dec 2004
01:38:08 +0000), Rich Teer wrote:
>> Interesting. What happened to X? To Apache? To other software not under GPL
>
> I'm amazed you allow that non-GPL filth onto your system.
We want "free as in freedom" software. The GPL is the best of the
current licenses, but X, Apache, Mozilla, and other software is released
with free licenses. This is in complete contrast to the current state of,
for example, Java and Solaris.
--
"Making Solaris free would be a substantial contribution to [free
software], but since we already have software (GNU/Linux) that does
more or less the same thing, it would not be a major contribution..."
-- Richard M. Stallman. Founder, Free Software Foundation.
|
|
|
0
|
|
|
|
Reply
|
Hamilcar
|
12/23/2004 2:06:30 AM
|
|
Joerg Schilling wrote:
> In article <UJoyd.568418$%k.434086@pd7tw2no>,
> Philip Callan <callanca@shaw.ca> wrote:
>
>>Joerg Schilling wrote:
>>
>>>It seems that even you are one of the people who forget that on Linux
>>>you have _only_ the GPLd software while on Solaris you have both, the
>>>UNIX tools and the other free software for your choice......
>>>
>>
>>Okay, please list any 'UNIX tools' that Linux lacks, but Solaris has.
>
>
> There are many.
>
> tar, awk, lex, ....
>
> On Linux you have only gtar, gawk, flex, ....
>
Excuse me, I must be missing something (besides flex/lex)
I /have/ tar and awk, they point to programs that do the exact same
thing, and from what I can tell, the only difference is the licensing,
is there a functional difference?
Or are there options available on your 'Unix' commands, that are not
available on their freely licensed counterparts?
|
|
|
0
|
|
|
|
Reply
|
Philip
|
12/23/2004 2:14:42 AM
|
|
Hamilcar Barca wrote:
> In article <Pine.SOL.4.58.0412221736160.698@zaphod> (Thu, 23 Dec 2004
> 01:38:08 +0000), Rich Teer wrote:
>
>
>>>Interesting. What happened to X? To Apache? To other software not under GPL
>>
>>I'm amazed you allow that non-GPL filth onto your system.
>
>
> We want "free as in freedom" software. The GPL is the best of the
> current licenses, but X, Apache, Mozilla, and other software is released
> with free licenses. This is in complete contrast to the current state of,
> for example, Java and Solaris.
>
Well, you can moan and groan about it, or you can make the best of it.
--
The e-mail address in our reply-to line is reversed in an attempt to
minimize spam. Our true address is of the form che...@prodigy.net.
|
|
|
0
|
|
|
|
Reply
|
CJT
|
12/23/2004 2:15:31 AM
|
|
In article <mSpyd.568551$%k.568013@pd7tw2no>,
Philip Callan <callanca@shaw.ca> wrote:
>>>Okay, please list any 'UNIX tools' that Linux lacks, but Solaris has.
>>
>>
>> There are many.
>>
>> tar, awk, lex, ....
>>
>> On Linux you have only gtar, gawk, flex, ....
>>
>
>Excuse me, I must be missing something (besides flex/lex)
>
>I /have/ tar and awk, they point to programs that do the exact same
>thing, and from what I can tell, the only difference is the licensing,
>is there a functional difference?
This is what Linux users believe :-(
>Or are there options available on your 'Unix' commands, that are not
>available on their freely licensed counterparts?
The GNU tar on your Linux system (even is you use the latest) by default
still does not create a TAR archive but a proprietary archive format
that cannot be read correctly by a UNIX tar program.
--
EMail:joerg@schily.isdn.cs.tu-berlin.de (home) J�rg Schilling D-13353 Berlin
js@cs.tu-berlin.de (uni) If you don't have iso-8859-1
schilling@fokus.fraunhofer.de (work) chars I am J"org Schilling
URL: http://www.fokus.fraunhofer.de/usr/schilling ftp://ftp.berlios.de/pub/schily
|
|
|
0
|
|
|
|
Reply
|
js
|
12/23/2004 8:22:18 AM
|
|
Joerg Schilling wrote:
> The GNU tar on your Linux system (even is you use the latest) by default
> still does not create a TAR archive but a proprietary archive format
> that cannot be read correctly by a UNIX tar program.
>
Oh yeah, and as author of another tool performing the same function your
unbiased right? Found an older thread of you flogging this horse with PTB:
http://linux.derkeiler.com/Newsgroups/comp.os.linux.misc/2003-08/1538.html
So a few quirks, that occur in rare instances, and your complaining,
what a suprise, because it doesn't do it the way Solaris does.
And this paragraph:
http://www.fokus.gmd.de/research/cc/glone/employees/joerg.schilling/private/cdrecord.html
[Both RedHat and SuSE publish bastardized and defective variants of
cdrtools in their distributions. If you have problems on RedHat or SuSE
systems, first fetch a recent original cdrtools source, compile it
yourself and run the original instead of broken software that illegally
claims to be cdrecord. ]
Where you show how ignorant you are of what the GPL is about, which is
making a derivative if the original program doesnt do what you need.
People wanted DVD support without having to pay for it, or needing a
'key' to run it.
I'm glad to say cdrdao and dvd-rw tools have a new home.
http://lwn.net/Articles/98487/
|
|
|
0
|
|
|
|
Reply
|
Philip
|
12/23/2004 9:33:07 AM
|
|
In article <nhwyd.565399$nl.51439@pd7tw3no>,
Philip Callan <callanca@shaw.ca> wrote:
>Joerg Schilling wrote:
>
>> The GNU tar on your Linux system (even is you use the latest) by default
>> still does not create a TAR archive but a proprietary archive format
>> that cannot be read correctly by a UNIX tar program.
>>
>
>Oh yeah, and as author of another tool performing the same function your
>unbiased right? Found an older thread of you flogging this horse with PTB:
A typical answer from an uninformed zealot :-(
Check mysql..... they publish the source in a broken archive
that can only be read back using GNU tar.
But this is a result from the FSF way of thinking. Telling people that
open standards do not matter if there is only open source. If you like to
check the quality of your favorite tar implementation, check
ftp://ftp.berlios.de/pub/star/testscripts/
and the program "tartest" that comes with star sources.
--
EMail:joerg@schily.isdn.cs.tu-berlin.de (home) J�rg Schilling D-13353 Berlin
js@cs.tu-berlin.de (uni) If you don't have iso-8859-1
schilling@fokus.fraunhofer.de (work) chars I am J"org Schilling
URL: http://www.fokus.fraunhofer.de/usr/schilling ftp://ftp.berlios.de/pub/schily
|
|
|
0
|
|
|
|
Reply
|
js
|
12/23/2004 10:02:39 AM
|
|
Philip Callan wrote:
> Joerg Schilling wrote:
>
>> The GNU tar on your Linux system (even is you use the latest) by default
>> still does not create a TAR archive but a proprietary archive format
>> that cannot be read correctly by a UNIX tar program.
>>
>
> Oh yeah, and as author of another tool performing the same function your
> unbiased right? Found an older thread of you flogging this horse with PTB:
>
> http://linux.derkeiler.com/Newsgroups/comp.os.linux.misc/2003-08/1538.html
>
> So a few quirks, that occur in rare instances, and your complaining,
> what a suprise, because it doesn't do it the way Solaris does.
>
Rare occurrences? Every time I open a "tarball" created by GNU tar,
it will fail if the pathname of any file in the archive is longer than
IIRC 256 characters. Happens quite often in anything but the smallest
archives.
And I see the same ocurrence whether I use Solaris tar or AIX tar or
SGI's tar. And here I thought AIX was the bastard child. ;) The
fact remains that GNU tar is the implementation that does not follow
the published standard.
--
Coy Hile
hile@cse.psu.edu
|
|
|
0
|
|
|
|
Reply
|
Coy
|
12/23/2004 12:43:41 PM
|
|
In article <cqeehv$4on$1@neuromancer.cse.psu.edu>,
Coy Hile <hile@cse.psu.edu> wrote:
>Philip Callan wrote:
>> So a few quirks, that occur in rare instances, and your complaining,
>> what a suprise, because it doesn't do it the way Solaris does.
>>
>
>Rare occurrences? Every time I open a "tarball" created by GNU tar,
>it will fail if the pathname of any file in the archive is longer than
>IIRC 256 characters. Happens quite often in anything but the smallest
>archives.
It is 100 chars and this used to happen close to never in 1994 when
I first asked the GNU tar maintainer to introduce a standard compliant
behavior. If GNU tar had been fixed in 1994, nobody would complain now.
>SGI's tar. And here I thought AIX was the bastard child. ;) The
>fact remains that GNU tar is the implementation that does not follow
>the published standard.
Correct :-(
--
EMail:joerg@schily.isdn.cs.tu-berlin.de (home) J�rg Schilling D-13353 Berlin
js@cs.tu-berlin.de (uni) If you don't have iso-8859-1
schilling@fokus.fraunhofer.de (work) chars I am J"org Schilling
URL: http://www.fokus.fraunhofer.de/usr/schilling ftp://ftp.berlios.de/pub/schily
|
|
|
0
|
|
|
|
Reply
|
js
|
12/23/2004 12:58:42 PM
|
|
Joerg Schilling wrote:
>
> The GNU tar on your Linux system (even is you use the latest) by default
> still does not create a TAR archive but a proprietary archive format
> that cannot be read correctly by a UNIX tar program.
>
Joerg, only you could call an open format whose details are fully
available and that may be implemented freely by anyone as
"proprietary".
--
Christopher Mattern
"Which one you figure tracked us?"
"The ugly one, sir."
"...Could you be more specific?"
|
|
|
0
|
|
|
|
Reply
|
Chris
|
12/23/2004 1:26:28 PM
|
|
In article <xdadnen2LYQZWlfcRVn-hA@comcast.com>,
Chris Mattern <matternc@comcast.net> wrote:
>Joerg Schilling wrote:
>
>>
>> The GNU tar on your Linux system (even is you use the latest) by default
>> still does not create a TAR archive but a proprietary archive format
>> that cannot be read correctly by a UNIX tar program.
>>
>Joerg, only you could call an open format whose details are fully
>available and that may be implemented freely by anyone as
>"proprietary".
A format that is not even documented is proprietary regardless
of your definitions.
For an open format you need people from different places to agree on
and a description how the format needs to be implemented,
_both_ are missing for the GNU tar proprietray format.
On the other side, POSIX did finally agree in 1988 on a TAR format that
is widely usable in many implementations. This POSIX TAR format is not
proprietary.
--
EMail:joerg@schily.isdn.cs.tu-berlin.de (home) J�rg Schilling D-13353 Berlin
js@cs.tu-berlin.de (uni) If you don't have iso-8859-1
schilling@fokus.fraunhofer.de (work) chars I am J"org Schilling
URL: http://www.fokus.fraunhofer.de/usr/schilling ftp://ftp.berlios.de/pub/schily
|
|
|
0
|
|
|
|
Reply
|
js
|
12/23/2004 1:47:02 PM
|
|
Joerg Schilling wrote:
> In article <xdadnen2LYQZWlfcRVn-hA@comcast.com>,
> Chris Mattern <matternc@comcast.net> wrote:
>>Joerg Schilling wrote:
>>
>>>
>>> The GNU tar on your Linux system (even is you use the latest) by default
>>> still does not create a TAR archive but a proprietary archive format
>>> that cannot be read correctly by a UNIX tar program.
>>>
>>Joerg, only you could call an open format whose details are fully
>>available and that may be implemented freely by anyone as
>>"proprietary".
>
> A format that is not even documented is proprietary regardless
> of your definitions.
>
> For an open format you need people from different places to agree on
People all over the world who use the GNU tool set agree on this
format. How many different places do you need?
> and a description how the format needs to be implemented,
Guh? Are you claiming that the source for gtar is in fact NOT
available?
> _both_ are missing for the GNU tar proprietray format.
>
> On the other side, POSIX did finally agree in 1988 on a TAR format that
> is widely usable in many implementations. This POSIX TAR format is not
> proprietary.
>
--
Christopher Mattern
"Which one you figure tracked us?"
"The ugly one, sir."
"...Could you be more specific?"
|
|
|
0
|
|
|
|
Reply
|
Chris
|
12/23/2004 3:55:40 PM
|
|
Beardy wrote:
> Jeez... Can you bottle some of that wisdom and put it in a global
> crop-duster? We need it. Not (never) enuf people are playing the
> table-banging advocate of Solaris currently. Great to see that you are.
> Thanks.
It's just that I'm so sick & tired of the Linux hype, especially after
having to work on it and realizing how much I missed working on Solaris.
I got so sick of Linux, that when the opportunity came to work with
Solaris again, I was happy to quit my job.
Meanwhile the former employer has no clue why I quit, they seemed
unhappy to see me go because they got some nice Linux servers out of the
deal. Best of luck to them, in spite of everything, I did my job
professionally. But I look forward to working in a corporate
environment with Solaris again.
|
|
|
0
|
|
|
|
Reply
|
UNIX
|
12/23/2004 4:07:46 PM
|
|
Chris Mattern <matternc@comcast.net> writes:
> Joerg Schilling wrote:
>> A format that is not even documented is proprietary regardless
>> of your definitions.
>>
>> For an open format you need people from different places to agree on
>
> People all over the world who use the GNU tool set agree on this
> format. How many different places do you need?
So, following your arguments, whatever non-standard "improvement" Microsoft
does to TCP/IP protocols is automatically a standard. Never mind that it
broke all these 20+years old, well established and documented standars that
were used all over the world. After all, lot more people use MS Windows
than any form of UNIX...
Standard bodies (POSIX, ANSI, ISO, LSB, ...) are generally a good thing and
they encourage inter-operability.
Dragan
--
Dragan Cvetkovic,
To be or not to be is true. G. Boole No it isn't. L. E. J. Brouwer
!!! Sender/From address is bogus. Use reply-to one !!!
|
|
|
0
|
|
|
|
Reply
|
Dragan
|
12/23/2004 4:30:53 PM
|
|
Coy Hile wrote:
> Philip Callan wrote:
>
>> Joerg Schilling wrote:
>>
>>> The GNU tar on your Linux system (even is you use the latest) by default
>>> still does not create a TAR archive but a proprietary archive format
>>> that cannot be read correctly by a UNIX tar program.
>>>
>>
>> Oh yeah, and as author of another tool performing the same function
>> your unbiased right? Found an older thread of you flogging this horse
>> with PTB:
>>
>> http://linux.derkeiler.com/Newsgroups/comp.os.linux.misc/2003-08/1538.html
>>
>>
>> So a few quirks, that occur in rare instances, and your complaining,
>> what a suprise, because it doesn't do it the way Solaris does.
>>
>
> Rare occurrences? Every time I open a "tarball" created by GNU tar, it
> will fail if the pathname of any file in the archive is longer than IIRC
> 256 characters. Happens quite often in anything but the smallest archives.
Really? That's odd because to the best of my knowledge, I've never had a
tar fail cross-platform, probably because but I rarely use tar for large
backups.
I have too many DOS-era habits in that I normally use rar or zip for
compressing anything I plan on moving around.
I kind of wish I knew more about the supporting unix commands, so that I
could find a way to easily list all pathnames on my HD that currently
exceed 255 characters, because as it stands, with my text console at
80x25, I've yet to see a pathname long enough to wrap my screen THREE
lines, 'happens quite often' is a strech though, considering that
without doing some intensive digging, the most vocal complainer so far
has only been Joerg, the author of a competing tool.
I've seriously found 5-10 complaints on various boards, with most of the
'tar / gnutar incompatibilities' bringing up the 'star' project pages,
and Joerg's arguments with PTB on this issue, with a few other
complaints for good measure, most of them are 6 months old or older.
> And I see the same ocurrence whether I use Solaris tar or AIX tar or
> SGI's tar. And here I thought AIX was the bastard child. ;) The fact
> remains that GNU tar is the implementation that does not follow the
> published standard.
>
And here when I read the changelog, and see changes like this:
version 1.14 - Sergey Poznyakoff, 2004-05-11
* Added support for POSIX.1-2001 and ustar archive formats.
* New option --format allows to select the output archive format
* The default output format can be selected at configuration time
by presetting the environment variable DEFAULT_ARCHIVE_FORMAT.
Allowed values are GNU, V7, OLDGNU and POSIX.
And see that the page with status here:
http://www.gnu.org/software/tar/tar.html
Shows there have been 3 releases /SINCE/ that one that added support for
cross-compatibility between differing versions of tar.
So I don't think you and Joerg have much to complain about if Gnutar
/IS/ capable of making your portable tar files, but simply requires you
to specify theyoutput type...
|
|
|
0
|
|
|
|
Reply
|
Philip
|
12/23/2004 5:06:44 PM
|
|
Rich Teer wrote:
> On Thu, 23 Dec 2004, Peter [UTF-8] Köhlmann wrote:
>
>
>>Interesting. What happened to X? To Apache? To other software not under GPL
>
>
> I'm amazed you allow that non-GPL filth onto your system.
I've got a fair amount of BSD tools on my Gentoo system. tnftp is one.
OpenSSL/ssh is another. In fact, I've got tnftp replacing wget for
the file transfer program for Portage. There's also a fair amount of
public domain software that can be used on Linux... pdksh is one example.
|
|
|
0
|
|
|
|
Reply
|
 hackr_d (695)
|
12/23/2004 5:31:08 PM
|
|
In article <EWCyd.566636$nl.1144@pd7tw3no>,
Philip Callan <callanca@shaw.ca> wrote:
>And here when I read the changelog, and see changes like this:
>
>version 1.14 - Sergey Poznyakoff, 2004-05-11
>
>* Added support for POSIX.1-2001 and ustar archive formats.
>* New option --format allows to select the output archive format
>* The default output format can be selected at configuration time
> by presetting the environment variable DEFAULT_ARCHIVE_FORMAT.
> Allowed values are GNU, V7, OLDGNU and POSIX.
Well, the fact that after 10 years they finally managed to implement first
draft code for POSIX does not mean that GNU tar uses this format by default.
GNU tar still by default creates archives with a proprietary format.
--
EMail:joerg@schily.isdn.cs.tu-berlin.de (home) J�rg Schilling D-13353 Berlin
js@cs.tu-berlin.de (uni) If you don't have iso-8859-1
schilling@fokus.fraunhofer.de (work) chars I am J"org Schilling
URL: http://www.fokus.fraunhofer.de/usr/schilling ftp://ftp.berlios.de/pub/schily
|
|
|
0
|
|
|
|
Reply
|
js
|
12/23/2004 5:35:21 PM
|
|
Philip Callan <callanca@shaw.ca> writes:
> I kind of wish I knew more about the supporting unix commands, so that I
> could find a way to easily list all pathnames on my HD that currently
> exceed 255 characters, because as it stands, with my text console at 80x25,
> I've yet to see a pathname long enough to wrap my screen THREE lines,
Apparently, 100 characters are/were the problematic limit. The following
command will print all such paths/files
# find / 2> /dev/null | nawk '{if (length($0) > 100) print}'
Replace nawk with gawk if running on Linux.
E.g.
/usr/share/icons/HighContrastLargePrintInverse/48x48/mimetypes/gnome-mime-application-x-python-bytecode.png
is an example. Or
/usr/share/sgml/docbook/docbook-xsl-stylesheets-1.60.1/extensions/doc/saxon6/com/nwalsh/saxon6/NumberLinesEmitter.html
both on my S10 box and both apparently also available on Linux (docbook and gnome).
And these are system files. With users using names like
'Optional files to be copied to image source directory'
(part of JAlbum 4.1.4, btw), you can easily exceed the limit.
Dragan
--
Dragan Cvetkovic,
To be or not to be is true. G. Boole No it isn't. L. E. J. Brouwer
!!! Sender/From address is bogus. Use reply-to one !!!
|
|
|
0
|
|
|
|
Reply
|
Dragan
|
12/23/2004 6:27:45 PM
|
|
In article <lm7jn8c1jy.fsf@privacy.net>,
Dragan Cvetkovic <dragan_usenet@gmx.net> wrote:
>Philip Callan <callanca@shaw.ca> writes:
>
>> I kind of wish I knew more about the supporting unix commands, so that I
>> could find a way to easily list all pathnames on my HD that currently
>> exceed 255 characters, because as it stands, with my text console at 80x25,
>> I've yet to see a pathname long enough to wrap my screen THREE lines,
>
>Apparently, 100 characters are/were the problematic limit. The following
>command will print all such paths/files
This is what I already wrote and what you may read in
ftp://ftp.berlios.de/pub/star/README.otherbugs
Note that with POSIX.1-1988, you hit a maximum limit of 256 chars (100 chars +
'/' + 255 chars split at a '/').
With POSIX.1-1988, you will hit the file name limit only in rare cases if
you usr tar to archive projects.
If you like to use tar for backups, you need POSIX.1-2001. Star supports
POSIX.1-2001 since August 2001 and supports (like other later POSIX.1-2001
implementations) up to PATH_MAX chars in a filename.
Next year with star-1.6, star will remove this limit completely and allow
arbitrary file name length. Then, star will be better than ufsrestore which is
limited to PATH_MAX-1.
--
EMail:joerg@schily.isdn.cs.tu-berlin.de (home) J�rg Schilling D-13353 Berlin
js@cs.tu-berlin.de (uni) If you don't have iso-8859-1
schilling@fokus.fraunhofer.de (work) chars I am J"org Schilling
URL: http://www.fokus.fraunhofer.de/usr/schilling ftp://ftp.berlios.de/pub/schily
|
|
|
0
|
|
|
|
Reply
|
js
|
12/23/2004 6:55:41 PM
|
|
In article <cqevkp$sjq$1@news.cs.tu-berlin.de> (Thu, 23 Dec 2004 17:35:21
+0000), Joerg Schilling wrote:
> GNU tar still by default creates archives with a proprietary format.
Repeating a lie doesn't make it true.
--
"When we speak of Free Software, we are referring to freedom, not price."
-- Richard M. Stallman. Founder, Free Software Foundation.
|
|
|
0
|
|
|
|
Reply
|
Hamilcar
|
12/23/2004 7:08:59 PM
|
|
Dragan Cvetkovic wrote:
> Philip Callan <callanca@shaw.ca> writes:
>
>
>>I kind of wish I knew more about the supporting unix commands, so that I
>>could find a way to easily list all pathnames on my HD that currently
>>exceed 255 characters, because as it stands, with my text console at 80x25,
>>I've yet to see a pathname long enough to wrap my screen THREE lines,
>
>
> Apparently, 100 characters are/were the problematic limit. The following
> command will print all such paths/files
>
> # find / 2> /dev/null | nawk '{if (length($0) > 100) print}'
>
> Replace nawk with gawk if running on Linux.
>
OUCH!
I ran that (with gawk obviously) and was actually suprised at how many
files fell into that category (invariably, the windows mounts were the
worst, what with C:\Program Files\Blah Blah Publisher\Blah Blah
developer\Application\skins\level1\00012949.pcx etc etc etc) but I don't
need 99% of tar's functions to back those up, they were imaged last time
I installed XP and finished castrating it, removing all the services I
dont need, and ripping tcp/ip / network access out of it, I find a
Windows box without drivers for your ethernet card is pretty easy to
secure! :)
If I exclude the windows shares, then I'm left with libncurses5-devel's
documentation, and a whack of the applnk-mdk .desktop files, both off
the /etc branch, and copied into ~/.kde for the users.
So I can see how it could present problems, especially for those running
web servers etc, where through virtual hosts and the rest, their
pathnames end up being really really long ;)
Honestly, the .desktop files, 99% are probably the defaults that are
copied from /usr/share/ with any additions the user has added (custom
drive icons on desktop, shortcuts etc)
So the loss of shortcuts wouldnt bother me, but the fact that files in
the /etc tree fell into that category isn't reassuring.
I notice Joerg sidestepped my question when I commented about the
changes made to Gnu tar in March of this year, and three times since, if
it has fixed these problems he comments on, and if you check the
archives for gnu tar mailing list, you'll see people pointing out
features that Gnutar provides, but Solaris tar (at least 9 and under)
DOESNT.
He just avoided it, but hey, if your only selling point is 'I do x/y/z
functions that product B doesnt!' and Product B all of a sudden adds
features x/y/z kinda leaves him in a lurch for a point to crow about
doesn't it?
Kind of like how the MDK/SuSE extensions to cdrecord left his
cdrecord-dvd in a lurch (as least as far as his commercial sales were
concerned)
|
|
|
0
|
|
|
|
Reply
|
Philip
|
12/23/2004 7:12:33 PM
|
|
Dragan Cvetkovic wrote:
> Chris Mattern <matternc@comcast.net> writes:
>
>> Joerg Schilling wrote:
>
>>> A format that is not even documented is proprietary regardless
>>> of your definitions.
>>>
>>> For an open format you need people from different places to agree on
>>
>> People all over the world who use the GNU tool set agree on this
>> format. How many different places do you need?
>
> So, following your arguments, whatever non-standard "improvement"
> Microsoft does to TCP/IP protocols is automatically a standard.
No, because those "improvements" do *not* have their implementations
made available to the public. Joerg defined two points to be
non-proprietary. gtar meets them, although he tried to claim that
it did not. Microsoft's screwing with TCP/IP does not.
> Never mind
> that it broke all these 20+years old, well established and documented
> standars that were used all over the world. After all, lot more people use
> MS Windows than any form of UNIX...
>
> Standard bodies (POSIX, ANSI, ISO, LSB, ...) are generally a good thing
> and they encourage inter-operability.
As does open source.
>
> Dragan
>
--
Christopher Mattern
"Which one you figure tracked us?"
"The ugly one, sir."
"...Could you be more specific?"
|
|
|
0
|
|
|
|
Reply
|
Chris
|
12/23/2004 7:59:33 PM
|
|
-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1
On 23 Dec 2004 00:29:27 GMT,
Joerg Schilling <js@cs.tu-berlin.de> wrote:
> In article <Xdiyd.567352$%k.409308@pd7tw2no>,
> Philip Callan <callanca@shaw.ca> wrote:
>
>>Yeah, I know, they also had a half million people calling it 'Slowaris'
>>and security advisories like nuts, I used Sun workstations, and SGI, and
>>IBM AIX, and HP-UX, like you, I have experience with a lot of operating
>>systems, but unlike you I haven't coded for them, I've /used/ them, and
>>although Solaris 10 may be a great leap for Sun, it doesn't hold a
>>candle to the thousands of programs and support Home/Average users will
>>want, and will get, with Linux and repositories of GPL software.
>
> It seems that even you are one of the people who forget that on Linux
> you have _only_ the GPLd software while on Solaris you have both, the
> UNIX tools and the other free software for your choice......
>
Are you seriously trying to claim that only GPL software can be used on
a Linux system? or that only GPL software comes with the distro? In
either case, you are completely wrong.
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.2.5 (GNU/Linux)
iD8DBQFByyPOd90bcYOAWPYRAqMlAJ4rO8tLmi53LcLHwj9dZEMQ4mqvhQCdEgKl
ANhKGeXljnC55AZZ9r5ZGAc=
=2qTf
-----END PGP SIGNATURE-----
--
Jim Richardson http://www.eskimo.com/~warlock
You will obey or molten silver will be poured into your ears.
|
|
|
0
|
|
|
|
Reply
|
Jim
|
12/23/2004 8:10:54 PM
|
|
Jim Richardson <warlock@eskimo.com> writes:
> On 23 Dec 2004 00:29:27 GMT,
> Joerg Schilling <js@cs.tu-berlin.de> wrote:
>> In article <Xdiyd.567352$%k.409308@pd7tw2no>,
>> Philip Callan <callanca@shaw.ca> wrote:
[snip]
>>>although Solaris 10 may be a great leap for Sun, it doesn't hold a
>>>candle to the thousands of programs and support Home/Average users will
>>>want, and will get, with Linux and repositories of GPL software.
>>
>> It seems that even you are one of the people who forget that on Linux
>> you have _only_ the GPLd software while on Solaris you have both, the
>> UNIX tools and the other free software for your choice......
>>
>
> Are you seriously trying to claim that only GPL software can be used on
> a Linux system? or that only GPL software comes with the distro? In
> either case, you are completely wrong.
>
But why would they want to use anything but GPL software if they claim that
the only good licence is GPL? That would make them hypocritical, wouldn't
it and they wouldn't want to be that, right?
For the rest of us, we are free to use both GPL-ed and not-GPL-ed software
....
Dragan
--
Dragan Cvetkovic,
To be or not to be is true. G. Boole No it isn't. L. E. J. Brouwer
!!! Sender/From address is bogus. Use reply-to one !!!
|
|
|
0
|
|
|
|
Reply
|
Dragan
|
12/23/2004 8:17:21 PM
|
|
begin Dragan Cvetkovic wrote:
> Jim Richardson <warlock@eskimo.com> writes:
>
>> On 23 Dec 2004 00:29:27 GMT,
>> Joerg Schilling <js@cs.tu-berlin.de> wrote:
>>> In article <Xdiyd.567352$%k.409308@pd7tw2no>,
>>> Philip Callan <callanca@shaw.ca> wrote:
>
> [snip]
>
>>>>although Solaris 10 may be a great leap for Sun, it doesn't hold a
>>>>candle to the thousands of programs and support Home/Average users will
>>>>want, and will get, with Linux and repositories of GPL software.
>>>
>>> It seems that even you are one of the people who forget that on Linux
>>> you have _only_ the GPLd software while on Solaris you have both, the
>>> UNIX tools and the other free software for your choice......
>>>
>>
>> Are you seriously trying to claim that only GPL software can be used on
>> a Linux system? or that only GPL software comes with the distro? In
>> either case, you are completely wrong.
>>
>
> But why would they want to use anything but GPL software if they claim
> that the only good licence is GPL? That would make them hypocritical,
> wouldn't it and they wouldn't want to be that, right?
>
> For the rest of us, we are free to use both GPL-ed and not-GPL-ed software
> ...
>
> Dragan
>
>
I simply can't believe this show of utter stupidity. And these are the
people who try to hook us to Solaris? They are even dumber than wintrolls,
and those rival retarded bricks
--
What happens if a big asteroid hits Earth? Judging from realistic
simulations involving a sledge hammer and a common laboratory frog,
we can assume it will be pretty bad. --- Dave Barry
|
|
|
0
|
|
|
|
Reply
|
Peter
|
12/23/2004 8:49:15 PM
|
|
Peter K�hlmann <Peter.Koehlmann@t-online.de> writes:
> begin Dragan Cvetkovic wrote:
>> For the rest of us, we are free to use both GPL-ed and not-GPL-ed software
>> ...
>
> I simply can't believe this show of utter stupidity. And these are the
> people who try to hook us to Solaris? They are even dumber than wintrolls,
> and those rival retarded bricks
I assume this is a slightly longer form of now famous "idiot" phrase.
I don't mind being in the company of Rich and J�rg :-)
Dragan
--
Dragan Cvetkovic,
To be or not to be is true. G. Boole No it isn't. L. E. J. Brouwer
!!! Sender/From address is bogus. Use reply-to one !!!
|
|
|
0
|
|
|
|
Reply
|
Dragan
|
12/23/2004 8:52:00 PM
|
|
-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1
On 23 Dec 2004 17:35:21 GMT,
Joerg Schilling <js@cs.tu-berlin.de> wrote:
> In article <EWCyd.566636$nl.1144@pd7tw3no>,
> Philip Callan <callanca@shaw.ca> wrote:
>
>>And here when I read the changelog, and see changes like this:
>>
>>version 1.14 - Sergey Poznyakoff, 2004-05-11
>>
>>* Added support for POSIX.1-2001 and ustar archive formats.
>>* New option --format allows to select the output archive format
>>* The default output format can be selected at configuration time
>> by presetting the environment variable DEFAULT_ARCHIVE_FORMAT.
>> Allowed values are GNU, V7, OLDGNU and POSIX.
>
> Well, the fact that after 10 years they finally managed to implement first
> draft code for POSIX does not mean that GNU tar uses this format by default.
>
> GNU tar still by default creates archives with a proprietary format.
>
proprietary? you keep saying that word, I do not think it means what you
think.
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.2.5 (GNU/Linux)
iD8DBQFByyfzd90bcYOAWPYRAqicAJ9kI0om4JMjqKTML5EAQWm7DvR/zQCfawQt
XIEzs3wmx4qBe0hqbBc/O0o=
=9lHy
-----END PGP SIGNATURE-----
--
Jim Richardson http://www.eskimo.com/~warlock
Slow day. Practice crawling.
|
|
|
0
|
|
|
|
Reply
|
Jim
|
12/23/2004 9:10:52 PM
|
|
-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1
On Thu, 23 Dec 2004 15:17:21 -0500,
Dragan Cvetkovic <me@privacy.net> wrote:
> Jim Richardson <warlock@eskimo.com> writes:
>
>> On 23 Dec 2004 00:29:27 GMT,
>> Joerg Schilling <js@cs.tu-berlin.de> wrote:
>>> In article <Xdiyd.567352$%k.409308@pd7tw2no>,
>>> Philip Callan <callanca@shaw.ca> wrote:
>
> [snip]
>
>>>>although Solaris 10 may be a great leap for Sun, it doesn't hold a
>>>>candle to the thousands of programs and support Home/Average users will
>>>>want, and will get, with Linux and repositories of GPL software.
>>>
>>> It seems that even you are one of the people who forget that on Linux
>>> you have _only_ the GPLd software while on Solaris you have both, the
>>> UNIX tools and the other free software for your choice......
>>>
>>
>> Are you seriously trying to claim that only GPL software can be used on
>> a Linux system? or that only GPL software comes with the distro? In
>> either case, you are completely wrong.
>>
>
> But why would they want to use anything but GPL software if they claim that
> the only good licence is GPL? That would make them hypocritical, wouldn't
> it and they wouldn't want to be that, right?
>
> For the rest of us, we are free to use both GPL-ed and not-GPL-ed software
> ...
I was referring to the apparant claim by Mr Schilling, re: that
"on Linux you have _only_ the GPLd software" Which is of course, simply
false.
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.2.5 (GNU/Linux)
iD8DBQFByz20d90bcYOAWPYRAjDIAKDrNxtmC+1JL2ncqBrTo9b1CZ/S3gCeM/OJ
3FIvX+5D4AXOzs1yo84uJdQ=
=8KZK
-----END PGP SIGNATURE-----
--
Jim Richardson http://www.eskimo.com/~warlock
Q: What does it say on the bottom of Coke cans in North Dakota?
A: Open other end.
|
|
|
0
|
|
|
|
Reply
|
Jim
|
12/23/2004 10:10:54 PM
|
|
On Thu, 23 Dec 2004 15:17:21 -0500, Dragan Cvetkovic wrote:
> Jim Richardson <warlock@eskimo.com> writes:
>
>> On 23 Dec 2004 00:29:27 GMT,
>> Joerg Schilling <js@cs.tu-berlin.de> wrote:
>>> In article <Xdiyd.567352$%k.409308@pd7tw2no>,
>>> Philip Callan <callanca@shaw.ca> wrote:
>
> [snip]
>
>>>>although Solaris 10 may be a great leap for Sun, it doesn't hold a
>>>>candle to the thousands of programs and support Home/Average users will
>>>>want, and will get, with Linux and repositories of GPL software.
>>>
>>> It seems that even you are one of the people who forget that on Linux
>>> you have _only_ the GPLd software while on Solaris you have both, the
>>> UNIX tools and the other free software for your choice......
>>>
>>
>> Are you seriously trying to claim that only GPL software can be used on
>> a Linux system? or that only GPL software comes with the distro? In
>> either case, you are completely wrong.
>>
>
> But why would they want to use anything but GPL software if they claim that
> the only good licence is GPL? That would make them hypocritical, wouldn't
> it and they wouldn't want to be that, right?
>
> For the rest of us, we are free to use both GPL-ed and not-GPL-ed software
We prefer GPL and GPL compatible licenses yes, especially at the OS level.
Licenses that do not mix well and can screw us over we do not want mixed
with our OS thank you very much. And yes, we do *prefer* our other
software to be Open Source, preferably GPL or a license friendly to the
GPL. However, for non-systems software we are quite willing to accept
(with the exceptions of the few true fanatics) and even pay for high
quality closed source software. And fortunately, some very nice, high
quality commercial software does exist for Linux, some of the very best
software out there in fact. So don't worry your pretty little slowlaris
heads about being able to use CSS on Linux. Or even other Open Source
Licenses.
|
|
|
0
|
|
|
|
Reply
|
Liam
|
12/23/2004 10:25:29 PM
|
|
In article <lm3bxwbwha.fsf@privacy.net> (Thu, 23 Dec 2004 15:17:21 -0500),
Dragan Cvetkovic wrote:
> But why would they want to use anything but GPL software if they claim that
> the only good licence is GPL?
Nobody made that claim, besides yourself.
> That would make them hypocritical, wouldn't
> it and they wouldn't want to be that, right?
That would make you hypocritical, having made a false claim and then used
it to justify your baseless attack on others.
> For the rest of us, we are free to use both GPL-ed and not-GPL-ed software
Sure you are.
--
"But mark this: all the silly names by all the silly Linux morons
in the world won't stop Longhorn from being a HUGE commercial and
technical success for MS."
-- DFS
|
|
|
0
|
|
|
|
Reply
|
Hamilcar
|
12/23/2004 10:48:49 PM
|
|
Here in comp.unix.solaris,
js@cs.tu-berlin.de (Joerg Schilling) spake unto us, saying:
>Richard Steiner <rsteiner@visi.com> wrote:
>>js@cs.tu-berlin.de (Joerg Schilling) spake unto us, saying:
>>>Richard Steiner <rsteiner@visi.com> wrote:
>>>
>>>Do you _really_ like to make the proposal that the whole FreeBSD kernel
>>>should be relicensed under GPL?
>>
>>Why not? Release it under both licenses. That way, GPL'd code can be
>>added legitimately to the public version of the FreeBSD kernel, but the
>>BSDL portions can still be used in proprietary ventures.
>
>You cannot because you cannot do things that affect the rights of a
>third party.
Could you elaborate?
--
-Rich Steiner >>>---> http://www.visi.com/~rsteiner >>>---> Smyrna, GA USA
OS/2 + eCS + Linux + Win95 + DOS + PC/GEOS + Executor = PC Hobbyist Heaven!
WARNING: I've seen FIELDATA FORTRAN V and I know how to use it!
The Theorem Theorem: If If, Then Then.
|
|
|
0
|
|
|
|
Reply
|
rsteiner
|
12/24/2004 6:21:09 AM
|
|
Chris Mattern wrote:
> Joerg Schilling wrote:
>
>
>>The GNU tar on your Linux system (even is you use the latest) by default
>>still does not create a TAR archive but a proprietary archive format
>>that cannot be read correctly by a UNIX tar program.
>>
>
> Joerg, only you could call an open format whose details are fully
> available and that may be implemented freely by anyone as
> "proprietary".
it's still proprietary as it's not widely accepted solution. it's not
closed as everyone can read implementation details from code (it's
another issue if reader can understand the code or if the code is meant
to be read...).
toomas
--
Perfect day for scrubbing the floor and other exciting things.
|
|
|
0
|
|
|
|
Reply
|
Toomas
|
12/24/2004 7:29:11 AM
|
|
Hamilcar Barca wrote:
> Of course they won't. Without your "sufficient knowledge", no operating
> system development will occur anywhere.
You know what??? When YOU have just 10% of what Joerg knows, you can
consider yourself to be SOMEONE.
I'm so sick of rabid Linux dogs. Fortunately, rabid dogs die from
muscle cramps.
|
|
|
0
|
|
|
|
Reply
|
UNIX
|
12/24/2004 6:21:35 PM
|
|
Philip Callan wrote:
> systems, but unlike you I haven't coded for them, I've /used/ them, and
> although Solaris 10 may be a great leap for Sun, it doesn't hold a
> candle to the thousands of programs and support Home/Average users will
> want, and will get, with Linux and repositories of GPL software.
Pure and utter Linux bullshit babble. Solaris has 95% of Linux
applications ported to it, and the 5% that aren't either will be ported
in the future, or they're just Linux crap code written by PC-bucket kids
who have never touched or even seen real UNIX hardware.
> Because SCSI drives were exorbitantly priced, as the IDE/ATAPI drives
> came on the market, PC/Linux users started buying them a lot more, and
> there is no doubt in my mind that there are /far/ more PC users who burn
> DVD's than there are Sun users.
It's *NOT* about which platform you burn on, it's about which SW you use
to do it. Nero Burning ROM is a piece of shit compared to the cdrtools.
In fact, if I do burn media on Windows, I end up using cdrtools
anyway. Gee, I wonder why.
> Just in case that wasn't clear, more people own DVD-R/RW drives on their
> x86/x86_64 based PC's than there are Sun users, flat out. I recall
> seeing in one of the online reports that adpotion rate of DVD burners
> was just as fast, and in some countries, faster than CD burners.
Well, I also burn media on my x86/x86_64 PC running Solaris, as well.
So what does that mean as far as your "statistics"?
> Like I said before, all that matters to you is 'open' not 'free'
And what exactly is wrong with open vs. free? Anybody with half a brain
can tell you that open is way more important than free.
I see nothing wrong with Joerg wanting to make some cash out of his
knowledge/expertise/hard work, and neither should you... unless you're a
parasite.
> So you'll be in good company with the rest of the Sun crowd. None of
> them seem to make the distinction between 'open source' and 'free
> software' either.
Excuse me? Open source DOES NOT mean FREE SOFTWARE. Just because
something is open, that is not the same as free software. Likewise,
open source does not equate to Linux.
|
|
|
0
|
|
|
|
Reply
|
UNIX
|
12/24/2004 6:31:03 PM
|
|
Hamilcar Barca wrote:
> We want "free as in freedom" software. The GPL is the best of the
> current licenses, but X, Apache, Mozilla, and other software is released
> with free licenses. This is in complete contrast to the current state of,
> for example, Java and Solaris.
If you want "free as in freedom", the you want the BSD license, and not
GPL. When will you finally switch your brain on and start thinking for
yourself? Must others always tell you what to think?
|
|
|
0
|
|
|
|
Reply
|
UNIX
|
12/24/2004 6:33:39 PM
|
|
Jose Maria Lopez Hernandez wrote:
> More enterprise features? Cheaper than Redhat?
> Do you know how much it costs Solaris consultory? Or Solaris support?
Last time I checked, it was $993 per year, which I don't think is too
bad, considering what I get in return.
> Why you want their compiler? Everybody uses GCC under Solaris.
Because Sun's (or any other vendor, for that matter) compiler generates
better, faster code.
In fact, pretty much *ANY* compiler generates better code than GCC.
The idea behind GCC may be excellent, but its realization leaves much to
be desired.
|
|
|
0
|
|
|
|
Reply
|
UNIX
|
12/24/2004 6:37:57 PM
|
|
Hamilcar Barca wrote:
> You're stretching a tiny truth extremely thin to make your unimaginative
> point, Joerg. You've obviously failed to notice that the GNU tools are
> the Open Standard.
The GNU tools are *ANYTHING BUT THE STANDARD*. And AS IF that wasn't BAD
ENOUGH, the GNU tools don't follow any conventions or abide by any
standards.
|
|
|
0
|
|
|
|
Reply
|
UNIX
|
12/24/2004 6:40:05 PM
|
|
Philip Callan wrote:
> I /have/ tar and awk, they point to programs that do the exact same
> thing, and from what I can tell, the only difference is the licensing,
> is there a functional difference?
Apparently you do not know MUCH about those tools, because if you did,
you'd be pissed off to no end that they DON'T WORK the way they're
SUPPOSED TO WORK.
|
|
|
0
|
|
|
|
Reply
|
UNIX
|
12/24/2004 6:41:48 PM
|
|
Philip Callan wrote:
> Joerg Schilling wrote:
>
>> The GNU tar on your Linux system (even is you use the latest) by default
>> still does not create a TAR archive but a proprietary archive format
>> that cannot be read correctly by a UNIX tar program.
>>
>
> Oh yeah, and as author of another tool performing the same function your
> unbiased right? Found an older thread of you flogging this horse with PTB:
Joerg may or may not be biased, but what he states is true. It's not
Joerg's, mine, or anybody else's fault that you do not command enough
knowledge to know this.
|
|
|
0
|
|
|
|
Reply
|
UNIX
|
12/24/2004 6:43:33 PM
|
|
Chris Mattern wrote:
> Guh? Are you claiming that the source for gtar is in fact NOT
> available?
No, I believe he is logically claiming that source code isn't the same
as something being documented.
Why are these concepts so hard to understand?
Do we need to start a charity to buy you all UNIX computers so that you
finally will understand?
|
|
|
0
|
|
|
|
Reply
|
UNIX
|
12/24/2004 6:57:07 PM
|
|
Chris Mattern wrote:
>>Standard bodies (POSIX, ANSI, ISO, LSB, ...) are generally a good thing
>>and they encourage inter-operability.
>
>
> As does open source.
Maybe it does, but two wrongs don't make a right. Open source software
goes about it the wrong way.
|
|
|
0
|
|
|
|
Reply
|
UNIX
|
12/24/2004 6:59:02 PM
|
|
Dragan Cvetkovic wrote:
>>I simply can't believe this show of utter stupidity. And these are the
>>people who try to hook us to Solaris? They are even dumber than wintrolls,
>>and those rival retarded bricks
>
>
> I assume this is a slightly longer form of now famous "idiot" phrase.
>
> I don't mind being in the company of Rich and J�rg :-)
I will gladly be an "idiot" if that is what this means.
|
|
|
0
|
|
|
|
Reply
|
UNIX
|
12/24/2004 7:02:18 PM
|
|
UNIX admin wrote:
> Philip Callan wrote:
>
>> systems, but unlike you I haven't coded for them, I've /used/ them,
>> and although Solaris 10 may be a great leap for Sun, it doesn't hold a
>> candle to the thousands of programs and support Home/Average users
>> will want, and will get, with Linux and repositories of GPL software.
>
>
> Pure and utter Linux bullshit babble. Solaris has 95% of Linux
> applications ported to it, and the 5% that aren't either will be ported
> in the future, or they're just Linux crap code written by PC-bucket kids
> who have never touched or even seen real UNIX hardware.
>
Okay, so take OUT the Linux compatibility layer, and any of the GPL
tools they distribute Solaris 10 with, what is it?
Good old Slowaris, different license, with some new features (dtrace,
containers, zfs).
This is central to my point about commercial entities using the works of
others, not just to benefit themselves, but to try to HARM the platform
it was originally written for.
Right now one of Linux's best points over windows is the thousands of
GPL programs that were written to do things you either cannot or are
more difficult in Windows.
See, Microsoft can't take a bunch of the GPL windows software, and
package it together with the Windows CD, and then have any way to
counter the Out of the box productivity a Linux solution offers.
SUN, however, can 'open source' their Solaris, package a shitload of GPL
Linux software and a compatibility layer, and then try to offer a
package with support akin to Redhat, counting on /businesses/ wanting
the OS, and the applications /AND/ the support.
They are not concerned with the x86 home user, they are more concerned
with trying to get a swing back at all the inexpensive x86/Linux
machines that pushed all the old SUN hardware off the racks.
Trying to say 'look you don't need to buy our through the nose expensive
hardware anymore'
And the only reason they have a half-decent x86 offering WILL be their
Linux compatibility layer.
Solaris/x86 is a joke.
>> Because SCSI drives were exorbitantly priced, as the IDE/ATAPI drives
>> came on the market, PC/Linux users started buying them a lot more, and
>> there is no doubt in my mind that there are /far/ more PC users who
>> burn DVD's than there are Sun users.
>
>
> It's *NOT* about which platform you burn on,
It is when I'm making a point counter to Joergs statement 'Solaris users
did write DVD using cdrecord since March 1998 long before Linux users
could afford DVD writers.'
Just because people were doing something FIRST on a platform, doesn't
make them the defacto 'way things should be done' things change, and
Linux users most likely account for the largest slice of cdrecord's
userbase. Majority rules in this case.
> it's about which SW you use
> to do it.
Yeah, in my case, cdrdao and dvdrw-tools now.
> Nero Burning ROM is a piece of shit compared to the cdrtools.
Uhh, yeah, but Nero isn't 'just' a burner, its a ripper, encoder, gui
etc....
For your truly 'average' Windows user, who knows *of* the command line,
but rarely uses it, Nero is a good product, and has a far easier
learning curve for non-cli people.
> In fact, if I do burn media on Windows, I end up using cdrtools
> anyway. Gee, I wonder why.
Because on Windows at least, it's the best solution?
Or because you use it on multiple platforms, and even if it was lacking
in some way, you still like the ability to use the same program and same
comamnds everywhere?
>
>> Just in case that wasn't clear, more people own DVD-R/RW drives on
>> their x86/x86_64 based PC's than there are Sun users, flat out. I
>> recall seeing in one of the online reports that adpotion rate of DVD
>> burners was just as fast, and in some countries, faster than CD burners.
>
>
> Well, I also burn media on my x86/x86_64 PC running Solaris, as well.
> So what does that mean as far as your "statistics"?
Your 1 for the x86/burner crowd, 1 for the 'sun users crowd', 0 for the
Sun/Sparc crowd.
And given that you say 'x86_64' it's a new machine, so /your/ burning
habits aren't relevant to the period Joerg is crowing about in 1998,
when he says 'Solaris users did write DVD using cdrecord since March
1998 long before Linux users could afford DVD writers.'
I cant even recall if an x86 version of Solaris was available in 1998....
>
>> Like I said before, all that matters to you is 'open' not 'free'
>
>
> And what exactly is wrong with open vs. free? Anybody with half a brain
> can tell you that open is way more important than free.
I agree, anyone with a half a brain says Open is more important that
Free, and they do! Jonathan Schwartz comes to mind first....
Those of use who use both halves realize that there is such a thing as
indirect benefits, and that 'Free' software, if it continues the growth
it has, will ensure that the infrastructure my children will use, will
remain Free, and under the control of the people, and not companies and
governments.
> I see nothing wrong with Joerg wanting to make some cash out of his
> knowledge/expertise/hard work, and neither should you... unless you're a
> parasite.
I see nothing wrong with him making money either, but his attitude
towards those developers that only did what the GPL lets them, and
modified his source in order to add functionality, only shows how little
he understands how the free software community is actually meant to work.
It's beginning to look like a serious case of GPL regret...
>
>> So you'll be in good company with the rest of the Sun crowd. None of
>> them seem to make the distinction between 'open source' and 'free
>> software' either.
>
>
> Excuse me? Open source DOES NOT mean FREE SOFTWARE. Just because
> something is open, that is not the same as free software. Likewise,
> open source does not equate to Linux.
I know it doesn't, I mis-typed, it should be 'None of them seem to be
able to make the distinction...' sorry.
And Linux is both OPEN and FREE, but I realize that the OSS community
extends far beyond the linux slice of the pie, for those clients who
don't care to switch to Linux, I give them a copy of the OpenCD, so they
can get a handle on the quality that OSS (most of it GPL) provides.
There is a reason GPL software can increase in quality so quickly,
because of the freedom that is inherent in the license, if a developer
*ahem* decides to NOT add a particular feature users desire, or chooses
to only do so in a commercial manner, OTHER can step in and do what
needs to be done.
|
|
|
0
|
|
|
|
Reply
|
Philip
|
12/24/2004 7:18:17 PM
|
|
UNIX admin wrote:
> Chris Mattern wrote:
>> Guh? Are you claiming that the source for gtar is in fact NOT
>> available?
>
> No, I believe he is logically claiming that source code isn't the same
> as something being documented.
>
> Why are these concepts so hard to understand?
>
> Do we need to start a charity to buy you all UNIX computers so that you
> finally will understand?
I have a Sun Blade 1500 on my desk at work and admin an F15K for my
paycheck. The 1500 is nice, but it's slow. I like Linux a lot, and
use it on my laptop and home PC.
--
Christopher Mattern
"Which one you figure tracked us?"
"The ugly one, sir."
"...Could you be more specific?"
|
|
|
0
|
|
|
|
Reply
|
Chris
|
12/24/2004 7:20:30 PM
|
|
Joerg Schilling wrote:
> The fact that there are still people that move towards Linux does not
> matter, these people just don't know enough about Linux.
This is very true. At my previous employer they were gong-ho about
Linux, while at the same nobody knew enough about it or UNIX in general
to be able to make any educated decisions.
> Unfortumately not to Solaris but to MacOS X. The reason
Well, that's not such a shame. MacOS seems to be a good, well rounded
and finished OS. Even if it does not abide by the UNIX standards (I
don't know if it does or doesn't), unlike Linux, it's consistent in
whatever it does.
And consistency is one of the most important things, somcething that
Linux sorely lacks.
|
|
|
0
|
|
|
|
Reply
|
UNIX
|
12/24/2004 7:21:14 PM
|
|
Hamilcar Barca wrote:
> It's an assumption supported fully by the evidence: Sun's floating of its
> new C[sic]DDL is evidence they intend to claim to support "open source"
> [sic] while retaining full control.
Opem sourcing something doesn't mean you automatically release control
over it. If that were so, then Linux wouldn't qualify as open source
either, as Linus Torvalds controls it [the kernel]. And let's not
forget, in words many a Linux zealot: "Linux is not an OS, it's a kernel!"
|
|
| | | |
Search |
Post Question |
Groups |
Stream |
About |
Register