f



Trolltech QT license question

  Can anyone here that is familiar with licensing QT from Trolltech
give me a moment?

  I've searched through past newsgroup posts and it would seem from all
the discussions that my group would need a commercial license of QT for
our project; due to the fact that we plan a wide distribution of our
proprietary software.  However we are planning to distribute it to our
associates for 'free'.  What I would like to know is the cost of
software to end users an issue in determining whether or not we need to
secure a license for QT?

  Clues would be appreciated.

Thanks

0
barryc
2/14/2006 9:51:28 PM
comp.os.linux.development.apps 5216 articles. 1 followers. Post Follow

273 Replies
778 Views

Similar Articles

[PageSpeed] 57

barryc.ctr@gmail.com writes:

>   Can anyone here that is familiar with licensing QT from Trolltech
> give me a moment?
>
>   I've searched through past newsgroup posts and it would seem from all
> the discussions that my group would need a commercial license of QT for
> our project; due to the fact that we plan a wide distribution of our
> proprietary software.  However we are planning to distribute it to our
> associates for 'free'.  What I would like to know is the cost of
> software to end users an issue in determining whether or not we need to
> secure a license for QT?

Why don't you write to Trolltech and ask them?

-- 
M�ns Rullg�rd
mru@inprovide.com
0
iso
2/14/2006 10:29:30 PM
barryc.ctr@gmail.com wrote:
>   Can anyone here that is familiar with licensing QT from Trolltech
> give me a moment?
> 
>   I've searched through past newsgroup posts and it would seem from all
> the discussions that my group would need a commercial license of QT for
> our project; due to the fact that we plan a wide distribution of our
> proprietary software.  However we are planning to distribute it to our
> associates for 'free'.  What I would like to know is the cost of
> software to end users an issue in determining whether or not we need to
> secure a license for QT?
> 
>   Clues would be appreciated.

 From what I understand, if you do not distribute the source code,
then you must pay for their Qt license.

Why don't you just switch to wxWindows?  :-)

Carlos
--
0
Carlos
2/15/2006 2:29:09 AM
"M�ns Rullg�rd" <mru@inprovide.com> wrote in message 
news:yw1xfymlk20l.fsf@agrajag.inprovide.com...

> Why don't you write to Trolltech and ask them?

    They're not exactly an unbiased party.

    DS


0
David
2/15/2006 3:28:21 AM
M�ns Rullg�rd writes:
> Why don't you write to Trolltech and ask them?

DS writes:
> They're not exactly an unbiased party.

So what?  They're the copyright owner.
-- 
John Hasler 
john@dhh.gt.org
Dancing Horse Hill
Elmwood, WI USA
0
John
2/15/2006 3:54:13 AM
David Schwartz wrote:
> "M�ns Rullg�rd" <mru@inprovide.com> wrote in message 
> news:yw1xfymlk20l.fsf@agrajag.inprovide.com...
> 
>> Why don't you write to Trolltech and ask them?
> 
>     They're not exactly an unbiased party.
They're quite helpful regarding licensing anyway.

The short answer is that unless you make your own
project GPL compatible - which involves making the
source code available to the customers, you need
the Qt commercial license.
0
ISO
2/15/2006 8:01:34 AM
"David Schwartz" <davids@webmaster.com> writes:

> "M�ns Rullg�rd" <mru@inprovide.com> wrote in message 
> news:yw1xfymlk20l.fsf@agrajag.inprovide.com...
>
>> Why don't you write to Trolltech and ask them?
>
>     They're not exactly an unbiased party.

They are the copyright holders, and they probably have better lawyers
should they wish to sue.  That's what matters in reality.

-- 
M�ns Rullg�rd
mru@inprovide.com
0
iso
2/15/2006 8:26:03 AM
"John Hasler" <john@dhh.gt.org> wrote in message 
news:87wtfxwa3e.fsf@toncho.dhh.gt.org...

> M�ns Rullg�rd writes:

>> Why don't you write to Trolltech and ask them?

> DS writes:

>> They're not exactly an unbiased party.

> So what?  They're the copyright owner.

    Exactly.

    Copyright owners will frequently argue that you should purchase a 
commercial license from then even when your use is well within what is 
permitted by their non-commercial (free) license. Asking the copyright owner 
if you need to pay for a license is like asking a barber if you need a 
haircut. He has the most information, and you should listen to his argument, 
but you may very well want a second opinion.

    DS


0
David
2/15/2006 2:59:34 PM
David Schwartz wrote:

>     Copyright owners will frequently argue that you should purchase a
> commercial license from then even when your use is well within what is
> permitted by their non-commercial (free) license. Asking the copyright
> owner if you need to pay for a license is like asking a barber if you need
> a haircut. He has the most information, and you should listen to his
> argument, but you may very well want a second opinion.

But if you plan to do something that they say the license does not allow to
do, is convenient to ask that second opinion from a lawyer, not a
newsgroup.

-- 
Salu2
0
ISO
2/15/2006 3:27:46 PM
"Juli�n Albo" <JULIANALBO@terra.es> wrote in message 
news:43f34871_1@x-privat.org...

> David Schwartz wrote:

>>     Copyright owners will frequently argue that you should purchase a
>> commercial license from then even when your use is well within what is
>> permitted by their non-commercial (free) license. Asking the copyright
>> owner if you need to pay for a license is like asking a barber if you 
>> need
>> a haircut. He has the most information, and you should listen to his
>> argument, but you may very well want a second opinion.

> But if you plan to do something that they say the license does not allow 
> to
> do, is convenient to ask that second opinion from a lawyer, not a
> newsgroup.

    Sure, but you might be able to save yourself some money if everyone 
agrees with them. You might even save yourself some money if you can collect 
arguments why the copyright owner is incorrect. Both of these solutions are 
better than going to a lawyer knowing nothing.

    It might even be dangerous to go to the copyright holder directly 
without going to a lawyer. Saying the wrong thing may box you into needing a 
commercial license when a slightly better presentation might have saved you.

    Similarly, going to your own lawyer with nothing is a good way to lose a 
good chunk of money. Going to a lawyer with a collection of opinions from 
other people (maybe even some lawyers) could save you a good chunk of 
change.

    Going to the lawyer or to the copyright holder could easily be a much 
worse move than going to a newsgroup. At worse, from the newsgroup, you get 
nothing.

    DS


0
David
2/16/2006 4:55:13 AM
Stork replied to:

>I've searched through past newsgroup posts and it would seem from all
>the discussions that my group would need a commercial license of QT for
>our project; due to the fact that we plan a wide distribution of our
>proprietary software.  However we are planning to distribute it to our
>associates for 'free'.

You need to have a commercial license.  The QT license for open source
requires that you use a GPL or GPL equivalent open source license in
order to qualify for free distribution.    You are distributing a
closed source, commercial application, for sale, and therefor, you
should pony up.

Now, for the ones that you give away for your own associates, if they
are employees, you -might- not need a license for them.  But, if your
intent is to give away promotional copies of your software without
opening it up, I would say that you need to have the license.

This is a fairly clear cut case of copyright violation, if you don't.

0
stork
2/17/2006 2:43:10 PM
"stork" <stork@storkyak.com> wrote in message 
news:1140187390.778508.15900@g47g2000cwa.googlegroups.com...

> You need to have a commercial license.  The QT license for open source
> requires that you use a GPL or GPL equivalent open source license in
> order to qualify for free distribution.    You are distributing a
> closed source, commercial application, for sale, and therefor, you
> should pony up.
>
> Now, for the ones that you give away for your own associates, if they
> are employees, you -might- not need a license for them.  But, if your
> intent is to give away promotional copies of your software without
> opening it up, I would say that you need to have the license.
>
> This is a fairly clear cut case of copyright violation, if you don't.

    You are assuming his product will be a derived work of QT. It may be 
possible to arrange his project so that it is not a derived work but a mere 
aggregate, dynamically linking to the QT library without containing any of 
it. In that case, he can use the library under the GPL, distributing only 
the source code to the library.

    Note that this is *NOT* as simple as dynamically linking. For example, 
if your source code includes TrollTech header files, your compiled 
executable may well be a derivative work even if it dynamically links. It 
may take quite a bit of isolation to ensure that your executable does not 
contain any of the covered work.

    DS


0
David
2/17/2006 11:01:35 PM
David Schwartz wrote:
> "stork" <stork@storkyak.com> wrote in message 
> news:1140187390.778508.15900@g47g2000cwa.googlegroups.com...
> 
>> You need to have a commercial license.  The QT license for open source
>> requires that you use a GPL or GPL equivalent open source license in
>> order to qualify for free distribution.    You are distributing a
>> closed source, commercial application, for sale, and therefor, you
>> should pony up.
>>
>> Now, for the ones that you give away for your own associates, if they
>> are employees, you -might- not need a license for them.  But, if your
>> intent is to give away promotional copies of your software without
>> opening it up, I would say that you need to have the license.
>>
>> This is a fairly clear cut case of copyright violation, if you don't.
> 
>     You are assuming his product will be a derived work of QT. It may be 
> possible to arrange his project so that it is not a derived work but a mere 
> aggregate, dynamically linking to the QT library without containing any of 
> it. In that case, he can use the library under the GPL, distributing only 
> the source code to the library.
> 
>     Note that this is *NOT* as simple as dynamically linking. For example, 
> if your source code includes TrollTech header files, your compiled 
> executable may well be a derivative work even if it dynamically links. It 
> may take quite a bit of isolation to ensure that your executable does not 
> contain any of the covered work

It's still an open question wether this will still count as "derived work".
0
UTF
2/17/2006 11:09:47 PM
DS writes:
> It may take quite a bit of isolation...

It may also take quite a bit of work by your copyright lawyer.  A license
might be cheaper.

Or, you could save yourself time and money by licensing the software under
the GPL.

-- 
John Hasler 
john@dhh.gt.org
Dancing Horse Hill
Elmwood, WI USA
0
John
2/17/2006 11:24:40 PM
Stork replied to:

>You are assuming his product will be a derived work of QT. <snip>
> Note that this is *NOT* as simple as dynamically linking <snip>

Quite so.  But... there's some discussion on the GNU site that such
inclusion of a library is in fact a derived work.  If you take that
interpretation of the GPL, then, obviously, you must be open source.

Even so, I think that it is safe to assume that setting up an aggregate
as you describe is extremely difficult.   To be non-derived, you would
probably have to have a clean room implementation of Trolltech header
files determined by using the Trolltech binaries.  Simply looking at
the source and copying it over is probably a copywrite violation.

0
stork
2/18/2006 2:20:13 AM
Stork replied to:

> Copyright owners will frequently argue that you should purchase a
> commercial license from then even when your use is well within what is
> permitted by their non-commercial (free) license

That's pretty silly.  There is no need for them to argue anything about
their own licensing.  If they want you to purchase a commercial license
for a situation, they will simply arrange their license so that you
must do so.

0
stork
2/18/2006 2:22:41 AM
stork writes:
> To be non-derived, you would probably have to have a clean room
> implementation of Trolltech header files determined by using the
> Trolltech binaries.

Header files that contain only interface specifications are probably not
protected work at all in the US.  I doubt that the OP wants to be a test
case, though.
-- 
John Hasler 
john@dhh.gt.org
Dancing Horse Hill
Elmwood, WI USA
0
John
2/18/2006 3:40:58 AM
""Nils O. Sel�sdal"" <noselasd@asgaard.removethis.homelinux.org> wrote in 
message news:43f657bb$1@news.wineasy.se...

> It's still an open question wether this will still count as "derived 
> work".

    No, that's a closed question. It's not.

    If it was, you'd get the ridiculous conclusion that every program that 
uses the C standard I/O library is a derived work of every implementation of 
the C standard I/O library, even if the library was made later.

    DS


0
David
2/18/2006 4:01:38 AM
"stork" <stork@storkyak.com> wrote in message 
news:1140229213.049830.4690@g44g2000cwa.googlegroups.com...

> Quite so.  But... there's some discussion on the GNU site that such
> inclusion of a library is in fact a derived work.  If you take that
> interpretation of the GPL, then, obviously, you must be open source.

    That's obviously false. If that were so, executables could be derived 
works from libraries even if the library was made after the executable and 
they contained no common code at all. That's obviously nonsense.

    DS


0
David
2/18/2006 4:02:31 AM
"stork" <stork@storkyak.com> wrote in message 
news:1140229361.539788.291500@g47g2000cwa.googlegroups.com...

> Stork replied to:

>> Copyright owners will frequently argue that you should purchase a
>> commercial license from then even when your use is well within what is
>> permitted by their non-commercial (free) license

> That's pretty silly.  There is no need for them to argue anything about
> their own licensing.  If they want you to purchase a commercial license
> for a situation, they will simply arrange their license so that you
> must do so.

    That has tradeoffs that may be unacceptable. For example, dual licensing 
under GPL has benefits in that you will get a lot of free publicity and 
freely-contributed improvements to your product. People may start off in 
development using your product because it's free, easy to get ahold of, and 
they may already be familiar with it. Then you hope that they don't go 
through the effort of changing to another product even if it's slightly 
cheaper than your commercial license. This is not incompatible with a 
business policy of encouraging people to get the commercial license even if 
they don't really need it. I have personally experienced this from several 
well-known companies that provide libraries or applications in both GPL and 
commercial forms.

    DS


0
David
2/18/2006 4:05:20 AM
"John Hasler" <john@dhh.gt.org> wrote in message 
news:87mzgpe3lh.fsf@toncho.dhh.gt.org...

> stork writes:

>> To be non-derived, you would probably have to have a clean room
>> implementation of Trolltech header files determined by using the
>> Trolltech binaries.

> Header files that contain only interface specifications are probably not
> protected work at all in the US.  I doubt that the OP wants to be a test
> case, though.

    In fact, it should be pretty clear that if you take only as much as you 
need to work with the library, you should be okay. Copyright protects 
against cases where there is more than one way to do the same thing. If 
there is only one way to make your code work with their library, then there 
is no creativity involved in choosing it.

    So you can look at their header files, change as much as possible but 
still keep it working, and you should be in the clear as far as copyright 
goes. If you keep only what you need for compatability, you had no choice, 
and copyright only applies to cases where you have many choices and use 
creativity to pick one.

    IANAL, but I familiar with intellectual property law.

    DS


0
David
2/18/2006 7:43:20 AM
Stork replied to:

>So you can look at their header files, change as much as possible but
>still keep it working, and you should be in the clear as far as copyright
>goes

That's totally not correct.  You are infringing.  You do have a choice.
 You can choose to use another product and another vendor whose
licensing is compatible with your desires, or you can write your own
stuff.  You -absolutely- do not have the right to take someone else's
product willy nilly and modify outside of the terms to which it is
licensed for you.  That is probably the one thing under copywrite law
that both Bill Gates and Richard Stallman would agree on.

0
stork
2/18/2006 5:10:25 PM
David Schwartz wrote:
> "stork" <stork@storkyak.com> wrote in message 
> news:1140187390.778508.15900@g47g2000cwa.googlegroups.com...
> 
> 
>>You need to have a commercial license.  The QT license for open source
>>requires that you use a GPL or GPL equivalent open source license in
>>order to qualify for free distribution.    You are distributing a
>>closed source, commercial application, for sale, and therefor, you
>>should pony up.
>>
>>Now, for the ones that you give away for your own associates, if they
>>are employees, you -might- not need a license for them.  But, if your
>>intent is to give away promotional copies of your software without
>>opening it up, I would say that you need to have the license.
>>
>>This is a fairly clear cut case of copyright violation, if you don't.
> 
> 
>     You are assuming his product will be a derived work of QT. 

I don't think Derived work, Intellectual property, or copyright have
any role in here.

 From what I understand, Trolltech requires you to purchase the
right to *use* Qt to create your application, unless you are
going to create an *open source* application -- from what I
understand, the fact that your application is "gratis" has
absolutely nothing to do with their conditions.

If your application is closed-source and uses Qt, then you must
purchase from them a license.

Copy-n-pasted from Trolltech's website:

" If you would like to write and publish freely redistributable source 
code, you can use the Qt Open Source Edition, now available across all 
supported platforms. If you do not want to open source your work, you 
should use the supported evaluation version instead.

Code written using the Open Source Editions must be available to all of 
the program's users. Users must also be free to modify the source and 
redistribute it internally and externally free of charge (according to 
the GNU GPL license). "

The above refers to "code written using" -- not "code that is a
derived work".

I think it is pretty clear to me (that one should choose wxWindows
instead  ;-)).

Carlos
--
0
Carlos
2/18/2006 8:41:45 PM
Carlos Moreno <moreno_at_mochima_dot_com@mailinator.com> writes:

> David Schwartz wrote:
>> "stork" <stork@storkyak.com> wrote in message
>> news:1140187390.778508.15900@g47g2000cwa.googlegroups.com...
>>
>>>You need to have a commercial license.  The QT license for open source
>>>requires that you use a GPL or GPL equivalent open source license in
>>>order to qualify for free distribution.    You are distributing a
>>>closed source, commercial application, for sale, and therefor, you
>>>should pony up.
>>>
>>>Now, for the ones that you give away for your own associates, if they
>>>are employees, you -might- not need a license for them.  But, if your
>>>intent is to give away promotional copies of your software without
>>>opening it up, I would say that you need to have the license.
>>>
>>>This is a fairly clear cut case of copyright violation, if you don't.
>>     You are assuming his product will be a derived work of QT.
>
> I don't think Derived work, Intellectual property, or copyright have
> any role in here.
>
>  From what I understand, Trolltech requires you to purchase the
> right to *use* Qt to create your application, unless you are
> going to create an *open source* application -- from what I
> understand, the fact that your application is "gratis" has
> absolutely nothing to do with their conditions.

I thought QT was available under the GPL, in which case it the letter
of the GPL is what counts, nothing else.  The GPL is all about derived
works and copyright.

-- 
M�ns Rullg�rd
mru@inprovide.com
0
iso
2/18/2006 9:22:29 PM
M�ns Rullg�rd writes:
> I thought QT was available under the GPL...

It is now.  It originally was under a special license that said that you
could use it only in GPL software but placed restrictions on what you could
do with the software.  Since it imposed additional restrictions this meant
that it could not be used with GPL software without permission of all the
authors of that software.

> ...in which case it the letter of the GPL is what counts, nothing else.

Right.  In practice, though, unless your planned use is unquestionably well
with in the boundaries of the GPL you would be well advised to check with
the copyright owner.  If he says no then you might want to talk to your
lawyer before proceeding.
-- 
John Hasler 
john@dhh.gt.org
Dancing Horse Hill
Elmwood, WI USA
0
John
2/18/2006 10:26:42 PM
M�ns Rullg�rd wrote:
> Carlos Moreno <moreno_at_mochima_dot_com@mailinator.com> writes:
> 
> 
>>David Schwartz wrote:
>>
>>>"stork" <stork@storkyak.com> wrote in message
>>>news:1140187390.778508.15900@g47g2000cwa.googlegroups.com...
>>>
>>>
>>>>You need to have a commercial license.  The QT license for open source
>>>>requires that you use a GPL or GPL equivalent open source license in
>>>>order to qualify for free distribution.    You are distributing a
>>>>closed source, commercial application, for sale, and therefor, you
>>>>should pony up.
>>>>
>>>>Now, for the ones that you give away for your own associates, if they
>>>>are employees, you -might- not need a license for them.  But, if your
>>>>intent is to give away promotional copies of your software without
>>>>opening it up, I would say that you need to have the license.
>>>>
>>>>This is a fairly clear cut case of copyright violation, if you don't.
>>>
>>>    You are assuming his product will be a derived work of QT.
>>
>>I don't think Derived work, Intellectual property, or copyright have
>>any role in here.
>>
>> From what I understand, Trolltech requires you to purchase the
>>right to *use* Qt to create your application, unless you are
>>going to create an *open source* application -- from what I
>>understand, the fact that your application is "gratis" has
>>absolutely nothing to do with their conditions.
> 
> 
> I thought QT was available under the GPL, in which case it the letter
> of the GPL is what counts, nothing else.  The GPL is all about derived
> works and copyright.

My understanding of the issue is that the GPL says what you must
do if you create a "derived work", and it does not say anything
otherwise.  Qt's licensing scheme *does say* something for the
otherwise case -- if you are going to use Qt to create an
application that you will distribute under the terms of the GPL,
then you can use Qt "GPL edition";  *in any other case* you must
purchase Qt "supported edition" -- that's the difference; if you
create an application that links to some GPL'd library, the GPL
does not impose any constraints on you;  Qt license does (again,
the way I understand it).

And again, it has absolutely nothing to do with the fact that
you sell your application or not -- it has to do exclusively
with the fact that you distribute the source code under the
GPL or not.

Carlos
--
0
Carlos
2/18/2006 11:21:42 PM
"stork" <stork@storkyak.com> wrote in message 
news:1140282625.102031.244670@g43g2000cwa.googlegroups.com...

> Stork replied to:

>>So you can look at their header files, change as much as possible but
>>still keep it working, and you should be in the clear as far as copyright
>>goes

> That's totally not correct.  You are infringing.  You do have a choice.
> You can choose to use another product and another vendor whose
> licensing is compatible with your desires, or you can write your own
> stuff.

    That argument has never worked in any court of law. See, for example, 
Lexmark v. Smartek. In this case, Smartek "stole" some code from Lexmark 
printer cartridges. Their defense to the copyright portion (the DMCA portion 
is not relevant) was that this was the only code they could use that would 
allow their cartridges to work with Lexmark printers. Your argument is 
essentially that they have no need to work with Lexmark printers and could 
use some other printer if they wanted to. This argument didn't impress the 
court.

> You -absolutely- do not have the right to take someone else's
> product willy nilly and modify outside of the terms to which it is
> licensed for you.  That is probably the one thing under copywrite law
> that both Bill Gates and Richard Stallman would agree on.

    That is true. However, copyright only applies when there's more than one 
way to do something. If there's only one way to use a shared library, that 
one way is not protected by copyright. There was only way to make a printer 
cartridge that would work in a Lexmark printer, so Smartek was permitted to 
"steal" Lexmark's code despite its copyright.

    The license (assuming it's not an EULA or shrink-wrap agreement, which 
the GPL is not) only applies to things that copyright prohibits you from 
doing.

    DS


0
David
2/18/2006 11:38:13 PM
"Carlos Moreno" <moreno_at_mochima_dot_com@mailinator.com> wrote in message 
news:fDLJf.34239$KC4.1172497@wagner.videotron.net...

> The above refers to "code written using" -- not "code that is a
> derived work".

    It doesn't matter. TrollTech can refer to "code written by people who 
have ever heard of TrollTech" and it would have no legal force. The way 
TrollTech gets to tell me what I can and can't do is by copyright and 
copyright only applies to derived works.

    DS


0
David
2/18/2006 11:39:23 PM
"Carlos Moreno" <moreno_at_mochima_dot_com@mailinator.com> wrote in message 
news:bZNJf.64759$pf1.1206802@weber.videotron.net...

> My understanding of the issue is that the GPL says what you must
> do if you create a "derived work", and it does not say anything
> otherwise.

    Right.

> Qt's licensing scheme *does say* something for the
> otherwise case -- if you are going to use Qt to create an
> application that you will distribute under the terms of the GPL,
> then you can use Qt "GPL edition";  *in any other case* you must
> purchase Qt "supported edition" -- that's the difference; if you
> create an application that links to some GPL'd library, the GPL
> does not impose any constraints on you;  Qt license does (again,
> the way I understand it).

    If this were true, then the code would not be available under the GPL. 
This would be a "GPL with additional restrictions" which is not compatible 
with the GPL.

> And again, it has absolutely nothing to do with the fact that
> you sell your application or not -- it has to do exclusively
> with the fact that you distribute the source code under the
> GPL or not.

    The GPL does not always require you to distribute your source code. 
There is no such thing as "available under the GPL if and only if you 
distribute your source code even if the GPL doesn't require you to", this 
would be "GPL with additional restrictions" which is incompatible with the 
GPL.

    If code is available under the GPL, then you may use and modify it as 
permitted by the GPL and the only restrictions you must comply with are 
those in the GPL. A commercial license can only *add* rights that you don't 
have under the GPL.

    DS


0
David
2/18/2006 11:42:53 PM
    Sorry to reply to myself, but let me make this really clear. Nobody can 
say, "you may use this work under the terms of the GPL if and only if you do 
X" where "X" is a restriction not found in the GPL itself.

    Even if they could, they couldn't. For example, suppose FooTech 
distributes a product FooMaster, under the GPL, however you must also agree 
not to sell your product commercially. If I get a copy of FooMaster and add 
some features to it, I can then distribute my product under the terms of the 
GPL. Anyone who got my copy of my modified FooMaster would be only subject 
to the terms of the GPL and would not have agreed with FooTech not to sell 
the product commercially.

    So even if you could work in some additional restrictions, you'd have no 
way to impose them on those who received your code from a third party who 
distributed it to them only under the GPL.

    The most egregious example of this type of abuse was a company that made 
their own version of a GPL'd product and would only sell it to people who 
had agreed not to request a copy of the source code!

    So if you get QT from someone other than TrollTech, redistributed under 
the GPL, you are subject only to restrictions in the GPL, period.

    DS


0
David
2/18/2006 11:50:57 PM
Carlos writes:
> Qt's licensing scheme *does say* something for the otherwise case -- if
> you are going to use Qt to create an application that you will distribute
> under the terms of the GPL, then you can use Qt "GPL edition"...

You are overcomplicating it.  TrollTech will license QT to you under the
terms of the GPL: it is included in most Linux distributions.  If the GPL
is not acceptable to you they will license a copy to you under other
terms.  No "licensing scheme" is involved: just a copyright owner
exercising his copyright.  Any copyright owner can do this.
-- 
John Hasler 
john@dhh.gt.org
Dancing Horse Hill
Elmwood, WI USA
0
John
2/19/2006 12:11:34 AM
David Schwartz writes:
> Sorry to reply to myself, but let me make this really clear. Nobody can
> say, "you may use this work under the terms of the GPL if and only if you
> do X" where "X" is a restriction not found in the GPL itself.

Of course they can.  It is perfectly legal and enforceable to distribute
your program with a file named "copyright" which says something like

 Flooble Version .001, a program to foobar the baz.  
 Copyright David Schwartz 2006
 This program is distributed under the terms of the GPL (see COPYING)
 with the following additional restrictions:
 ...
 ...

You probably do want to make one of your "additional restrictions"
something like "If you distribute you must include a copy of this file"
and you should mention this file in every source file.

Of course, Flooble's licensing terms are not GPL compatible and you had
better not include any GPLd code in it without the written permission of
the copyright owner.

> So if you get QT from someone other than TrollTech, redistributed under
> the GPL, you are subject only to restrictions in the GPL, period.

And if you get QT from TrollTech you are also subject only to restrictions
in the GPL, period (unless you negotiated other terms, which I understand
they are happy to do).  QT has been distributed under the unmodified GPL
for years.
-- 
John Hasler 
john@dhh.gt.org
Dancing Horse Hill
Elmwood, WI USA
0
John
2/19/2006 12:27:33 AM
"David Schwartz" <davids@webmaster.com> wrote in message 
news:dt8b4v$grv$1@nntp.webmaster.com...

>> That's totally not correct.  You are infringing.  You do have a choice.
>> You can choose to use another product and another vendor whose
>> licensing is compatible with your desires, or you can write your own
>> stuff.

    Here are the relevant excerpts. Again, my position is that if you take 
from header files only what's necessary to interoperate with the shared 
library, you are not violating copyright. These are all from the Lexmark 
case: 
http://www.eff.org/legal/cases/Lexmark_v_Static_Control/20041026_Ruling.pdf

    The crux of my position is basically that if there is only one way to be 
compatible with the shared library, then there is no creative expression 
involved in choosing that one way. Copyright only protects creative choices 
where there is more than one way to do the same thing.

    The argument that you could use another product or vendor just doesn't 
work. With respect to the header files, using another product or vendor 
would be expressing different ideas or processes. Copyright only applies 
when there is more than one way to express the same idea or effect the same 
process. If there is no expressively different way to express the same idea 
as the header files express, they are not copyrightable.

To the extent compatibility requires that a particular code sequence be 
included in the component device to permit its use, the merger and sc�nes � 
faire doctrines generally preclude the code sequence from obtaining 
copyright protection. See Sega Enters., 977 F.2d at 1524 ("When specific 
instructions, even though previously copyrighted, are the only and essential 
means of accomplishing a given task, their later use by another will not 
amount to infringement.") (quoting National Commission on New Technological 
Uses of Copyrighted Works, Final Report 20 (1979)) (emphasis added); Atari 
Games Corp. v. Nintendo of Am., Inc., Nos. 88-4805 & 89-0027, 1993 WL 
207548, at *1 (N.D. Cal. May 18, 1993) ("Atari III") ("Program code that is 
strictly necessary to achieve current compatibility presents a merger 
problem, almost by definition, and is thus excluded from the scope of any 
copyright.").
....
17 U.S.C. � 107. With respect to computer programs, "fair use doctrine 
preserves public access to the ideas and functional elements embedded in 
copyrighted computer software programs." Sony Computer Entm't, Inc. v. 
Connectix Corp., 203 F.3d 596, 603 (9th Cir. 2000).
....
When a work itself constitutes merely an idea, process or method of 
operation, or when any discernible expression is inseparable from the idea 
itself, or when external factors dictate the form of expression, copyright 
protection does not extend to the work. See Bateman, 79 F.3d at 1546 n.28 
("Compatibility and other functionality challenges to originality . . . are 
applied so as to deny copyright protection to a particular work or portion 
of a work.") (emphasis added); Mason, 967 F.2d at 138 n.5 (rejecting 
argument that merger doctrine applies only to question of infringement and 
noting that "this court has applied the merger doctrine to the question of 
copyrightability").

....

Even aside from Dr. Goldberg's testimony that the Toner Loading Program is 
the most efficient means of calculating toner levels, the alternatives 
suggested by Dr. Maggs do not appear to support the district court's initial 
conclusion that the program is expressive. Dr. Maggs' first and third 
suggestions-that different equations and values or a different means of 
measuring toner level altogether could have been used-do not appear to 
represent alternative means of expressing the ideas or methods of operations 
embodied in the Toner Loading Program; they appear to be different ideas or 
methods of operation altogether. Selection from among competing ideas or 
methods of operation generally does not result in copyright-protectable 
expression. See Bateman, 79 F.3d at 1546 n.29 ("Generally there is more than 
one method of operation or process that can be used to perform a particular 
computer program function. However, methods of operation and processes are 
not copyrightable."); see also Am. Dental Ass'n v. Delta Dental Plans Ass'n, 
126 F.3d 977, 980 (7th Cir. 1997) ("A lamp may be entirely original, but if 
the novel elements are also functional the lamp cannot be copyrighted."). 
Nor would the use of a "lookup table" appear to differ meaningfully from the 
use of other equations directly. Instead of executing a mathematical formula 
on a given input, this program merely "looks up" in a data table the output 
of that same formula for the given input value. Finally, Dr. Maggs' fourth 
suggestion-that the same equations could be reordered-does not appear to 
show originality because such alterations may be too trivial to support a 
finding of creative expression. See M.M. Bus. Forms, 472 F.2d at 1140 
(holding that paraphrasing of words on legal forms does not contain the 
requisite originality for copyright protection).

....

    Note that this case is not perfectly analogous to the case of header 
files and a shared library, however these arguments should apply. Note that 
you can only take what is necessary for your code to work with the shared 
library, no more. You do not have to change function names, variable names, 
function orders, or things like that. They are trivial functional elements. 
It is the creative expression you must change if you can and still 
interoperate.

    DS


0
David
2/19/2006 12:57:26 AM
Stork replied to:

> That's obviously false. If that were so, executables could be derived
>works from libraries even if the library was made after the executable and
>they contained no common code at all. That's obviously nonsense.

Wrongo.

"You may not copy, modify, sublicense, or distribute the Program
except as expressly provided under this License.  Any attempt
otherwise to copy, modify, sublicense or distribute the Program is
void, and will automatically terminate your rights under this License."

"This General Public License does not permit incorporating your program
into
proprietary programs.  If your program is a subroutine library, you may
consider it more useful to permit linking proprietary applications with
the
library.  If this is what you want to do, use the GNU Library General
Public License instead of this License."

The license is extremely clear. Once again, if you want to distribute a
proprietary application, contact the IP holder and pay them.  In your
case, pony up for Trolltech.

0
stork
2/19/2006 4:35:42 AM
"stork" <stork@storkyak.com> writes:

> Stork replied to:
>
>>That's obviously false. If that were so, executables could be
>>derived works from libraries even if the library was made after the
>>executable and they contained no common code at all. That's
>>obviously nonsense.
>
> Wrongo.
>
> "You may not copy, modify, sublicense, or distribute the Program
> except as expressly provided under this License.  Any attempt
> otherwise to copy, modify, sublicense or distribute the Program is
> void, and will automatically terminate your rights under this License."
>
> "This General Public License does not permit incorporating your
> program into proprietary programs.  If your program is a subroutine
> library, you may consider it more useful to permit linking
> proprietary applications with the library.  If this is what you want
> to do, use the GNU Library General Public License instead of this
> License."

I'm afraid you're the one who's wrong.  In all dictionaries I've ever
seen, "incorporate" is defined similarly to this (from m-w.com):

  1 a : to unite or work into something already existent so as to form
  an indistinguishable whole b : to blend or combine thoroughly
  2 a : to form into a legal corporation b : to admit to membership in
  a corporate body
  3 : to give material form to : EMBODY

It should be clear that definition 1 is the one that applies here.  On
my system dynamic linking does in no way result in an "indistinguishable
whole" containing the application and the library.  The entire point
with dynamic linking is *not* to do this.

> The license is extremely clear.

I'd say it's extremely fuzzy.

-- 
M�ns Rullg�rd
mru@inprovide.com
0
iso
2/19/2006 11:30:15 AM
Stork replied to:

>On my system dynamic linking does in no way result in an "indistinguishable
>whole" containing the application and the library

My point is that the Free Software Foundation has taken the stance that
dynamic linking is incorporation, and I would assume that Trolltech
would as well.  Both of them, and many other commercial open source
vendors, believe that if you are using, consuming, incorporating, ie:
benefiting from their product, for a closed source commercial purpose,
then you need to pursue a licensing option outside of the GPL because
you are violating it.  Even Microsoft is beginning to make this
argument with their tool suite.  I believe they now have clauses that
explicitly prohibit you from making GPL software with their
redistributables.

The argument you are making is that you don't really have to follow a
license.  There's no court in the USA, especially given the current
administration's appointees, that is going to side with a consumer
playing licensing games with a vendor's product.  That is, if Trolltech
sues you for copyright infringement, you are going to go to court
system stuffed with judges from 12 years of Reagan Bush I, and now 5
years of Bush II.  Good luck.

As a practical matter, what is so hard about honoring the spirit of
what Trolltech is trying to do?  It's really simple.  If you are going
to write free software, then Qt is free.  If you aren't, then you have
to pay them.  If you are benefiting from Qt in a commercial product,
then you should pay Trolltech, and, if you are not willing to pay them,
then, use another product.

You can argue the legalisms as much as someone might try and throw up a
bunch of made a reasons they don't have to pay taxes, but, you would
still be trying to rip off Trolltech.

0
stork
2/19/2006 1:01:59 PM
Stork replied to:

>Here are the relevant excerpts. Again, my position is that if you take
>from header files only what's necessary to interoperate with the shared
>library, you are not violating copyright. These are all from the Lexmark
>case:

That's a heck of a read.  Damn, if there was a subsequent ruling that
said header files or software interfaces in general were not
copywritable, then, yes, you would be right.   Suddenly both the GPL
and Microsoft, which depend on tying, would be stopped dead in their
respective tracks.

The thing about this ruling is that it suggests interfaces aren't
copywritable, but, it doesn't make the slam dunk ruling.  But, to think
that the Supreme Court refused to hear the case, as much as goes, this
is this is settled law.   I guess if you've got the money for the
lawyers, copy away.  But, if you've got the legal money, why not pay
Trolltech for their work?

0
stork
2/19/2006 1:25:53 PM
Stork replied to:

>There's no court in the USA, especially given the current
>administration's appointees, that is going to side with a consumer
>playing licensing games with a vendor's product

Had to reply to myself to point out that the 6th circuit did in fact
quash Lexmark on a very similar issue, and the Supreme Court refused to
hear the appeal on a technicality.

0
stork
2/19/2006 1:42:04 PM
Stork writes:
> Stork replied to:

Why can't you tell us who you are responding to?
-- 
John Hasler 
john@dhh.gt.org
Dancing Horse Hill
Elmwood, WI USA
0
John
2/19/2006 1:51:40 PM
"stork" <stork@storkyak.com> writes:

> Stork replied to:
>
>>On my system dynamic linking does in no way result in an "indistinguishable
>>whole" containing the application and the library
>
> My point is that the Free Software Foundation has taken the stance that
> dynamic linking is incorporation, and I would assume that Trolltech
> would as well.  Both of them, and many other commercial open source
> vendors, believe that if you are using, consuming, incorporating, ie:

Belief is dangerous.  Many people seem to believe that drawing a
picture of a guy with a beard is a deadly sin.  Does that make it
true?  Not in my part of the world.

> benefiting from their product, for a closed source commercial purpose,
> then you need to pursue a licensing option outside of the GPL because
> you are violating it.  Even Microsoft is beginning to make this
> argument with their tool suite.  I believe they now have clauses that
> explicitly prohibit you from making GPL software with their
> redistributables.
>
> The argument you are making is that you don't really have to follow a
> license.

You have to follow the letter of the license, not what the licensor
says he believes it means.  If the licensor made a mistake and granted
you rights he didn't intend, his problem.

> There's no court in the USA, especially given the current
> administration's appointees, that is going to side with a consumer
> playing licensing games with a vendor's product.  That is, if
> Trolltech sues you for copyright infringement, you are going to go
> to court system stuffed with judges from 12 years of Reagan Bush I,
> and now 5 years of Bush II.  Good luck.

That's the reason I suggested to the OP that he ask Trolltech about
it.

> As a practical matter, what is so hard about honoring the spirit of
> what Trolltech is trying to do?  It's really simple.  If you are going
> to write free software, then Qt is free.  If you aren't, then you have
> to pay them.  If you are benefiting from Qt in a commercial product,
> then you should pay Trolltech, and, if you are not willing to pay them,
> then, use another product.

Trolltech, and everyone else, should make their licenses say exactly
what they intend, not use something that sounds about right, and later
say "well, but we really meant..." when someone takes advantage of the
mistake.

-- 
M�ns Rullg�rd
mru@inprovide.com
0
iso
2/19/2006 2:33:05 PM
Stork replied to:

>Trolltech, and everyone else, should make their licenses say exactly
>what they intend, not use something that sounds about right, and later
>say "well, but we really meant..." when someone takes advantage of the
>mistake.

If I wanted all this lawyering around, well, I'd program for Windows.
The choice is to act honorably or not.

0
stork
2/19/2006 4:14:01 PM
M�ns Rullg�rd writes:
> Trolltech ... should ... not use something that sounds about right, and
> later say "well, but we really meant..."

Do you have some evidence that they are doing so?
-- 
John Hasler 
john@dhh.gt.org
Dancing Horse Hill
Elmwood, WI USA
0
John
2/19/2006 4:35:58 PM
"stork" <stork@storkyak.com> writes:

> Stork replied to:
>
>>Trolltech, and everyone else, should make their licenses say exactly
>>what they intend, not use something that sounds about right, and later
>>say "well, but we really meant..." when someone takes advantage of the
>>mistake.
>
> If I wanted all this lawyering around, well, I'd program for Windows.
> The choice is to act honorably or not.

Do you propose that we act "honorably" with respect to SCO too?

-- 
M�ns Rullg�rd
mru@inprovide.com
0
iso
2/19/2006 4:59:52 PM
On 2006-02-19, M�ns Rullg�rd wrote:
> "stork" <stork@storkyak.com> writes:
>> Stork replied to:
>>
>>>Trolltech, and everyone else, should make their licenses say exactly
>>>what they intend, not use something that sounds about right, and later
>>>say "well, but we really meant..." when someone takes advantage of the
>>>mistake.
>>
>> If I wanted all this lawyering around, well, I'd program for Windows.
>> The choice is to act honorably or not.
>
> Do you propose that we act "honorably" with respect to SCO too?

   Of course!

   If you don't, you are no better than they are.

-- 
   Chris F.A. Johnson, author   |    <http://cfaj.freeshell.org>
   Shell Scripting Recipes:     |  My code in this post, if any,
   A Problem-Solution Approach  |          is released under the
   2005, Apress                 |     GNU General Public Licence
0
Chris
2/19/2006 5:09:44 PM
John Hasler <john@dhh.gt.org> writes:

> M�ns Rullg�rd writes:
>> Trolltech ... should ... not use something that sounds about right, and
>> later say "well, but we really meant..."
>
> Do you have some evidence that they are doing so?

I don't know what intentions Trolltech had when they chose the GPL;
they just happen to be serving as an example in this thread.  Evidence
of the FSF are doing exactly this is everywhere, most notably in the
GPL FAQ.

-- 
M�ns Rullg�rd
mru@inprovide.com
0
iso
2/19/2006 5:58:39 PM
"John Hasler" <john@dhh.gt.org> wrote in message 
news:87pslkchvu.fsf@toncho.dhh.gt.org...

> David Schwartz writes:

>> Sorry to reply to myself, but let me make this really clear. Nobody can
>> say, "you may use this work under the terms of the GPL if and only if you
>> do X" where "X" is a restriction not found in the GPL itself.

> Of course they can.  It is perfectly legal and enforceable to distribute
> your program with a file named "copyright" which says something like
>
> Flooble Version .001, a program to foobar the baz.
> Copyright David Schwartz 2006
> This program is distributed under the terms of the GPL (see COPYING)
> with the following additional restrictions:

    The GPL specifically says there can be no additional restrictions, so 
you would have two licensing terms that directly contradicted each other. 
(See, for example, section 6)

    Even if you could, such a product would not actually be licensed under 
the GPL because you could not distribute it under the GPL's terms. It could 
not be combined with other GPL'd software or used in GPL'd projects.

    Such a thing would *NEVER* be considered to be GPL'd.

    DS


0
David
2/19/2006 7:05:25 PM
"stork" <stork@storkyak.com> wrote in message 
news:1140323742.181657.167020@g43g2000cwa.googlegroups.com...

> Stork replied to:
>
>> That's obviously false. If that were so, executables could be derived
>>works from libraries even if the library was made after the executable and
>>they contained no common code at all. That's obviously nonsense.
>
> Wrongo.
>
> "You may not copy, modify, sublicense, or distribute the Program
> except as expressly provided under this License.  Any attempt
> otherwise to copy, modify, sublicense or distribute the Program is
> void, and will automatically terminate your rights under this License."
>
> "This General Public License does not permit incorporating your program
> into
> proprietary programs.  If your program is a subroutine library, you may
> consider it more useful to permit linking proprietary applications with
> the
> library.  If this is what you want to do, use the GNU Library General
> Public License instead of this License."
>
> The license is extremely clear. Once again, if you want to distribute a
> proprietary application, contact the IP holder and pay them.  In your
> case, pony up for Trolltech.

    I already addressed this argument. It's 100% totally wrong. The license 
can say "if you even thing about this program, you must distribute every 
program you ever write under the GPL" and it would have no legal force. The 
license cannot set its own scope, copyright law does.

    DS


0
David
2/19/2006 7:06:09 PM
"stork" <stork@storkyak.com> wrote in message 
news:1140354119.492985.52880@g14g2000cwa.googlegroups.com...

> As a practical matter, what is so hard about honoring the spirit of
> what Trolltech is trying to do?  It's really simple.  If you are going
> to write free software, then Qt is free.  If you aren't, then you have
> to pay them.  If you are benefiting from Qt in a commercial product,
> then you should pay Trolltech, and, if you are not willing to pay them,
> then, use another product.
>
> You can argue the legalisms as much as someone might try and throw up a
> bunch of made a reasons they don't have to pay taxes, but, you would
> still be trying to rip off Trolltech.

    That's is such total bullshit. If Trolltech is trying to capitalize on 
the popularity of the GPL while refusing to honor its spirit, they are the 
ones you are ripping people off. However, I don't think Trolltech acts this 
way and it's kind of strange to see people arguing it on their behalf. I 
would imagine Trolltech has no problem with you using their software if you 
comply with the actual requirements of the actual GPL.

    And what the hell is up with the "free software" community arguing that 
taking the absolute minimum you need to interoperate isn't fair use even 
when courts seem to think it is?

    DS


0
David
2/19/2006 7:09:01 PM
"stork" <stork@storkyak.com> wrote in message 
news:1140355552.971503.268220@o13g2000cwo.googlegroups.com...

> Stork replied to:

>>Here are the relevant excerpts. Again, my position is that if you take
>>from header files only what's necessary to interoperate with the shared
>>library, you are not violating copyright. These are all from the Lexmark
>>case:

> That's a heck of a read.  Damn, if there was a subsequent ruling that
> said header files or software interfaces in general were not
> copywritable, then, yes, you would be right.

    Since that's not my position, how would it make me right?

> Suddenly both the GPL
> and Microsoft, which depend on tying, would be stopped dead in their
> respective tracks.

    Huh?

> The thing about this ruling is that it suggests interfaces aren't
> copywritable, but, it doesn't make the slam dunk ruling.

    I never argued that interfaces weren't copyrightable. However, *IF* the 
interface contains no creative copyrightable elements and just consists of 
the minimum straightforward functional code necessary to talk to the other 
end, it is *obviously* not copyrightable.

> But, to think
> that the Supreme Court refused to hear the case, as much as goes, this
> is this is settled law.   I guess if you've got the money for the
> lawyers, copy away.  But, if you've got the legal money, why not pay
> Trolltech for their work?

    The Supreme Court has heard similar cases. I only cited this one because 
its language is the most compact.

    I have had companies that dual license under GPL and commercial licenses 
come right out and tell me that I'm probably right legally but it's cheaper 
to license than to fight. They were right, and we licensed.

    DS


0
David
2/19/2006 7:11:15 PM
"M�ns Rullg�rd" <mru@inprovide.com> wrote in message 
news:yw1xwtfrb580.fsf@agrajag.inprovide.com...

> I don't know what intentions Trolltech had when they chose the GPL;
> they just happen to be serving as an example in this thread.  Evidence
> of the FSF are doing exactly this is everywhere, most notably in the
> GPL FAQ.

    They're wrong and they'll lose. It fails the most obvious common sense 
analysis.

    For example, suppose there's a bunch of code out there that dynamically 
links to "libmalloc". I write a brand new program, compile/link it to 
"libmalloc", and now all that existing code can dynamically link to my 
library. Someone could even distribute those existing progams with my 
"libmalloc" as the default "libmalloc" on the system.

    Now, can you possibly argue that those programs, written before I even 
started on my "libmalloc", are somehow derivative works of my program? Even 
though they were made first? Even though they contain no part of my 
"libmalloc" *at* *all*?

    Copyright does not protect minimal functional elements. If you want to 
make headers that interoperate with a shared library, and they are truly 
minimal, the only creative expression they contain is the names of the 
functions and variables and the order they are in. Not only could someone 
trivially change those things, but courts have held that that's not 
copyrightable expression anyway, especially when the order is arbitrary or 
functional rather than creative and the names are purely functional.

    DS


0
David
2/19/2006 7:18:05 PM
I wrote:
> Of course they can.  It is perfectly legal and enforceable to distribute
> your program with a file named "copyright" which says something like
>
> Flooble Version .001, a program to foobar the baz.
> Copyright David Schwartz 2006
> This program is distributed under the terms of the GPL (see COPYING)
> with the following additional restrictions:
> ...
> ...

DS writes:
> The GPL specifically says there can be no additional restrictions, so you
> would have two licensing terms that directly contradicted each other.

The GPL specifically says that _licensees_ cannot add restrictions.  It is
a model license, not a law.  It cannot tell you as sole copyright owner
what you may or may not do.  The example I gave makes it very clear that
the "additional restrictions" override anything in the GPL they conflict
with.

> Even if you could, such a product would not actually be licensed under
> the GPL because you could not distribute it under the GPL's terms. It
> could not be combined with other GPL'd software or used in GPL'd
> projects.

> Such a thing would *NEVER* be considered to be GPL'd.

Of course not.  So what?
-- 
John Hasler 
john@dhh.gt.org
Dancing Horse Hill
Elmwood, WI USA
0
John
2/19/2006 7:51:50 PM
"John Hasler" <john@dhh.gt.org> wrote in message 
news:87vevbazzd.fsf@toncho.dhh.gt.org...

> DS writes:

>> The GPL specifically says there can be no additional restrictions, so you
>> would have two licensing terms that directly contradicted each other.

> The GPL specifically says that _licensees_ cannot add restrictions.  It is
> a model license, not a law.

    Right, and it's the licensees that redistribute the software. If 
licensees cannot add restrictions, then those who receive the software 
*from* licensees have no additional restrictions. I did not get Qt from 
Trolltech. I got it from a licensee. I will gladly mail you a copy. I cannot 
impose any restrictions on you and neither can Trolltech.

> It cannot tell you as sole copyright owner
> what you may or may not do.

    It cannot tell you as sole copyright owner that you cannot grant 
additional rights. It can definitely prevent you from imposing restrictions 
on those who take advantage of the rights granted by the license. Of course, 
you are free not to license under the GPL.

> The example I gave makes it very clear that
> the "additional restrictions" override anything in the GPL they conflict
> with.

    Irrelevent, since there's no way to impose these "additional 
restrictions" on people who receive the work from third parties under the 
GPL.

>> Even if you could, such a product would not actually be licensed under
>> the GPL because you could not distribute it under the GPL's terms. It
>> could not be combined with other GPL'd software or used in GPL'd
>> projects.
>
>> Such a thing would *NEVER* be considered to be GPL'd.
>
> Of course not.  So what?

    Qt is in fact GPL'd, so unless this is some bizarre hypothetical that 
you find interesting, it's not relevant.

    DS


0
David
2/19/2006 8:25:50 PM
"David Schwartz" <davids@webmaster.com> writes:

> "M�ns Rullg�rd" <mru@inprovide.com> wrote in message 
> news:yw1xwtfrb580.fsf@agrajag.inprovide.com...
>
>> I don't know what intentions Trolltech had when they chose the GPL;
>> they just happen to be serving as an example in this thread.  Evidence
>> of the FSF are doing exactly this is everywhere, most notably in the
>> GPL FAQ.
>
>     They're wrong and they'll lose. It fails the most obvious common sense 
> analysis.
>
>     For example, suppose there's a bunch of code out there that dynamically 
> links to "libmalloc". I write a brand new program, compile/link it to 
> "libmalloc", and now all that existing code can dynamically link to my 
> library. Someone could even distribute those existing progams with my 
> "libmalloc" as the default "libmalloc" on the system.
>
>     Now, can you possibly argue that those programs, written before I even 
> started on my "libmalloc", are somehow derivative works of my program? Even 
> though they were made first? Even though they contain no part of my 
> "libmalloc" *at* *all*?

I've attempted that very argument on the debian-legal mailing list.
All I met with was hand-waving and accusations of trolling from the
list regulars.

-- 
M�ns Rullg�rd
mru@inprovide.com
0
iso
2/19/2006 9:55:32 PM
Stork replied to:

> I have had companies that dual license under GPL and commercial licenses
>come right out and tell me that I'm probably right legally but it's cheaper
>to license than to fight. They were right, and we licensed.

But what ethical planet do you come from that you would think of even
doing that?  If the software these people use is so good that you can
use it, do you not even think of entering into a fair agreement with
them?  Your first inclination is to completely fuck over the very
people that are making your product possible, and use some lame excuse
of copywrite law to cover the fact that you cannot be relied on to
enter an agreement in good faith because you have none.

0
stork
2/19/2006 10:00:44 PM
Storkyak replied to:

>They're wrong and they'll lose. It fails the most obvious common sense
>analysis.

Let's cut away all of this obfuscating crap and call it for what it is:

Vendor provides a product.  You choose to use the product, and benefit
from it.  As such you are on the receiving end of a license, which you
then choose to break.  No amount of supreme court citations or self
proclaimed expertise can paper over the fact that you are using
something without any return in kind for those that provided it to you.
 You look at the GPL license and use an excuse of copywrite law in
order to break a contract made supposedly in good faith, all largely so
you can save a few bucks.  This isn't about copywrite law.  This is
about you not wanting to pay someone for the software that they
delivered, either by honoring the ideals of their contract or by paying
them. 

Cut away all the crap, and you are stealing.  Plain and simple.

0
stork
2/19/2006 10:04:26 PM
"stork" <stork@storkyak.com> writes:

> Storkyak replied to:
>
>>They're wrong and they'll lose. It fails the most obvious common sense
>>analysis.
>
> Let's cut away all of this obfuscating crap and call it for what it is:
>
> Vendor provides a product.  You choose to use the product, and benefit
> from it.  As such you are on the receiving end of a license, which you
> then choose to break.  No amount of supreme court citations or self
> proclaimed expertise can paper over the fact that you are using
> something without any return in kind for those that provided it to you.

The product being provided under the GPL means that you do not need to
pay to use it.  If the vendor wants to be paid for the product, he
shouldn't be giving it away for free.  It's really that simple.

-- 
M�ns Rullg�rd
mru@inprovide.com
0
iso
2/19/2006 10:29:03 PM
"M�ns Rullg�rd" <mru@inprovide.com> wrote in message 
news:yw1xslqfau97.fsf@agrajag.inprovide.com...

> I've attempted that very argument on the debian-legal mailing list.
> All I met with was hand-waving and accusations of trolling from the
> list regulars.

    The problem is people with no legal education who assume that 
intellectual property law works the way they naively assume it must work or 
the way they think it should work. To an extent, it does, but it also has 
some odd quirks that are not well understood outside the legal community.

    For example, numerous people have suggested to me the use of a 
copyrighted string as a "lock out" code to prevent interoperability. Heck, I 
saw a business plan built around this concept. Of course it won't work --  
copyright is based upon there being more than one way to express the *same* 
idea and protecting your choice of ways.

    DS


0
David
2/21/2006 3:49:00 AM
"stork" <stork@storkyak.com> wrote in message 
news:1140386444.137066.288770@o13g2000cwo.googlegroups.com...

> Stork replied to:

>> I have had companies that dual license under GPL and commercial licenses
>>come right out and tell me that I'm probably right legally but it's 
>>cheaper
>>to license than to fight. They were right, and we licensed.

> But what ethical planet do you come from that you would think of even
> doing that?

    The planet where people are not entitled to claim ownership of what is 
not theirs.

> If the software these people use is so good that you can
> use it, do you not even think of entering into a fair agreement with
> them?

    The software got so good because of people who contributed to the 
program *because* of its licensing terms.

> Your first inclination is to completely fuck over the very
> people that are making your product possible, and use some lame excuse
> of copywrite law to cover the fact that you cannot be relied on to
> enter an agreement in good faith because you have none.

    They are the ones who are not honoring in good faith the terms of the 
license they chose to use to develop and popularize their product.

    DS


0
David
2/21/2006 3:50:10 AM
"M�ns Rullg�rd" <mru@inprovide.com> wrote in message 
news:yw1xlkw7aspc.fsf@agrajag.inprovide.com...

> The product being provided under the GPL means that you do not need to
> pay to use it.  If the vendor wants to be paid for the product, he
> shouldn't be giving it away for free.  It's really that simple.

    Exactly. The vendor wants to distribute the product under the GPL to get 
the benefit of people popularizing, testing, and even contributing to their 
product. Then when people try to use the product for free, the very purpose 
of the GPL, they argue that you need a commercial license even though you in 
fact do not.

    I know these seem like obscure technical arguments to non-legal types. 
But they're not. They're absolutely fundamental to the notion of fair use.

    The supreme irony is that it's a disgusting alliance between the free 
software people and the most radical intellectual property nazis that are 
arguing that taking just what you need to interoperate isn't fair use.

    DS


0
David
2/21/2006 3:52:29 AM
    Let me say one more thing:

    The GPL is supposed to be a license that grants additional rights you 
wouldn't otherwise have. When you receive a product under the GPL, you are 
supposed to have every single right you'd have if you bought the program 
plus some additional rights. Further, the GPL is designed to allow you to 
grant those same rights that you have to others with ease.

    To argue that the GPL doesn't allow you to do something would be 
necessarily to argue that you could not do that same thing if you bought the 
program. Fair use, first sale and all the sources of rights are based simply 
on you lawfully acquiring a copy of the work, they make no distinction 
between buying something and getting it for free.

    The GPL grants you some *additional* rights that have special conditions 
(for example, you must distribute source code). However, you should be able 
to exercise every right you have if you bought the product without having to 
comply with any of those special conditions. (This is because the GPL is not 
a EULA or shrink wrap agreement.)

    If a library is distributed under the GPL, then anyone can lawfully 
acquire a copy and gain all the rights they would have if they bought that 
program without even complying with any of the requirements of the GPL 
because these are rights that come simply from lawfully acquiring a copy.

    Everything I've said above is universally agreed upon by knowledgeable 
people. It's the FSF's position. It's my position. It's what we all know and 
understand before we start arguing. Don't argue with the above unless you're 
sure you know more about the law than I do. ;)

    So if you have to release your source code if you distribute a program 
that uses a library distributed under the GPL, it would only be because you 
did not have the right to do that under rights of first sale or fair use. In 
other words, that would mean that even if you and I both bought a commercial 
library, we could not exchange source code that used that library without 
permission of the library author.

    The point is, the legal position the free software community should be 
taking is that if you take only what you need to interoperate, that's fair 
use. Interoperating is an idea, not an expression of an idea. By taking the 
opposition position, the free software community is seriously damaging fair 
use and leading to a world where projects like samba would be illegal 
without Microsoft's permission. That's not the world I want to live in.

    I want to live in a world where copryight and patent laws are balanced 
and fair use is extensive. However, the FSF wants more control for copyright 
owners because that's what makes the GPL work.

    Choose your sides carefully.

    DS

    PS: I am, almost, saying that an interface is an idea, not an expression 
of an idea, and should not be copyrightable. If you have to use a particular 
interface to interoperate with a particular product, then there is no 
multiple choices, the pre-condition of copyrightability. This is a slight 
simplification of my view, but it's the gist of it.


0
David
2/21/2006 4:18:29 AM
Stork replied to:

>The planet where people are not entitled to claim ownership of what is
>not theirs.

The software is theirs, not yours.

>The software got so good because of people who contributed to the
>program *because* of its licensing terms

No, the software got so good because it is a -better product- made by
-better developers-.  There's plenty of free GUI alternatives out there
for you.  You could have used wxWidgets, for example.

>They are the ones who are not honoring in good faith the terms of the
>license they chose to use to develop and popularize their product.

No, you are just making up cheesy excuses to steal their work.  The
hole in your argument is that you say that licensing terms made their
product better, rather than its merits, yet, you feel you can't use
someone else's product.  wxWidgets has a "use for anything" licensing
scheme. .. heck, for that matter, why didn't you just write your
product for Windows, which not only has a closed source library set,
but a damned good one, and, on top of that, you don't have to pay any
royalties to use their stuff at all, as users pay the tax.  What the
difference between Linux and Windows really is, is that in Linux
developers pay for the libraries they use - either by licensing in
kind, ala GPL, or by royalties / fees to people such as troll tech.  On
the other hand, in Windows, users pay for the vast majority of Windows
development.  Yeah Visual Studio is hideously expensive in its
Professional and Enterprise glory, but for Visual Studio money you are
still only partially towards paying Trolltech, unless of course, you
advocate stealing their software, like you do.

0
stork
2/23/2006 5:17:24 AM
Stork replied to:

>I did not get Qt from Trolltech. I got it from a licensee. I will gladly mail you a copy. I cannot impose any restrictions on you and neither >can Trolltech.

So, if someone hands me a copy of your company's program, and I post it
on the Internet for free, that would be the same thing, now wouldn't
it?  I mean, what's the difference between how you would treat
Trolltech and how you would expect your own firm to be treated.

0
stork
2/23/2006 5:19:22 AM
Stork replied to:

>If the vendor wants to be paid for the product, he
>shouldn't be giving it away for free.  It's really that simple.

They aren't.  Remember, they are charging a licensing fee that you
don't want to pay.

0
stork
2/23/2006 5:23:07 AM
Stork replied to:

>The point is, the legal position the free software community should be
>taking is that if you take only what you need to interoperate, that's fair
>use. Interoperating is an idea, not an expression of an idea

But Aye, there's the rub.  Interoperating is essentially all you really
need out of a library.  If you make interoperability fair use, then,
you basically introduce the MS model over to Linux - because the OS
vendors would be the only ones that could realistically recover
developmental costs, and your whole Linux Open Source dream goes out
the window.

0
stork
2/23/2006 5:33:24 AM
Stork replied to:

> However, the FSF wants more control for copyright
>owners because that's what makes the GPL work

I'm a copyright owner, my wife is an artist and my best friend is a
patent attorney.  Thus, I lean more towards stronger copyright
protections for owners and less towards "fair use".  If you want to use
something, you should pay for it.  Otherwise, don't use it.  If you can
get a reduced rate by doing some open source thing, then so be it,
choose between a GPL license offering or another, traditional, license
offering.  But don't be walking into the candy store thinking that you
can bust open a bag of M&Ms because it is unfair that that you can't
get a bag of just the yellow ones.

The paradoxical position, that I would argue, is that open source
software actually won't be practical until we have DRM on Linux.  That
way, an open source developer would have more cost recovery options
because one could use micropayments to determine the use of a library
or even a few lines of source code.

FSF errs, I would agree with you, in that, their knee jerk reaction
against DRM conflicts with their own ideology.

0
stork
2/23/2006 5:42:37 AM
Stork replied to:

>The supreme irony is that it's a disgusting alliance between the free
>software people and the most radical intellectual property nazis that are
>arguing that taking just what you need to interoperate isn't fair use.

Politics makes strange bedfellows, and I don't even know that I believe
in "fair use".  I think if you want to use someone's work, you should
pay for it.

0
stork
2/23/2006 5:48:50 AM
"stork" <stork@storkyak.com> wrote in message 
news:1140671962.398346.209010@i39g2000cwa.googlegroups.com...
> Stork replied to:
>
>>I did not get Qt from Trolltech. I got it from a licensee. I will gladly 
>>mail you a copy. I cannot impose any restrictions on you and neither >can 
>>Trolltech.

> So, if someone hands me a copy of your company's program, and I post it
> on the Internet for free, that would be the same thing, now wouldn't
> it?  I mean, what's the difference between how you would treat
> Trolltech and how you would expect your own firm to be treated.

    Are you reading anything that I'm saying? I'm talking about the 
consequences of the GPL, not things that aren't available under the GPL.

    DS


0
David
2/23/2006 6:21:42 AM
"stork" <stork@storkyak.com> wrote in message 
news:1140672804.783966.268780@f14g2000cwb.googlegroups.com...

> Stork replied to:

>>The point is, the legal position the free software community should be
>>taking is that if you take only what you need to interoperate, that's fair
>>use. Interoperating is an idea, not an expression of an idea

> But Aye, there's the rub.  Interoperating is essentially all you really
> need out of a library.

    Umm, no, you also need the right to use the library, for example.

> If you make interoperability fair use, then,
> you basically introduce the MS model over to Linux - because the OS
> vendors would be the only ones that could realistically recover
> developmental costs, and your whole Linux Open Source dream goes out
> the window.

    Huh? What dream are you talking about?

    DS


0
David
2/23/2006 6:22:24 AM
"stork" <stork@storkyak.com> writes:

> I'm a copyright owner, my wife is an artist and my best friend is a
> patent attorney.  Thus, I lean more towards stronger copyright
> protections for owners and less towards "fair use".  If you want to use
> something, you should pay for it.  

You are profoundly missing the point(s):

1. Wildly uninhibited free use is the baseline.  As citizens we
   jointly agreed to impose restrictions on that in order to encourage
   innovation.  Thus patent and copyright law.  However, we as
   citizens recognize that eventually we as a society and civilization
   need uninhibited use.  Thus the time limits on patent and copyright
   law.  Despite Disney-purchased extensions to the time limits, this
   is not about rights in perpetuity.

2. Having established time-limited rights, you are welcome to play by
   any rules you wish.  Prohibit all reuse.  Require $1B for each use.
   Require reciprocal rights to the other guy's patents.  Or...the
   conditions of OSI-compliant licenses.  It is dealer's choice.

   "You should pay for it" is a personal decision by the innovator,
   not a feature of patent or copyright law.

3. TrollTech provides QT under commercial and GPL licenses.  That is
   their choice.  It is up to others to decide if they really want to
   use a linked library under GPL.  Personally I do not.

-- 
harry.g.george@boeing.com
6-6M21 BCA CompArch Design Engineering
Phone: (425) 294-4718
0
Harry
2/23/2006 2:35:45 PM
Stork replied to:

>Wildly uninhibited free use is the baseline

For all practical purposes, it is not, any more.  Congress has made it
fairly clear that they are more in favor of extending rights to the
authors and creators over the last 50 years than they are in protecting
wildly uninhibited use.  To wit, any copyright made by any of us now
will outlive us AND our children.

>Or...the    conditions of OSI-compliant licenses.  It is dealer's choice

Bingo, that is all I'm saying.

>TrollTech provides QT under commercial and GPL licenses.  That is  their choice.  It is up to others to decide if they really want to use a linked library under GPL

All I'm saying is, if Trolltech says that its their work, and that you
have to GPL in order to even link, then, that's their licensing intent
and it is probably fundamental to their business model that they ask
that.  I think its reasonable to have a licensing position that says,
if you are making money with our stuff, we want a piece of it,
otherwise, we don't care.  That's pretty fair and it is sorta
fundamental to economics.  Microsoft doesn't ask that, they just charge
you.

0
stork
2/23/2006 4:55:20 PM
Harry George <harry.g.george@boeing.com> writes:

> "stork" <stork@storkyak.com> writes:
>
>> I'm a copyright owner, my wife is an artist and my best friend is a
>> patent attorney.  Thus, I lean more towards stronger copyright
>> protections for owners and less towards "fair use".  If you want to use
>> something, you should pay for it.  
>
> You are profoundly missing the point(s):

[...]

> 3. TrollTech provides QT under commercial and GPL licenses.  That is
>    their choice.  It is up to others to decide if they really want to
>    use a linked library under GPL.  Personally I do not.

I came across a piece of funny reading:
http://business.timesonline.co.uk/article/0%2c%2c9075-2051196%2c00.html

-- 
M�ns Rullg�rd
mru@inprovide.com
0
iso
2/23/2006 10:04:21 PM
"stork" <stork@storkyak.com> writes:

> Stork replied to:
> 
> >Wildly uninhibited free use is the baseline
> 
> For all practical purposes, it is not, any more.  

I'm speaking from global, historical perspective, not from the
confines of current US law.  As in:
http://www.seanet.com/~hgg9140/politics/law/index.html
 

> Congress has made it
> fairly clear that they are more in favor of extending rights to the
> authors and creators over the last 50 years than they are in protecting
> wildly uninhibited use.  To wit, any copyright made by any of us now
> will outlive us AND our children.
> 

Thus my reference to Disney.  Liberty is a more powerful meme than
DMCA or patent law or copyright.   Eventually it wins.

-- 
harry.g.george@boeing.com
6-6M21 BCA CompArch Design Engineering
Phone: (425) 294-4718
0
Harry
2/24/2006 3:07:59 PM
Stork replied to:

>Liberty is a more powerful meme than DMCA or patent law or copyright.   Eventually it wins.

And somewhere along the line, liberty does not mean liberty to steal.

0
stork
2/24/2006 3:46:13 PM
Stork replies to:

>Are you reading anything that I'm saying? I'm talking about the
>consequences of the GPL, not things that aren't available under the GPL

No, you are arguing that you have a right to use a piece of software
irrespective of the vendor's license terms within your understanding of
the law and not within the confines of any moral obligation to enter
into a contract in good faith.  Thus, the law is all that matters and
not any morals that support it.  Other people can play by the same
rules that you establish as well, and with outcomes not to your liking.


If you can make up a definition of fair use that suits you, then,
certainly I could make one up that suits me.  I could, in theory, given
sufficient funding, make the law be anything I want, either via a court
fight or flat out legislation.  All of that could be avoided, as I said
earlier, if you were just willing to honor the spirit of the vendor's
offering, and, if you didn't like it, use some other vendor's products.
 That's how the free market works, ya know.

0
stork
2/24/2006 3:53:29 PM
"stork" <stork@storkyak.com> writes:

> Stork replied to:
>
>>Liberty is a more powerful meme than DMCA or patent law or
>>copyright.  Eventually it wins.
>
> And somewhere along the line, liberty does not mean liberty to steal.

Nor does it permit extortion.

-- 
M�ns Rullg�rd
mru@inprovide.com
0
iso
2/24/2006 6:33:57 PM
"stork" <stork@storkyak.com> writes:

> Stork replies to:
>
>>Are you reading anything that I'm saying? I'm talking about the
>>consequences of the GPL, not things that aren't available under the GPL
>
> No, you are arguing that you have a right to use a piece of software
> irrespective of the vendor's license terms within your understanding of
> the law and not within the confines of any moral obligation to enter
> into a contract in good faith.  Thus, the law is all that matters and
> not any morals that support it.  Other people can play by the same
> rules that you establish as well, and with outcomes not to your liking.
>
> If you can make up a definition of fair use that suits you, then,
> certainly I could make one up that suits me.  I could, in theory, given
> sufficient funding, make the law be anything I want, either via a court
> fight or flat out legislation.  All of that could be avoided, as I said
> earlier, if you were just willing to honor the spirit of the vendor's
> offering, and, if you didn't like it, use some other vendor's products.
>  That's how the free market works, ya know.

Stork, do you by any chance work for the MPAA?  Your views on fair use
and similar concepts suggest that this might be the case.

-- 
M�ns Rullg�rd
mru@inprovide.com
0
iso
2/24/2006 6:37:30 PM
Stork replied to:

> I already addressed this argument. It's 100% totally wrong

No, not at all. The license is a contract.  You either agree to its
terms in order to receive a good or service, or you do not.

0
stork
2/24/2006 8:04:40 PM
Stork replied to:

>Nor does it permit extortion.

Where is the extortion in the case of GUIs?  There's nothing that
requires you to use Trolltech's tools, particularly in the case of
Linux.  Now, if Trolltech had some bizarre set of patents that
conferred upon them a monopoly, then I would agree with you.  But they
really don't, or, whatever they have isn't really significant or useful.

0
stork
2/24/2006 8:06:20 PM
Stork replied to:

>That's is such total bullshit. If Trolltech is trying to capitalize on
>the popularity of the GPL while refusing to honor its spirit, they are the
>ones you are ripping people off

There's nothing about a person's marketing that conferrs upon you the
right to steal their stuff.  You either like the product or you don't.
If you like the product, then why not enter into a contract in good
faith, that is, honor the license?

>  And what the hell is up with the "free software" community arguing that
>taking the absolute minimum you need to interoperate isn't fair use even
>when courts seem to think it is?

I'm not a member of the Free Software Community.  I have several closed
source products for Windows and I'm willing to make open source
products for Linux, because I have a business model that I hope will
support it.  I believe in private property and strong intellectual
property.

0
stork
2/24/2006 8:09:12 PM
"stork" <stork@storkyak.com> writes:

> Stork replied to:
>
>> I already addressed this argument. It's 100% totally wrong
>
> No, not at all. The license is a contract.  You either agree to its
> terms in order to receive a good or service, or you do not.

The terms allow you use and distribute the library without limit, free
of charge.  Why do you not want to agree with this?

-- 
M�ns Rullg�rd
mru@inprovide.com
0
iso
2/24/2006 8:49:38 PM
Stork replied to:

>The terms allow you use and distribute the library without limit, free
>of charge.  Why do you not want to agree with this

IF and only IF you either make your application GPL, or, pay the per
developer seat fee.

0
stork
2/25/2006 1:06:57 AM
"stork" <stork@storkyak.com> wrote in message 
news:1140796409.462513.64350@t39g2000cwt.googlegroups.com...

> Stork replies to:

>>Are you reading anything that I'm saying? I'm talking about the
>>consequences of the GPL, not things that aren't available under the GPL

> No, you are arguing that you have a right to use a piece of software
> irrespective of the vendor's license terms within your understanding of
> the law and not within the confines of any moral obligation to enter
> into a contract in good faith.

    That is correct. If there is no EULA or shrink-wrap agreement and you 
didn't sign any license and you legally acquired the software, then you have 
the right to the normal and expected use of that. Otherwise, if you bought a 
CD, you couldn't play. Or if I dropped copies of a poem from an airplane, I 
could sue everyone who read it.

> Thus, the law is all that matters and
> not any morals that support it.  Other people can play by the same
> rules that you establish as well, and with outcomes not to your liking.

    It's not my fault that you don't understand the morals that support the 
law in this area. I can try to educate you, but it's a long road and 
requires a lot of information that you probably don't have. It probably 
didn't occur to you that someone might drop a poem from an airplane and then 
sue everyone who read it, but this is the reason that is *must* be the case 
that if you get something for free, you have the right to normal use of it.

    I can go through all these issues with you case by case, but it gets 
very technical. We can stick to one or two, and we'll probably get further.

    But it is the law (called "first sale") and it makes moral sense that if 
you legally acquire a copyrighted work (without shrink-wrap agreement, EULA, 
or signed contract), you have the right to the normal use of that work. The 
GPL is not a shrink-wrap agreement, EULA, or signed contract.

> If you can make up a definition of fair use that suits you, then,
> certainly I could make one up that suits me.  I could, in theory, given
> sufficient funding, make the law be anything I want, either via a court
> fight or flat out legislation.  All of that could be avoided, as I said
> earlier, if you were just willing to honor the spirit of the vendor's
> offering, and, if you didn't like it, use some other vendor's products.
> That's how the free market works, ya know.

    The law happens to be right in this area. In this hypothetical, it would 
be the vendor that would be abusing both the letter and spirit of the law by 
dropping copies of their poem from airplanes and suing anyone who read them.

    You cannot give something away then charge for its use without a 
contract. That makes legal and moral sense.

    DS


0
David
2/25/2006 8:44:39 AM
"stork" <stork@storkyak.com> wrote in message 
news:1140811480.182282.104830@u72g2000cwu.googlegroups.com...
> Stork replied to:

>> I already addressed this argument. It's 100% totally wrong

> No, not at all. The license is a contract.  You either agree to its
> terms in order to receive a good or service, or you do not.

    You aren't reading what I'm saying or you have no framework with which 
to understand it. I'll try again to explain it, but my patience is running 
out.

    The GPL gives you *additional* rights, beyond the rights you get just 
because you legally acquired a work. You do not need to accept the GPL for 
normal use, which would be everything you can do with a work that you 
purchased.

    You do not have to agree to the GPL to receive a GPL'd work. Receiving 
the work is what gives you the right to use it in its normal and expected 
use. (Which, for a library, is to develop programs that use it.)

    This is not even a subject of debate. Nobody with any basic knowledge of 
law argues that you must agree to the GPL to use a work in its normal and 
expected way, not even the FSF.

    DS


0
David
2/25/2006 8:49:21 AM
"stork" <stork@storkyak.com> wrote in message 
news:1140811752.507667.308670@z34g2000cwc.googlegroups.com...

> Stork replied to:

>>That's is such total bullshit. If Trolltech is trying to capitalize on
>>the popularity of the GPL while refusing to honor its spirit, they are the
>>ones you are ripping people off

> There's nothing about a person's marketing that conferrs upon you the
> right to steal their stuff.  You either like the product or you don't.
> If you like the product, then why not enter into a contract in good
> faith, that is, honor the license?

    I'm not stealing it, they're giving it to me.

>>  And what the hell is up with the "free software" community arguing that
>>taking the absolute minimum you need to interoperate isn't fair use even
>>when courts seem to think it is?

> I'm not a member of the Free Software Community.  I have several closed
> source products for Windows and I'm willing to make open source
> products for Linux, because I have a business model that I hope will
> support it.  I believe in private property and strong intellectual
> property.

    Me too.

    DS


0
David
2/25/2006 8:50:01 AM
"stork" <stork@storkyak.com> wrote in message 
news:1140829617.451541.69710@e56g2000cwe.googlegroups.com...

> Stork replied to:

>>The terms allow you use and distribute the library without limit, free
>>of charge.  Why do you not want to agree with this

> IF and only IF you either make your application GPL, or, pay the per
> developer seat fee.

    That is simply not true. Where does the GPL say that if you want to 
distribute a library, you must distribute the source code to an application 
that uses it?

    Bluntly, you don't seem to have any idea what you're talking about. You 
don't even understand any of the things that all sides *agree* on. You're 
out on your own somewhere in the middle of nowhere.

    DS


0
David
2/25/2006 8:51:57 AM
Stork replied to:

> You cannot give something away then charge for its use without a
>contract. That makes legal and moral sense.

I see your point in as much as, if an end user has a piece of software
on their machine, they should be allowed to buy applications that tap
into it.  I simply disagree.  If the user did not want that facility on
their machine, they can reject the installation of that software -
particularly under Linux, and that software's existence confers no
right to you.  As an artist though, I think it is useful to be able to
carve out a section of digital real estate and charge tolls across it
on a user's PC, so, your argument makes no sense whatsover.

But.... that's not really your case, now, is it....

In your case the breach of contract is actually on the developer's PC.
In order for you to effectively and realistically develop against that
system, you have to use all the tools and documentation that Trolltech
developed.  Specifically, you are going to use all the Qt stuff in
KDevelop, and then just back out only what "you need".  You fire up all
the Qt designer, examples, documentations, developing a product the
very rich set of tools that they provide, but not paying for their use,
per their contract that you very much did accept in KDevelop, and then
you break it... there's a reason that the Trolltech licenses are per
developer...

Most people look at KDevelop and say, geez, if I use all this KDE / Qt
stuff, for a commercial product, I have to fork over 3 grand per
developer [or something like that].  That's actually not too much more
what a full set of tools for a Windows developer will run these days -
particularly if you have TOAD and Visual Studio Enterprise and
Office.... so the money should not be prohibitive to a respectable
business.  You look at it, and say, geez, how can I use this stuff
without having to pay for it.  You know, you argue that someone put the
candy in a jar on the counter at the 7-11, so therefor under fair use,
you can just take it.  Sometimes you just have to walk away, if you
don't want it.

0
stork
2/25/2006 2:02:54 PM
"stork" <stork@storkyak.com> wrote in message 
news:1140876174.945089.11500@z34g2000cwc.googlegroups.com...

> In your case the breach of contract is actually on the developer's PC.
> In order for you to effectively and realistically develop against that
> system, you have to use all the tools and documentation that Trolltech
> developed.  Specifically, you are going to use all the Qt stuff in
> KDevelop, and then just back out only what "you need".  You fire up all
> the Qt designer, examples, documentations, developing a product the
> very rich set of tools that they provide, but not paying for their use,
> per their contract that you very much did accept in KDevelop, and then
> you break it... there's a reason that the Trolltech licenses are per
> developer...

    I apologize if I wasn't clear about something, and I can't stress this 
enough. I'm not saying anything special about Qt, KDeveloper, or Trolltech. 
I know very little about them. I was talking purely about the hypothetical 
case where a company distributes a library with both the GPL and a 
commercial license and was analyzing what use would be allowed without 
needing to purchase the commercial license.

    I have seen one company pressure companies into buying commercial 
licenses even when their own legal experts felt it wasn't needed, because it 
was debatable, and licensing is cheaper than defending a suit. The company 
was not Trolltech.

    Legally, morally, you cannot give something away and then charge people 
for using it without a contract. A license like the GPL is not a contract 
because you do not need to agree to it to get the thing it licenses.

    DS


0
David
2/25/2006 10:23:32 PM
Stork replied to:

> I was talking purely about the hypothetical
>case where a company distributes a library with both the GPL and a
>commercial license and was analyzing what use would be allowed without
>needing to purchase the commercial license

Ok.  For sake of argument then, in your case do you assume that the
library you want to use is distributed with an operating system, such
that, there's an expectation that if you deploy a product that it will
have this library. In other words, the library is already there and the
end user has the library licensed via some other mechanism?

0
stork
2/25/2006 10:46:41 PM
"stork" <stork@storkyak.com> writes:

> Stork replied to:
>
>> I was talking purely about the hypothetical
>>case where a company distributes a library with both the GPL and a
>>commercial license and was analyzing what use would be allowed without
>>needing to purchase the commercial license
>
> Ok.  For sake of argument then, in your case do you assume that the
> library you want to use is distributed with an operating system, such
> that, there's an expectation that if you deploy a product that it will
> have this library. In other words, the library is already there and the
> end user has the library licensed via some other mechanism?

Or you could just ask the user to download and install the library
himself before installing your software.

-- 
M�ns Rullg�rd
mru@inprovide.com
0
iso
2/25/2006 11:38:09 PM
"stork" <stork@storkyak.com> wrote in message 
news:1140907601.441699.49500@i39g2000cwa.googlegroups.com...

> Stork replied to:

>> I was talking purely about the hypothetical
>>case where a company distributes a library with both the GPL and a
>>commercial license and was analyzing what use would be allowed without
>>needing to purchase the commercial license

> Ok.  For sake of argument then, in your case do you assume that the
> library you want to use is distributed with an operating system, such
> that, there's an expectation that if you deploy a product that it will
> have this library. In other words, the library is already there and the
> end user has the library licensed via some other mechanism?

    It makes no difference. When you place something under the GPL and you 
sell or transfer it to at least one person, you have effectively made 
everyone a legal licensee of the work.

    I see no legal or moral difference between a person who has the work and 
a person who doesn't, so long as the work has been distributed to at least 
one person who is willing to distribute it. The issue is not one of physical 
possession of the covered work but one of possessing the right to use the 
covered work.

    Again, if you don't like that, don't place a work under the GPL.

    DS


0
David
2/27/2006 12:21:55 PM
"M�ns Rullg�rd" <mru@inprovide.com> wrote in message 
news:yw1x1wxr102m.fsf@agrajag.inprovide.com...

> Or you could just ask the user to download and install the library
> himself before installing your software.

    No need, the GPL gives you the right to distribute it to him.

    If your work is a derivative work of the covered work, you can't 
distribute your own work to him anyway. If your work is not a derivative 
work, the GPL does not require you to provide the source code to it.

"The 'Program', below, refers to any such program or work, and a "work based 
on the Program" means either the Program or any derivative work under 
copyright law: that is to say, a work containing the Program or a portion of 
it, either verbatim or with modifications and/or translated into another 
language."

    Note that the phrase "work containing the Program or a portion of it" is 
not intended to be definitive here, it's just meant to be a brief summary of 
what it means for a work to be a "derivative work". You can get some truly 
absurd results if you take this phrase literally!

    DS


0
David
2/27/2006 12:24:46 PM
Stork replied to:

>When you place something under the GPL and you
>sell or transfer it to at least one person, you have effectively made
>everyone a legal licensee of the work

Nope, because, as a copyright owner, I have the right to determine who
can use my work or not.  The GPL is not a transfer of copyright, as you
would argue that it is.  Besides, your viewpoint is silly.  You would
argue that because GM might give away a few hundred Pontiac G6's on
Oprah, that everyone should get it for free.

0
stork
2/28/2006 3:26:16 PM
Stork replied to:

>No need, the GPL gives you the right to distribute it to him.

Yes, but as the copyright holder, if I get wind of it, I can revoke
your license to do so.  MySQL has been able to enforce this point of
view in court, for this exact problem.

0
stork
2/28/2006 3:42:50 PM
On 2006-02-28, stork <stork@storkyak.com> wrote:
> Stork replied to:
>
>>No need, the GPL gives you the right to distribute it to him.
>
> Yes, but as the copyright holder, if I get wind of it, I can revoke
> your license to do so.  MySQL has been able to enforce this point of
> view in court, for this exact problem.

Based on what I can tell on google, this was NOT a revocation by the
copyright holder, but a GPL violation. And the court did not rule on the
GPL/copyright violation, only the trademark violation.
0
Jordan
2/28/2006 4:44:52 PM
Jordan Abel <random832@gmail.com> writes:

> On 2006-02-28, stork <stork@storkyak.com> wrote:
>> Stork replied to:
>>
>>>No need, the GPL gives you the right to distribute it to him.
>>
>> Yes, but as the copyright holder, if I get wind of it, I can revoke
>> your license to do so.  MySQL has been able to enforce this point of
>> view in court, for this exact problem.
>
> Based on what I can tell on google, this was NOT a revocation by the
> copyright holder, but a GPL violation. And the court did not rule on the
> GPL/copyright violation, only the trademark violation.

That sounds about right, which means that we still don't know for sure
that it was a violation.

-- 
M�ns Rullg�rd
mru@inprovide.com
0
iso
2/28/2006 7:20:24 PM
"stork" <stork@storkyak.com> writes:

> Stork replied to:
>
>>When you place something under the GPL and you
>>sell or transfer it to at least one person, you have effectively made
>>everyone a legal licensee of the work
>
> Nope, because, as a copyright owner, I have the right to determine who
> can use my work or not.

You have the right to determine who receives copies of your work.
What they do with the copies is purely their business.  By releasing
your work under the GPL you have given anyone who receives a copy the
right to create and distribute more copies.  Apparently you cannot
understand this.

> The GPL is not a transfer of copyright, as you would argue that it
> is.

Nobody has argued that.  The GPL is a grant of rights to create and
distribute derived works, no more and no less.

> Besides, your viewpoint is silly.  You would argue that because GM
> might give away a few hundred Pontiac G6's on Oprah, that everyone
> should get it for free.

Again, nobody is saying anything like that.  However, what you are
saying is comparable to saying that GM would have the power to
disallow anyone who received on of those cars to drive it to New York.

-- 
M�ns Rullg�rd
mru@inprovide.com
0
iso
2/28/2006 7:27:49 PM
"stork" <stork@storkyak.com> wrote in message 
news:1141140376.112689.282490@p10g2000cwp.googlegroups.com...

> Stork replied to:

>>When you place something under the GPL and you
>>sell or transfer it to at least one person, you have effectively made
>>everyone a legal licensee of the work

> Nope, because, as a copyright owner, I have the right to determine who
> can use my work or not.

    That is correct.

> The GPL is not a transfer of copyright, as you
> would argue that it is.

    No, it is not a transfer of copyright, it is a license to use, modify, 
distribute, and so on.

> Besides, your viewpoint is silly.  You would
> argue that because GM might give away a few hundred Pontiac G6's on
> Oprah, that everyone should get it for free.

    No, I would not argue that. I am arguing, however, that if they gave 
away a few hundred Pontiac G6's without restrictions, they cannot prevent me 
from driving one. Normally, Pontiac would have the right to say I cannot 
drive a G6, until they give one away without restrictions, then I have the 
right to drive one.

    DS


0
David
2/28/2006 8:23:58 PM
"stork" <stork@storkyak.com> wrote in message 
news:1141141370.093856.251970@u72g2000cwu.googlegroups.com...

> Stork replied to:

>>No need, the GPL gives you the right to distribute it to him.

> Yes, but as the copyright holder, if I get wind of it, I can revoke
> your license to do so.

    You cannot revoke a person's GPL rights unless they violate the GPL.

> MySQL has been able to enforce this point of
> view in court, for this exact problem.

    I have no idea what you're talking about. How about a reference so I can 
show that your understanding of the law is as accurate on this as it has 
been in the past.

    DS


0
David
2/28/2006 8:24:46 PM
On 2006-02-28, David Schwartz <davids@webmaster.com> wrote:
>     No, I would not argue that. I am arguing, however, that if they gave 
> away a few hundred Pontiac G6's without restrictions, they cannot prevent me 
> from driving one. Normally, Pontiac would have the right to say I cannot 
> drive a G6, until they give one away without restrictions, then I have the 
> right to drive one.

Right - because one of the people they gave it to might have decided to
let you drive it, and that's no longer Pontiac's business.
0
Jordan
2/28/2006 9:13:48 PM
"David Schwartz" <davids@webmaster.com> writes:

> "stork" <stork@storkyak.com> wrote in message 
> news:1141140376.112689.282490@p10g2000cwp.googlegroups.com...
>
>> Stork replied to:
>
>>>When you place something under the GPL and you
>>>sell or transfer it to at least one person, you have effectively made
>>>everyone a legal licensee of the work
>
>> Nope, because, as a copyright owner, I have the right to determine who
>> can use my work or not.
>
>     That is correct.
>
>> The GPL is not a transfer of copyright, as you
>> would argue that it is.
>
>     No, it is not a transfer of copyright, it is a license to use, modify, 
> distribute, and so on.
>
>> Besides, your viewpoint is silly.  You would
>> argue that because GM might give away a few hundred Pontiac G6's on
>> Oprah, that everyone should get it for free.
>
>     No, I would not argue that. I am arguing, however, that if they gave 
> away a few hundred Pontiac G6's without restrictions, they cannot prevent me 
> from driving one. Normally, Pontiac would have the right to say I cannot 
> drive a G6, until they give one away without restrictions, then I have the 
> right to drive one.
>
>     DS
>

-- 
M�ns Rullg�rd
mru@inprovide.com
0
iso
2/28/2006 9:31:42 PM
"David Schwartz" <davids@webmaster.com> writes:

> "stork" <stork@storkyak.com> wrote in message 
> news:1141140376.112689.282490@p10g2000cwp.googlegroups.com...
>
>> Stork replied to:
>
>>>When you place something under the GPL and you
>>>sell or transfer it to at least one person, you have effectively made
>>>everyone a legal licensee of the work
>
>> Nope, because, as a copyright owner, I have the right to determine who
>> can use my work or not.
>
>     That is correct.

Not quite.  The copyright owner, as the name indicates, controls the
right to make copies (and derivatives) of the work.  Once you have
acquired a copy by legal means, you may do with it whatever you see
fit.  The only thing you may not do is create and distribute more
copies (or derivative works).  There are no restrictions on use.

-- 
M�ns Rullg�rd
mru@inprovide.com
0
iso
2/28/2006 9:35:49 PM
"M�ns Rullg�rd" <mru@inprovide.com> wrote in message 
news:yw1xaccbtbd6.fsf@agrajag.inprovide.com...

> Not quite.  The copyright owner, as the name indicates, controls the
> right to make copies (and derivatives) of the work.  Once you have
> acquired a copy by legal means, you may do with it whatever you see
> fit.  The only thing you may not do is create and distribute more
> copies (or derivative works).  There are no restrictions on use.

    Anything that is necessary as part of the ordinary and reasonable use of 
a work is use. If the work is a dynamically linked library, and programs 
that use it are derivative works, then creating derivative works *is* part 
of the ordinary and reasonable use of the work, and so is legally considered 
use.

    There are two ways you can argue this:

    1) An application that uses a dynamically linked library is not a 
derivative work of the original work. In this case, the GPL does not require 
you to distribute the source code to the application if you distribute both 
the application and the library.

    2) An application that uses a dynamically linked library is a derivative 
work of the library. In this case, the creation of the application is part 
of the ordinary use of the library, and creating it does not invoke any GPL 
restrictions because the GPL only puts restrictions on rights that are not 
part of ordinary use.

    You can keep analyzing each case in detail, of course. And you can argue 
that in some cases applications are derivative works and in other cases they 
are not.

    First sale grants you the right to the oridinary and expected use of the 
covered work. If that ordinary and expected use requires the creation of 
derivative works, then you may do so. The right to use something includes 
the right to do anything necessary to use that something. I can cite quite a 
bit of case law on this if you don't believe me (but it should make perfect 
sense). If I grant you the right to drive my car, and driving my car 
requires you to insert the key in the ignition, then I have granted you the 
right to insert the key in the ignition even if that would normally be 
something you did not have the right to do.

    DS


0
David
2/28/2006 10:41:54 PM
"David Schwartz" <davids@webmaster.com> writes:

>     First sale grants you the right to the oridinary and expected use of the 
> covered work. If that ordinary and expected use requires the creation of 
> derivative works, then you may do so. The right to use something includes 
> the right to do anything necessary to use that something. I can cite quite a 
> bit of case law on this if you don't believe me (but it should make perfect 
> sense).

I believe you perfectly well.  It's that other fellow that doesn't.

-- 
M�ns Rullg�rd
mru@inprovide.com
0
iso
2/28/2006 11:48:25 PM
On 2006-02-28, stork <stork@storkyak.com> wrote:
> Nope, because, as a copyright owner, I have the right to determine who
> can use my work or not. 

Yes, and by releasing it under the GPL, you said "Hey, you can all use
this". And the GPL doesn't provide for any "takebacks".
0
Jordan
2/28/2006 11:51:15 PM
Stork replied to:

>An application that uses a dynamically linked library is not a
>derivative work of the original work. In this case, the GPL does not require
>you to distribute the source code to the application if you distribute both
>the application and the library.

If you use the LGPL, not the regular GPL.

>The right to use something includes
>the right to do anything necessary to use that something

"Anything necessary".  There are limits.

>If I grant you the right to drive my car, and driving my car
>requires you to insert the key in the ignition, then I have granted you the
>right to insert the key in the ignition

Ah, but, if I give you the right to drive my car, and driving the car
requires gasoline that I happen to sell, you don't have the right to
steal it in order to drive the car.

0
stork
3/1/2006 3:03:25 PM
On 2006-03-01, stork <stork@storkyak.com> wrote:
> Stork replied to:
>
>>An application that uses a dynamically linked library is not a
>>derivative work of the original work. In this case, the GPL does not require
>>you to distribute the source code to the application if you distribute both
>>the application and the library.
>
> If you use the LGPL, not the regular GPL.

Using one license or another doesn't change "An application that uses a
library is not a derived work of the library". You're right that the GPL
claims to require it, he's merely claiming that it has no legal leg to
stand on. (I'm not a lawyer, and he may in fact be wrong, but if he's
wrong then it applies to both licenses)
0
Jordan
3/1/2006 4:54:00 PM
Jordan Abel wrote:

> On 2006-03-01, stork <stork@storkyak.com> wrote:
>> Stork replied to:
>>
>>>An application that uses a dynamically linked library is not a
>>>derivative work of the original work. In this case, the GPL does not
>>>require you to distribute the source code to the application if you
>>>distribute both the application and the library.
>>
>> If you use the LGPL, not the regular GPL.
> 
> Using one license or another doesn't change "An application that uses a
> library is not a derived work of the library". 

You base that claim on what exactly?

< snip >
-- 
Who the fuck is General Failure, and why is he reading my harddisk?

0
Peter
3/1/2006 5:30:49 PM
On 2006-03-01, Peter K�hlmann <peter.koehlmann@t-online.de> wrote:
> Jordan Abel wrote:
>
>> On 2006-03-01, stork <stork@storkyak.com> wrote:
>>> Stork replied to:
>>>
>>>>An application that uses a dynamically linked library is not a
>>>>derivative work of the original work. In this case, the GPL does not
>>>>require you to distribute the source code to the application if you
>>>>distribute both the application and the library.
>>>
>>> If you use the LGPL, not the regular GPL.
>> 
>> Using one license or another doesn't change "An application that uses a
>> library is not a derived work of the library". 
>
> You base that claim on what exactly?

I didn't make the claim - I paraphrased the quoted claim [which you have
snipped the attribution on]. If it's true, it's equally true for the GPL
as it is for the LGPL.
0
Jordan
3/1/2006 6:27:55 PM
Stork replied to:

>Yes, and by releasing it under the GPL, you said "Hey, you can all use
>this". And the GPL doesn't provide for any "takebacks".

I say, by releasing something under the GPL, that you can all use this,
IF you follow the terms and release your own software GPL.  And, I can
just as easily say, I hereby revoke everyone's license to use my
copywrited stuff on Tuesdays... or only if you give me $2.37, It's -my-
stuff, not -yours-.  If I want to add something to the GPL, I can.  It
wouldn't be the "GPL", but, it's -my- stuff, and the GPL is nothing
more than a model that I use a template for my license.

0
stork
3/1/2006 7:27:31 PM
"stork" <stork@storkyak.com> wrote in message 
news:1141225402.099485.77770@i40g2000cwc.googlegroups.com...

> Stork replied to:

>>An application that uses a dynamically linked library is not a
>>derivative work of the original work. In this case, the GPL does not 
>>require
>>you to distribute the source code to the application if you distribute 
>>both
>>the application and the library.

> If you use the LGPL, not the regular GPL.

    Do you even read what I write? If it's not a derivative work, the GPL 
doesn't apply to it. See section 0 of the GPL.

>>The right to use something includes
>>the right to do anything necessary to use that something
>
> "Anything necessary".  There are limits.

    That's a nonsensical comment.

>>If I grant you the right to drive my car, and driving my car
>>requires you to insert the key in the ignition, then I have granted you 
>>the
>>right to insert the key in the ignition

> Ah, but, if I give you the right to drive my car, and driving the car
> requires gasoline that I happen to sell, you don't have the right to
> steal it in order to drive the car.

    You are the king of the meaningless analogy. A more accurate analogy 
would be that if you give me permission to use your car, I can "steal" the 
minimum gas necessary to get the car to a gas station. It's not stealing, 
permission to use a car is implicitly permission to do anything *necessary* 
for the ordinary, reasonable, and expected use of the car.

    I'm not sure why you're bothering to reply if you aren't going to 
include any substance at all. It's starting to look like trolling. Post a 
few "no" responses and see how long a response you can get from me. Sounds 
like fun.

    DS


0
David
3/1/2006 10:34:56 PM
"Peter K�hlmann" <peter.koehlmann@t-online.de> wrote in message 
news:du4lp7$mp8$01$1@news.t-online.com...

> Jordan Abel wrote:

>> Using one license or another doesn't change "An application that uses a
>> library is not a derived work of the library".

> You base that claim on what exactly?

    Obvious common sense.

    For example, suppose you write a program that dynamically links to 
"libc". I write a brand new clean room implementation of "libc" and run your 
program, allowing it to link to my "libc". Your program does not contain any 
of my "libc", how could it, it was written *first*. Yet the idea that an 
application that uses a dynamically linked library is a derivative work of 
that library would make your application a derivative work from my library. 
Even though they were developed independently. Even though they have no 
common code. Even though the alleged derivative was created first.

    There are other more technical reasons, but this one is the most obvious 
one.

    DS


0
David
3/1/2006 10:37:04 PM
"stork" <stork@storkyak.com> wrote in message 
news:1141241251.888514.10760@j33g2000cwa.googlegroups.com...

> Stork replied to:

>>Yes, and by releasing it under the GPL, you said "Hey, you can all use
>>this". And the GPL doesn't provide for any "takebacks".

> I say, by releasing something under the GPL, that you can all use this,
> IF you follow the terms and release your own software GPL.

    THAT IS NOT WHAT THE GPL SAYS. If you think that's what the GPL says, 
you are terribly mistaken. Perhaps that's the source of your 
misunderstanding -- you think the GPL is a EULA or a shrink-wrap agreement.

> And, I can
> just as easily say, I hereby revoke everyone's license to use my
> copywrited stuff on Tuesdays... or only if you give me $2.37, It's -my-
> stuff, not -yours-.  If I want to add something to the GPL, I can.  It
> wouldn't be the "GPL", but, it's -my- stuff, and the GPL is nothing
> more than a model that I use a template for my license.

    Actually, that's not true. The GPL itself is copyrighted and the license 
agreement does not permit you to modify it.

    But this thread has actually not been about the GPL. It's really about 
first sale rights. If what you said was correct, then a person could post 
their poem on a billboard and then sue everyone who read that poem. But of 
course, you're not right, and everything you've said is just your naive view 
about how the world should work and has little to nothing to do with how it 
does work.

    If you give away copies of a copyrighted work without a EULA, 
shrink-wrap agreement, click-through agreement, or signed contract, then you 
have also given away certain rights to the work. This includes the right to 
the ordinary, necessary, expected use of that work.

    DS


0
David
3/1/2006 10:40:07 PM
David Schwartz wrote:

> 
> "Peter K�hlmann" <peter.koehlmann@t-online.de> wrote in message
> news:du4lp7$mp8$01$1@news.t-online.com...
> 
>> Jordan Abel wrote:
> 
>>> Using one license or another doesn't change "An application that uses a
>>> library is not a derived work of the library".
> 
>> You base that claim on what exactly?
> 
>     Obvious common sense.
> 
>     For example, suppose you write a program that dynamically links to
> "libc". I write a brand new clean room implementation of "libc" and run
> your program, allowing it to link to my "libc". Your program does not
> contain any of my "libc", how could it, it was written *first*. Yet the
> idea that an application that uses a dynamically linked library is a
> derivative work of that library would make your application a derivative
> work from my library. Even though they were developed independently. Even
> though they have no common code. Even though the alleged derivative was
> created first.
> 
>     There are other more technical reasons, but this one is the most
>     obvious
> one.
> 
>     DS

You are clearly the king of bullshit claims
-- 
You're not my type.  For that matter, you're not even my species

0
Peter
3/1/2006 11:11:16 PM
Peter K�hlmann <peter.koehlmann@t-online.de> writes:

> David Schwartz wrote:
>
>> 
>> "Peter K?ann" <peter.koehlmann@t-online.de> wrote in message
>> news:du4lp7$mp8$01$1@news.t-online.com...
>> 
>>> Jordan Abel wrote:
>> 
>>>> Using one license or another doesn't change "An application that uses a
>>>> library is not a derived work of the library".
>> 
>>> You base that claim on what exactly?
>> 
>>     Obvious common sense.
>> 
>>     For example, suppose you write a program that dynamically links to
>> "libc". I write a brand new clean room implementation of "libc" and run
>> your program, allowing it to link to my "libc". Your program does not
>> contain any of my "libc", how could it, it was written *first*. Yet the
>> idea that an application that uses a dynamically linked library is a
>> derivative work of that library would make your application a derivative
>> work from my library. Even though they were developed independently. Even
>> though they have no common code. Even though the alleged derivative was
>> created first.
>> 
>>     There are other more technical reasons, but this one is the most
>>     obvious
>> one.
>> 
>>     DS
>
> You are clearly the king of bullshit claims

Care to elaborate?  I think he's making perfect sense.

-- 
M�ns Rullg�rd
mru@inprovide.com
0
iso
3/2/2006 12:47:27 AM
On 2006-03-01, stork <stork@storkyak.com> wrote:
> Stork replied to:
>
>>Yes, and by releasing it under the GPL, you said "Hey, you can all use
>>this". And the GPL doesn't provide for any "takebacks".
>
> I say, by releasing something under the GPL, that you can all use this,
> IF you follow the terms and release your own software GPL. 

Well... sort of.

For one thing, they don't have to release their software under the GPL
if their software is not derived from yours.  For another, the text of
the GPL contains a declaration of belief that no license is required to
use software that is legitimately required, and it's arguable that since
that statement implies a specific intent, that even if such a license
_IS_ required, the statement constitutes an unrestricted nonrevocable
license for use. (IANAL)

> And, I can just as easily say, I hereby revoke everyone's license to
> use my copywrited stuff on Tuesdays... or only if you give me $2.37,
> It's -my- stuff, not -yours-.  If I want to add something to the GPL,
> I can.  It wouldn't be the "GPL", but, it's -my- stuff, and the GPL is
> nothing more than a model that I use a template for my license.
>
0
Jordan
3/2/2006 12:52:36 AM
Måns Rullgård wrote:

> Peter Köhlmann <peter.koehlmann@t-online.de> writes:
> 
>> David Schwartz wrote:
>>
>>> 
>>> "Peter K?ann" <peter.koehlmann@t-online.de> wrote in message
>>> news:du4lp7$mp8$01$1@news.t-online.com...
>>> 
>>>> Jordan Abel wrote:
>>> 
>>>>> Using one license or another doesn't change "An application that uses
>>>>> a library is not a derived work of the library".
>>> 
>>>> You base that claim on what exactly?
>>> 
>>>     Obvious common sense.
>>> 
>>>     For example, suppose you write a program that dynamically links to
>>> "libc". I write a brand new clean room implementation of "libc" and run
>>> your program, allowing it to link to my "libc". Your program does not
>>> contain any of my "libc", how could it, it was written *first*. Yet the
>>> idea that an application that uses a dynamically linked library is a
>>> derivative work of that library would make your application a derivative
>>> work from my library. Even though they were developed independently.
>>> Even though they have no common code. Even though the alleged derivative
>>> was created first.
>>> 
>>>     There are other more technical reasons, but this one is the most
>>>     obvious
>>> one.
>>> 
>>>     DS
>>
>> You are clearly the king of bullshit claims
> 
> Care to elaborate?  I think he's making perfect sense.
> 

Really? You think that "deriving" an app from some library and then
reimplementing that library should now make sense when claiming "derivative
work" of the later library? Why not just "derivative work" of the first
library and be done with it?
How about some even more convoluted claim?
-- 
Experience is what causes a person to make new mistakes instead of 
old ones.

0
Peter
3/2/2006 1:21:28 AM
Peter K�hlmann <peter.koehlmann@t-online.de> writes:

> M�ns Rullg�rd wrote:
>
>> Peter K�hlmann <peter.koehlmann@t-online.de> writes:
>> 
>>> David Schwartz wrote:
>>>
>>>> 
>>>> "Peter K?ann" <peter.koehlmann@t-online.de> wrote in message
>>>> news:du4lp7$mp8$01$1@news.t-online.com...
>>>> 
>>>>> Jordan Abel wrote:
>>>> 
>>>>>> Using one license or another doesn't change "An application that uses
>>>>>> a library is not a derived work of the library".
>>>> 
>>>>> You base that claim on what exactly?
>>>> 
>>>>     Obvious common sense.
>>>> 
>>>>     For example, suppose you write a program that dynamically links to
>>>> "libc". I write a brand new clean room implementation of "libc" and run
>>>> your program, allowing it to link to my "libc". Your program does not
>>>> contain any of my "libc", how could it, it was written *first*. Yet the
>>>> idea that an application that uses a dynamically linked library is a
>>>> derivative work of that library would make your application a derivative
>>>> work from my library. Even though they were developed independently.
>>>> Even though they have no common code. Even though the alleged derivative
>>>> was created first.
>>>> 
>>>>     There are other more technical reasons, but this one is the most
>>>>     obvious
>>>> one.
>>>> 
>>>>     DS
>>>
>>> You are clearly the king of bullshit claims
>> 
>> Care to elaborate?  I think he's making perfect sense.
>> 
>
> Really? You think that "deriving" an app from some library and then
> reimplementing that library should now make sense when claiming "derivative
> work" of the later library? Why not just "derivative work" of the first
> library and be done with it?

David is saying that if linking to a library creates a work derived
from that library, that would mean that a program could be derived
from a library that wasn't written at the time the program was
created.  That would obviously be absurd, so linking to a library
cannot by itself create a derived work.  Do you still disagree?

> How about some even more convoluted claim?

I can't see how "dynamic linking doesn't automatically create a
derived work" can be considered a convoluted claim.

-- 
M�ns Rullg�rd
mru@inprovide.com
0
iso
3/2/2006 2:11:40 AM
Måns Rullgård wrote:

> Peter Köhlmann <peter.koehlmann@t-online.de> writes:
> 
>> Måns Rullgård wrote:
>>
>>> Peter Köhlmann <peter.koehlmann@t-online.de> writes:
>>> 
>>>> David Schwartz wrote:
>>>>
>>>>> 
>>>>> "Peter K?ann" <peter.koehlmann@t-online.de> wrote in message
>>>>> news:du4lp7$mp8$01$1@news.t-online.com...
>>>>> 
>>>>>> Jordan Abel wrote:
>>>>> 
>>>>>>> Using one license or another doesn't change "An application that
>>>>>>> uses a library is not a derived work of the library".
>>>>> 
>>>>>> You base that claim on what exactly?
>>>>> 
>>>>>     Obvious common sense.
>>>>> 
>>>>>     For example, suppose you write a program that dynamically links to
>>>>> "libc". I write a brand new clean room implementation of "libc" and
>>>>> run your program, allowing it to link to my "libc". Your program does
>>>>> not contain any of my "libc", how could it, it was written *first*.
>>>>> Yet the idea that an application that uses a dynamically linked
>>>>> library is a derivative work of that library would make your
>>>>> application a derivative work from my library. Even though they were
>>>>> developed independently. Even though they have no common code. Even
>>>>> though the alleged derivative was created first.
>>>>> 
>>>>>     There are other more technical reasons, but this one is the most
>>>>>     obvious
>>>>> one.
>>>>> 
>>>>>     DS
>>>>
>>>> You are clearly the king of bullshit claims
>>> 
>>> Care to elaborate?  I think he's making perfect sense.
>>> 
>>
>> Really? You think that "deriving" an app from some library and then
>> reimplementing that library should now make sense when claiming
>> "derivative work" of the later library? Why not just "derivative work" of
>> the first library and be done with it?
> 
> David is saying that if linking to a library creates a work derived
> from that library, that would mean that a program could be derived
> from a library that wasn't written at the time the program was
> created.  That would obviously be absurd, so linking to a library
> cannot by itself create a derived work.  Do you still disagree?
> 

Yes

>> How about some even more convoluted claim?
> 
> I can't see how "dynamic linking doesn't automatically create a
> derived work" can be considered a convoluted claim.
> 

Fine. Look closer
-- 
Individualists unite!

0
Peter
3/2/2006 2:20:54 AM
"Peter K�hlmann" <peter.koehlmann@t-online.de> wrote in message 
news:du5hbm$13c$03$1@news.t-online.com...

> Really? You think that "deriving" an app from some library and then
> reimplementing that library should now make sense when claiming 
> "derivative
> work" of the later library? Why not just "derivative work" of the first
> library and be done with it?
> How about some even more convoluted claim?

    I honestly can't figure out what you're trying to say here. I've really 
made an honest effort. It may be that you don't understand that the term 
"derivative work" has a precise technical meaning. It may be something else.

    In any event, that was just selected as the easiest-to-understand 
reason. One work cannot contain the creative elements of another if the two 
works were developed entirely independently. If that's not common sense, I 
don't know what is.

    There are other reasons as well, and in the post where I cite from the 
Lexmark case, I go into detail about another one.

    Note that I am not saying that it is impossible for an application to be 
a derivative work of a library it dynamically links to. I am saying that the 
argument that it must always be the case fails the giggle test.

    DS


0
David
3/2/2006 3:06:10 AM
Peter K�hlmann <peter.koehlmann@t-online.de> writes:

> M�ns Rullg�rd wrote:
>
>> Peter K�hlmann <peter.koehlmann@t-online.de> writes:
>> 
>>> M�ns Rullg�rd wrote:
>>>
>>>> Peter K�hlmann <peter.koehlmann@t-online.de> writes:
>>>> 
>>>>> David Schwartz wrote:
>>>>>
>>>>>> 
>>>>>> "Peter K?ann" <peter.koehlmann@t-online.de> wrote in message
>>>>>> news:du4lp7$mp8$01$1@news.t-online.com...
>>>>>> 
>>>>>>> Jordan Abel wrote:
>>>>>> 
>>>>>>>> Using one license or another doesn't change "An application that
>>>>>>>> uses a library is not a derived work of the library".
>>>>>> 
>>>>>>> You base that claim on what exactly?
>>>>>> 
>>>>>>     Obvious common sense.
>>>>>> 
>>>>>>     For example, suppose you write a program that dynamically links to
>>>>>> "libc". I write a brand new clean room implementation of "libc" and
>>>>>> run your program, allowing it to link to my "libc". Your program does
>>>>>> not contain any of my "libc", how could it, it was written *first*.
>>>>>> Yet the idea that an application that uses a dynamically linked
>>>>>> library is a derivative work of that library would make your
>>>>>> application a derivative work from my library. Even though they were
>>>>>> developed independently. Even though they have no common code. Even
>>>>>> though the alleged derivative was created first.
>>>>>> 
>>>>>>     There are other more technical reasons, but this one is the most
>>>>>>     obvious
>>>>>> one.
>>>>>> 
>>>>>>     DS
>>>>>
>>>>> You are clearly the king of bullshit claims
>>>> 
>>>> Care to elaborate?  I think he's making perfect sense.
>>>> 
>>>
>>> Really? You think that "deriving" an app from some library and then
>>> reimplementing that library should now make sense when claiming
>>> "derivative work" of the later library? Why not just "derivative work" of
>>> the first library and be done with it?
>> 
>> David is saying that if linking to a library creates a work derived
>> from that library, that would mean that a program could be derived
>> from a library that wasn't written at the time the program was
>> created.  That would obviously be absurd, so linking to a library
>> cannot by itself create a derived work.  Do you still disagree?
>
> Yes

Consider these two events happening, in this order:

1. I write a program, P,  that uses some interface, I.
2. You write a library, L, implementing I.

Are you trying to say that P is derived from L?

>>> How about some even more convoluted claim?
>> 
>> I can't see how "dynamic linking doesn't automatically create a
>> derived work" can be considered a convoluted claim.
>
> Fine. Look closer

Does "convoluted" mean "I don't like it" in German?

-- 
M�ns Rullg�rd
mru@inprovide.com
0
iso
3/2/2006 3:06:29 AM
"Peter K�hlmann" <peter.koehlmann@t-online.de> wrote in message 
news:du5kr5$8p0$03$2@news.t-online.com...

>> David is saying that if linking to a library creates a work derived
>> from that library, that would mean that a program could be derived
>> from a library that wasn't written at the time the program was
>> created.  That would obviously be absurd, so linking to a library
>> cannot by itself create a derived work.  Do you still disagree?

> Yes

    Simply saying you don't agree without showing why isn't particularly 
helpful. It's worse than the AOL "me too". When presented with a clear and 
concise rational argument, responding with "I don't agree" is breaking the 
rules of reasonable debate.

    If you are not capable of rational debate, that's fine, I'll just say 
that everytime you demonstrate it. If you are capable of rational debate, 
please follow the rules that everyone else follows.

    DS


0
David
3/2/2006 3:08:06 AM
"M�ns Rullg�rd" <mru@inprovide.com> wrote in message 
news:yw1x7j7dr3xf.fsf@agrajag.inprovide.com...

> David is saying that if linking to a library creates a work derived
> from that library, that would mean that a program could be derived
> from a library that wasn't written at the time the program was
> created.  That would obviously be absurd, so linking to a library
> cannot by itself create a derived work.  Do you still disagree?

    Thank you for explaining my argument better than I did. You do have one 
minor technical error though. One could argue that linking to a library by 
itself creates a derived work -- the application linked to the library in 
the computer's memory. What I am saying is that it can't make the 
application *itself* a derivative work.

    I would change your phrase "so linking to a library cannot by itself 
create a derived work" to "so that an application can link to a library does 
not by itself make the application a derivative work of the library".

    Note that I disagree with the argument that the program in memory is a 
derivative work, but the argument you're summarizing is not sufficient to 
demolish it.

    For the record, I would argue that with only those exceptions specified 
in the law (such as translation), only *creative* combination can create a 
derivative work, automatic combination creates a mere aggregate. Second, the 
in-use copy of a program in a computer's memory is not sufficiently 
permanent to count as a "work". (Just as the reflection of a book on your 
retinas when you read it is not legally a "copy" of the work.) Third, 
anything necessary for use is use, and you can't use a computer program 
without loading it into memory.

    DS


0
David
3/2/2006 3:28:06 AM
Stork replied to:

>    Do you even read what I write? If it's not a derivative work, the GPL
>doesn't apply to it. See section 0 of the GPL.

This License applies to any program or other work which contains a
notice placed by the copyright holder saying it may be distributed
under the terms of this General Public License. The "Program", below,
refers to any such program or work, and a "work based on the Program"
means either the Program or any derivative work under copyright law:
that is to say, a work containing the Program or a portion of it,
either verbatim or with modifications and/or translated into another
language. (Hereinafter, translation is included without limitation in
the term "modification".) Each licensee is addressed as "you".

Again, the argument that FSF makes is that dynamic linking is a
derivative work.  You keep blurring the argument.  The GPL clearly
intends to apply to libraries and programs that use them must also be
GPL, and if you don't want that, use the GPL.  Whether or not that this
is legally actionable is another matter.  The INTENT of the GPL is to
make every program that uses a GPL work also GPL, and that is what you
are arguing against.

>You are the king of the meaningless analogy.

Your analogy is totally wrong, not mine.  The issue is one of
ownership.  You imply that if I give you the car to use, that I
automatically give you ownership of -everything- related to the car.

>It's starting to look like trolling

You're the one that is trolling.  I'm just the one seeing through your
FUD. Fundamentally, all you are trying to do is argue that you don't
have to GPL your work if you use something else that is GPL, and that's
just not the case if you go by the spirit of the license.  If you want
to go and try and lawyer your way out of it, you can, but that's
certainly a bad faith negotiation in its own right.

0
stork
3/2/2006 4:39:36 AM
Stork replied to:

>If what you said was correct, then a person could post
>their poem on a billboard and then sue everyone who read that poem

No, you are shifting the argument.  You argue that if a person posts a
poem on a bulletin board, that you can sell whatever is on that board
yourself. To wit, go ahead and argue that you can sell NFL logo'd
t-shirts because someone gave one to you.

0
stork
3/2/2006 4:42:03 AM
Stork replied to:

>If you give away copies of a copyrighted work without a EULA,
>shrink-wrap agreement, click-through agreement, or signed contract

I'm assuming that my work would have the GPL used as a EULA in the
above fashion, in which case, its a contract for you to use my stuff,
and, doing anything else is a breach.

>Actually, that's not true. The GPL itself is copyrighted and the license
>agreement does not permit you to modify it.

As you would say, the GPL is an artistic expression.  I can create a
license with the same effect as the GPL and with my own terms...so, the
point is actually academic.

> If what you said was correct, then a person could post
>their poem on a billboard and then sue everyone who read that poem

No, what I'm saying is, if someone puts their poem on a bulletin board,
a driver could not copy that poem down and resell it, and that most
certainly is the case.

0
stork
3/2/2006 4:46:19 AM
Stork replied to:

>David is saying that if linking to a library creates a work derived
>from that library, that would mean that a program could be derived
>from a library that wasn't written at the time the program was
>created.

Yes, but the problem with David's argument is that he then uses this to
justify stealing code from a library that most certainly exists at the
time his work was created.  In other words, he's arguing a situation
that doesn't apply to what he's trying to do.  If, you accept his
premise that the existence of the library is a prerequisite to creating
a derived work, then, the correct logical conclusion is that because
the header files he wants to copy from exists, then, he is creating a
derived work.  It's almost like he's saying that given A->B, then
!A->B, which makes no sense whatsoever.

He's trolling, and a damned good one.  My hat goes off to him.

>That would obviously be absurd, so linking to a library
>cannot by itself create a derived work.  Do you still disagree?

Yes, see above.

0
stork
3/2/2006 4:52:06 AM
"stork" <stork@storkyak.com> wrote in message 
news:1141274376.954786.226280@z34g2000cwc.googlegroups.com...

> Again, the argument that FSF makes is that dynamic linking is a
> derivative work.  You keep blurring the argument.

    I'm not "blurring" it. I've demonstrated that it's obviously false.

> The GPL clearly
> intends to apply to libraries and programs that use them must also be
> GPL, and if you don't want that, use the GPL.

    That doesn't matter because the GPL cannot set its own scope.

>Whether or not that this
> is legally actionable is another matter.  The INTENT of the GPL is to
> make every program that uses a GPL work also GPL, and that is what you
> are arguing against.

    No, that's not the intent of the GPL. If the intent was to do so, the 
GPL would have been worded to do that, rather than clearly stating that it 
puts no restrictions on use. See the second half of section zero.

>>You are the king of the meaningless analogy.

> Your analogy is totally wrong, not mine.  The issue is one of
> ownership.  You imply that if I give you the car to use, that I
> automatically give you ownership of -everything- related to the car.

    Why does it matter what I imply? Why not address what I actually *said*? 
I'm not being vague or indirect. If I implied something you don't agree 
with, wonderful, but what about what I actually *said*? Are you just going 
to totally ignore that?

>>It's starting to look like trolling

> You're the one that is trolling.  I'm just the one seeing through your
> FUD. Fundamentally, all you are trying to do is argue that you don't
> have to GPL your work if you use something else that is GPL, and that's
> just not the case if you go by the spirit of the license.  If you want
> to go and try and lawyer your way out of it, you can, but that's
> certainly a bad faith negotiation in its own right.

    If you want to refute my arguments, that's fine. But summarizing them 
badly isn't an argument.

    DS


0
David
3/2/2006 4:58:22 AM
"stork" <stork@storkyak.com> wrote in message 
news:1141274523.223213.235890@z34g2000cwc.googlegroups.com...

> Stork replied to:

>>If what you said was correct, then a person could post
>>their poem on a billboard and then sue everyone who read that poem

> No, you are shifting the argument.  You argue that if a person posts a
> poem on a bulletin board, that you can sell whatever is on that board
> yourself.

    Umm, no. Nobody is talking about selling the work that is GPL'd.

> To wit, go ahead and argue that you can sell NFL logo'd
> t-shirts because someone gave one to you.

    Again, no. Nobody is talking about transferring a copy of a work to 
someone who is not already a licensee.

    Why not stick to what I actually said instead of summarizing it badly?

    DS


0
David
3/2/2006 4:59:30 AM
"stork" <stork@storkyak.com> wrote in message 
news:1141274779.059581.123490@j33g2000cwa.googlegroups.com...

> Stork replied to:

>>If you give away copies of a copyrighted work without a EULA,
>>shrink-wrap agreement, click-through agreement, or signed contract

> I'm assuming that my work would have the GPL used as a EULA in the
> above fashion, in which case, its a contract for you to use my stuff,
> and, doing anything else is a breach.

    The GPL is not an EULA.

>>Actually, that's not true. The GPL itself is copyrighted and the license
>>agreement does not permit you to modify it.

> As you would say, the GPL is an artistic expression.  I can create a
> license with the same effect as the GPL and with my own terms...so, the
> point is actually academic.

    No, it's not academic. We are talking strictly about works that are 
subject to the GPL. If you want to talk about something else, that's fine, 
but since we have genuine disagreements about GPL'd works, why can't we 
stick to that?

>> If what you said was correct, then a person could post
>>their poem on a billboard and then sue everyone who read that poem

> No, what I'm saying is, if someone puts their poem on a bulletin board,
> a driver could not copy that poem down and resell it, and that most
> certainly is the case.

    I agree. He would only be entitled to ordinary use. However, nobody is 
talking about selling a work that is covered by the GPL. So I can't figure 
out for the life of my why you bring that up.

    DS


0
David
3/2/2006 5:00:54 AM
"stork" <stork@storkyak.com> wrote in message 
news:1141275125.973345.5560@t39g2000cwt.googlegroups.com...

> Stork replied to:

>>David is saying that if linking to a library creates a work derived
>>from that library, that would mean that a program could be derived
>>from a library that wasn't written at the time the program was
>>created.

> Yes, but the problem with David's argument is that he then uses this to

    You are ignoring the rules of rational debate. You cannot disagree with 
a reasoned argument because you don't like what I'm going to do with it 
later. You have to reserve that disagreement for the later part.

> If, you accept his
> premise that the existence of the library is a prerequisite to creating
> a derived work, then, the correct logical conclusion is that because
> the header files he wants to copy from exists, then, he is creating a
> derived work.

    What?! It is a prerequisite to you driving my car that I own a car. 
Therefore if I own a car you are driving it?

> He's trolling, and a damned good one.  My hat goes off to him.

    Pot. Kettle. Black.

    DS


0
David
3/2/2006 5:02:45 AM
Peter K�hlmann <peter.koehlmann@t-online.de> writes:
> >>
> >> You are clearly the king of bullshit claims
> > 
> > Care to elaborate?  I think he's making perfect sense.
> > 
> 
> Really? You think that "deriving" an app from some library and then
> reimplementing that library should now make sense when claiming "derivative
> work" of the later library? Why not just "derivative work" of the first
> library and be done with it?
> How about some even more convoluted claim?

He's making a proof by contradiction, to demonstrate that the GPL's
claim that a program linked with glibc is a derived work is absurd.
You appear to be agreeing with him, by agreeing the result is absurd.
-- 
Joseph J. Pfeiffer, Jr., Ph.D.       Phone -- (505) 646-1605
Department of Computer Science       FAX   -- (505) 646-1002
New Mexico State University          http://www.cs.nmsu.edu/~pfeiffer
0
Joe
3/2/2006 5:08:53 AM
"David Schwartz" <davids@webmaster.com> writes:

> "M�ns Rullg�rd" <mru@inprovide.com> wrote in message 
> news:yw1x7j7dr3xf.fsf@agrajag.inprovide.com...
>
>> David is saying that if linking to a library creates a work derived
>> from that library, that would mean that a program could be derived
>> from a library that wasn't written at the time the program was
>> created.  That would obviously be absurd, so linking to a library
>> cannot by itself create a derived work.  Do you still disagree?
>
>     Thank you for explaining my argument better than I did. You do have one 
> minor technical error though. One could argue that linking to a library by 
> itself creates a derived work -- the application linked to the library in 
> the computer's memory. What I am saying is that it can't make the 
> application *itself* a derivative work.
>
>     I would change your phrase "so linking to a library cannot by itself 
> create a derived work" to "so that an application can link to a library does 
> not by itself make the application a derivative work of the library".

That's even better.

>     Note that I disagree with the argument that the program in memory is a 
> derivative work, but the argument you're summarizing is not sufficient to 
> demolish it.
>
>     For the record, I would argue that with only those exceptions specified 
> in the law (such as translation), only *creative* combination can create a 
> derivative work, automatic combination creates a mere aggregate. Second, the 
> in-use copy of a program in a computer's memory is not sufficiently 
> permanent to count as a "work". (Just as the reflection of a book on your 
> retinas when you read it is not legally a "copy" of the work.) Third, 
> anything necessary for use is use, and you can't use a computer program 
> without loading it into memory.

IIRC, some countries copyright law even makes the right to copy a
computer program to memory for execution explicit.

-- 
M�ns Rullg�rd
mru@inprovide.com
0
iso
3/2/2006 8:38:20 AM
Måns Rullgård wrote:

> Peter Köhlmann <peter.koehlmann@t-online.de> writes:
> 
>> Måns Rullgård wrote:
>>
>>> Peter Köhlmann <peter.koehlmann@t-online.de> writes:
>>> 
>>>> Måns Rullgård wrote:
>>>>
>>>>> Peter Köhlmann <peter.koehlmann@t-online.de> writes:
>>>>> 
>>>>>> David Schwartz wrote:
>>>>>>
>>>>>>> 
>>>>>>> "Peter K?ann" <peter.koehlmann@t-online.de> wrote in message
>>>>>>> news:du4lp7$mp8$01$1@news.t-online.com...
>>>>>>> 
>>>>>>>> Jordan Abel wrote:
>>>>>>> 
>>>>>>>>> Using one license or another doesn't change "An application that
>>>>>>>>> uses a library is not a derived work of the library".
>>>>>>> 
>>>>>>>> You base that claim on what exactly?
>>>>>>> 
>>>>>>>     Obvious common sense.
>>>>>>> 
>>>>>>>     For example, suppose you write a program that dynamically links
>>>>>>>     to
>>>>>>> "libc". I write a brand new clean room implementation of "libc" and
>>>>>>> run your program, allowing it to link to my "libc". Your program
>>>>>>> does not contain any of my "libc", how could it, it was written
>>>>>>> *first*. Yet the idea that an application that uses a dynamically
>>>>>>> linked library is a derivative work of that library would make your
>>>>>>> application a derivative work from my library. Even though they were
>>>>>>> developed independently. Even though they have no common code. Even
>>>>>>> though the alleged derivative was created first.
>>>>>>> 
>>>>>>>     There are other more technical reasons, but this one is the most
>>>>>>>     obvious
>>>>>>> one.
>>>>>>> 
>>>>>>>     DS
>>>>>>
>>>>>> You are clearly the king of bullshit claims
>>>>> 
>>>>> Care to elaborate?  I think he's making perfect sense.
>>>>> 
>>>>
>>>> Really? You think that "deriving" an app from some library and then
>>>> reimplementing that library should now make sense when claiming
>>>> "derivative work" of the later library? Why not just "derivative work"
>>>> of the first library and be done with it?
>>> 
>>> David is saying that if linking to a library creates a work derived
>>> from that library, that would mean that a program could be derived
>>> from a library that wasn't written at the time the program was
>>> created.  That would obviously be absurd, so linking to a library
>>> cannot by itself create a derived work.  Do you still disagree?
>>
>> Yes
> 
> Consider these two events happening, in this order:
> 

I did consider it

> 1. I write a program, P,  that uses some interface, I.
> 2. You write a library, L, implementing I.
> 
> Are you trying to say that P is derived from L?
> 

No. It is derived from I
L is totally irrelevant to the situation
And by changing the wording to "that uses some interface" instead of
"derived from" you show that your only intention is to be dishonest
P is either derived from I, in which case the GPL applies, or it is not. In
which case this absurd "discussion" is even more absurd

>>>> How about some even more convoluted claim?
>>> 
>>> I can't see how "dynamic linking doesn't automatically create a
>>> derived work" can be considered a convoluted claim.
>>
>> Fine. Look closer
> 
> Does "convoluted" mean "I don't like it" in German?
> 

No. It means you are busyly setting up hordes of strawmen to knock down
-- 
Another name for a Windows tutorial is crash course

0
Peter
3/2/2006 9:14:09 AM
"Peter K�hlmann" <peter.koehlmann@t-online.de> wrote in message 
news:du6d1u$umf$03$1@news.t-online.com...

> I did consider it
>
>> 1. I write a program, P,  that uses some interface, I.
>> 2. You write a library, L, implementing I.
>>
>> Are you trying to say that P is derived from L?
>>
>
> No. It is derived from I

    It may be "derived from" in the normal English usage of those words, but 
is not (necessarily) a derivative work of for copyright purposes.

    I, the interface, probably does not exist as a distinct protectable 
work. It is an idea, not the expression of an idea.

    The proof is that this same interface can be expressed in P and 
expressed in L. Each of which are distinct (and presumably created and 
copyrightable) expressions of, among other things, the same idea I.

    DS


0
David
3/2/2006 7:19:07 PM
Peter K�hlmann <peter.koehlmann@t-online.de> writes:

> M�ns Rullg�rd wrote:
>
>> Consider these two events happening, in this order:
>> 
>
> I did consider it
>
>> 1. I write a program, P,  that uses some interface, I.
>> 2. You write a library, L, implementing I.
>> 
>> Are you trying to say that P is derived from L?
>
> No. It is derived from I

Correct.

> L is totally irrelevant to the situation

L is the only thing we need to determine if P is derived from.

> And by changing the wording to "that uses some interface" instead of
> "derived from" you show that your only intention is to be dishonest

I trying to use as precise a wording as I can.

> P is either derived from I, in which case the GPL applies, or it is
> not. In which case this absurd "discussion" is even more absurd

Suppose I is the standard C API as published by ISO or the Java class
library specification published by Sun.  Neither of these places any
restrictions on applications using them.

In the first case, L would be a C library implementation, e.g. BSD
libc, Sun libc, or glibc.  If P is derived from any of these
implementation, it must be derived from all of them, since it will
work equally well with either.  In the second case, L would be any
Java class library implementation, e.g. Sun's own, IBM's, or GNU
Classpath.  Again, an application that uses the Java class library
interface will work with any implementation.  The license of an
individual implementation cannot possibly make any difference
whatsoever.  Are you still pretending not to understand?

>>>>> How about some even more convoluted claim?
>>>> 
>>>> I can't see how "dynamic linking doesn't automatically create a
>>>> derived work" can be considered a convoluted claim.
>>>
>>> Fine. Look closer
>> 
>> Does "convoluted" mean "I don't like it" in German?
>
> No. It means you are busyly setting up hordes of strawmen to knock down

Hmm, I should probably get out those German textbooks again.

-- 
M�ns Rullg�rd
mru@inprovide.com
0
iso
3/2/2006 7:29:29 PM
David Schwartz wrote:

> 
> "Peter K�hlmann" <peter.koehlmann@t-online.de> wrote in message
> news:du6d1u$umf$03$1@news.t-online.com...
> 
>> I did consider it
>>
>>> 1. I write a program, P,  that uses some interface, I.
>>> 2. You write a library, L, implementing I.
>>>
>>> Are you trying to say that P is derived from L?
>>>
>>
>> No. It is derived from I
> 
>     It may be "derived from" in the normal English usage of those words,
>     but
> is not (necessarily) a derivative work of for copyright purposes.
> 
>     I, the interface, probably does not exist as a distinct protectable
> work. It is an idea, not the expression of an idea.
> 
>     The proof is that this same interface can be expressed in P and
> expressed in L. Each of which are distinct (and presumably created and
> copyrightable) expressions of, among other things, the same idea I.
> 
>     DS

And all that is still meaningless in the context of the discussion
-- 
Warning: You have moved the mouse. 
Windows will reboot now to make the change permanent

0
Peter
3/2/2006 7:55:58 PM
On 2006-03-02, M�ns Rullg�rd <mru@inprovide.com> wrote:
> "David Schwartz" <davids@webmaster.com> writes:
>
>> "M�ns Rullg�rd" <mru@inprovide.com> wrote in message 
>> news:yw1x7j7dr3xf.fsf@agrajag.inprovide.com...
>>
>>> David is saying that if linking to a library creates a work derived
>>> from that library, that would mean that a program could be derived
>>> from a library that wasn't written at the time the program was
>>> created.  That would obviously be absurd, so linking to a library
>>> cannot by itself create a derived work.  Do you still disagree?
>>
>>     Thank you for explaining my argument better than I did. You do have one 
>> minor technical error though. One could argue that linking to a library by 
>> itself creates a derived work -- the application linked to the library in 
>> the computer's memory. What I am saying is that it can't make the 
>> application *itself* a derivative work.
>>
>>     I would change your phrase "so linking to a library cannot by itself 
>> create a derived work" to "so that an application can link to a library does 
>> not by itself make the application a derivative work of the library".
>
> That's even better.
>
>>     Note that I disagree with the argument that the program in memory is a 
>> derivative work, but the argument you're summarizing is not sufficient to 
>> demolish it.
>>
>>     For the record, I would argue that with only those exceptions specified 
>> in the law (such as translation), only *creative* combination can create a 
>> derivative work, automatic combination creates a mere aggregate. Second, the 
>> in-use copy of a program in a computer's memory is not sufficiently 
>> permanent to count as a "work". (Just as the reflection of a book on your 
>> retinas when you read it is not legally a "copy" of the work.) Third, 
>> anything necessary for use is use, and you can't use a computer program 
>> without loading it into memory.
>
> IIRC, some countries copyright law even makes the right to copy a
> computer program to memory for execution explicit.

For example, that of the United States. (Not in so many words, but the
relevant section was IIRC _specifically_ aimed to close the loophole
that allowed EULAs.)

IANAL
0
Jordan
3/3/2006 1:25:09 AM
Stork replied to:

> but is not (necessarily) a derivative work of for copyright purposes.

All computer software is by necessity a derivative work.  Copyright law
designed for books is not the same as intellectual property for
computers.   You don't need to read "On the Road" to make use of "The
Stand", but you do need header files and support files from an
operating system to make an application.

0
stork
3/3/2006 3:26:05 PM
Stork replied to:

>If I implied something you don't agree with, wonderful, but what about what I actually *said*? Are you just going
>to totally ignore that?
>If you want to refute my arguments, that's fine. But summarizing them
>badly isn't an argument

What's to summarize when you aren't saying anything or even being clear
about what you are saying?  It's not my fault that you are a sloppy
writer who undermines his own credibility with circular logic.  I've
read what you have written and most of what you've got is a shrill and
random assault on some ideas that are taken for granted within the
industry.  That you've opened up some eyes to a possible different
perspective is interesting, but, ultimately, your ideas lack sufficient
backing and cohesiveness to be taken seriously.

1) Your legal position is not good.  You roll out with some mumbo jumbo
about how even though you are not a lawyer, you understand IP law,
then, you offer up a patently absurd case of copyright violation and
claim it valid by waving around Lexmark.  And, the best is that you
claim that companies "say" that you are right, but, you abide by their
interpretation of the GPL because you are afraid of getting... if you
were so right, then, why not take it to court?  For that matter, why
hasn't anyone else?  Do you think IBM and Sun lawyers are missing
something about the law that you are when crafting open source
licenses?  Ergot, your own views are hardly on the solid legal ground
that you claim to be.

2) You are using a definition of derivative work that is controversial
as if it your point of view were a statement of fact.   Most of your
arguments around the GPL revolve around your believe that one program
using another does not create a derivative work, when, common sense
within the industry is that, in fact, ALL software is a dervived work
of some kind or another, with the possible exception of the operating
system. The whole work is the network itself.  To even attempt to apply
copyright law that says one book or movie is independent of another to
computers is arguably absurd - because there is no such thing as an
independent work.  That's just the way computers work.   However,
having done so, this industry has the best world from an author's
perspective.  We have the network effects of imposing one's will on
derived works, and as a software author, I can't see how supporting
less author's rights would be beneficial to me.

3) You mix moral and legal arguments as if both were one and the same,
and, where you describe how you wish the universe would be, you refuse
to see all sides of the coin.  You think about what you want other
people to do for you, but, there's not much of a quid pro quo in how
they can in turn treat you.  I'll give you the benefit of the doubt and
assume you are just being asbergerly fixated rather than particularly
small or selfish, but you certainly don't come off as being fair
minded.

So...  other than the fact that you have no real basis within either
the law or computer science upon which to stake your claims, I'd say
you have all your bases covered.  But, until the courts or the congress
determine otherwise, the FSF, IBM's, Sun's, Oracles, MySQLs, the
various Linux foundations, and scores of other projects living within
the GPL as I described, not as you want, govern present practice, and
not you.

0
stork
3/3/2006 3:58:31 PM
Stork replied to:

>Yes, and by releasing it under the GPL, you said "Hey, you can all use
>this". And the GPL doesn't provide for any "takebacks".

No.  I said, you can use all of this if I can use all of yours.  If you
are using mine and I can't use yours, then, there's no deal, and I can
take my stuff back.  It's pretty clear cut.

0
stork
3/3/2006 4:03:43 PM
"stork" <stork@storkyak.com> writes:

> Stork replied to:
> 
> >Yes, and by releasing it under the GPL, you said "Hey, you can all use
> >this". And the GPL doesn't provide for any "takebacks".
> 
> No.  I said, you can use all of this if I can use all of yours.  If you
> are using mine and I can't use yours, then, there's no deal, and I can
> take my stuff back.  It's pretty clear cut.
> 

This has gone on long enough.  Given that stork's B.S. arrived in sync
with a massive Microsoft push to FUD the issues of licenses (I'm
seeing it many places), I'm thinking this may be more of same.  Time
to visit the kill file.

-- 
harry.g.george@boeing.com
6-6M21 BCA CompArch Design Engineering
Phone: (425) 294-4718
0
Harry
3/3/2006 5:58:00 PM
On 2006-03-03, stork <stork@storkyak.com> wrote:
> Stork replied to:
>
>>Yes, and by releasing it under the GPL, you said "Hey, you can all use
>>this". And the GPL doesn't provide for any "takebacks".
>
> No.  I said, you can use all of this if I can use all of yours.

That's not what the GPL says.

> If you are using mine and I can't use yours, then, there's no deal,
> and I can take my stuff back.  It's pretty clear cut.
0
Jordan
3/3/2006 7:46:08 PM
On 2006-03-03, stork <stork@storkyak.com> wrote:
> Stork replied to:
>
>> but is not (necessarily) a derivative work of for copyright purposes.
>
> All computer software is by necessity a derivative work.  Copyright law
> designed for books is not the same as intellectual property for
> computers.   You don't need to read "On the Road" to make use of "The
> Stand", but you do need header files and support files from an
> operating system to make an application.

Which is not necessarily sufficient to make it a derived work.
Particularly if the bits that are included are uncopyrightable [as
typical header files are] or the inclusion of those bits of files [as
statically linked libraries] is completely mechanical.
0
Jordan
3/3/2006 7:47:18 PM
"stork" <stork@storkyak.com> wrote in message 
news:1141399565.195444.15830@j33g2000cwa.googlegroups.com...

> Stork replied to:

>> but is not (necessarily) a derivative work of for copyright purposes.

> All computer software is by necessity a derivative work.  Copyright law
> designed for books is not the same as intellectual property for
> computers.   You don't need to read "On the Road" to make use of "The
> Stand", but you do need header files and support files from an
> operating system to make an application.

    It is funny how self-refuting your argument is. If you *need* to use a 
particular chunk of code to get something done, and there is no other way to 
do that same thing, then it is *not* copyrightable material and using it 
does *not* make a derivative work.

    The whole premise of copyright law is that there is more than one way to 
express the *same* idea and you choose one creatively.

    DS


0
David
3/3/2006 9:03:08 PM
"stork" <stork@storkyak.com> wrote in message 
news:1141401511.000662.47830@u72g2000cwu.googlegroups.com...

> Stork replied to:

>>If I implied something you don't agree with, wonderful, but what about 
>>what I actually *said*? Are you just going
>>to totally ignore that?
>>If you want to refute my arguments, that's fine. But summarizing them
>>badly isn't an argument

> 2) You are using a definition of derivative work that is controversial
> as if it your point of view were a statement of fact.   Most of your
> arguments around the GPL revolve around your believe that one program
[snip]

    Why don't you summarize in your own words what test we should use to 
determine if one piece of software is a derivative work of another.

> 3) You mix moral and legal arguments as if both were one and the same,
> and, where you describe how you wish the universe would be, you refuse
> to see all sides of the coin.  You think about what you want other
> people to do for you, but, there's not much of a quid pro quo in how
> they can in turn treat you.  I'll give you the benefit of the doubt and
> assume you are just being asbergerly fixated rather than particularly
> small or selfish, but you certainly don't come off as being fair
> minded.

    I don't mix them as if they were the same. I do use both, to show that 
my position makes sense both morally and legally. There are certainly areas 
of law that I do not agree with (for an almost-relevant example, look at the 
DMCA's absence of a 'fair use' exception), however, in the case at issue, I 
defend moral issues morally and legal issues legally.

    For example, I have said that under current copyright law, most 
interfaces cannot be copyrighted because they are ideas rather than 
expressions of an idea. However, I have made no attempt to defend this issue 
morally -- in fact, I've taken no position on whether there should be some 
type of protection for interfaces.

    DS


0
David
3/3/2006 9:06:18 PM
"stork" <stork@storkyak.com> wrote in message 
news:1141401823.147646.71470@u72g2000cwu.googlegroups.com...

> Stork replied to:

>>Yes, and by releasing it under the GPL, you said "Hey, you can all use
>>this". And the GPL doesn't provide for any "takebacks".

> No.  I said, you can use all of this if I can use all of yours.  If you
> are using mine and I can't use yours, then, there's no deal, and I can
> take my stuff back.  It's pretty clear cut.

    There are certainly licenses that say that, but the GPL is not one of 
them. That could only be done by a shrink-wrap, click-through, or EULA. A 
mere copyright license just cannot do that because a copyright license 
cannot restrict first sale rights, and first sale rights allow normal use 
without any such requirements.

    For sound legal and moral reasons, if you want to restrict use, you must 
impose that restriction as a condition of receiving the work.

    DS


0
David
3/3/2006 9:08:34 PM
On 2006-03-03, David Schwartz <davids@webmaster.com> wrote:
>
> "stork" <stork@storkyak.com> wrote in message 
> news:1141401823.147646.71470@u72g2000cwu.googlegroups.com...
>
>> Stork replied to:
>
>>>Yes, and by releasing it under the GPL, you said "Hey, you can all use
>>>this". And the GPL doesn't provide for any "takebacks".
>
>> No.  I said, you can use all of this if I can use all of yours.  If you
>> are using mine and I can't use yours, then, there's no deal, and I can
>> take my stuff back.  It's pretty clear cut.
>
>     There are certainly licenses that say that, but the GPL is not one of 
> them. That could only be done by a shrink-wrap, click-through, or EULA. A 
> mere copyright license just cannot do that because a copyright license 
> cannot restrict first sale rights, and first sale rights allow normal use 
> without any such requirements.
>
>     For sound legal and moral reasons, if you want to restrict use, you must 
> impose that restriction as a condition of receiving the work.

It's possible that he is simply using a non-standard meaning of the word
use - i.e., using ONLY the meaning "use a library's API to develop a
program", to the exclusion of the meaning of actually running the code.

Charitably, he also doesn't seem to be able to imagine that someone
could use his GPLed library for one application, released under the GPL,
while developing other (proprietary) applications that have nothing to
do with his library.
0
Jordan
3/3/2006 10:53:12 PM
Stork replied to:

>Given that stork's B.S. arrived in sync
>with a massive Microsoft push to FUD the issues of licenses (I'm
>seeing it many places), I'm thinking this may be more of same.  Time
>to visit the kill file.

How can you say that I'm in bed with MS when I'm the one that is
arguing a pro-open source developer point of view of the GPL and
defending Trolltech's interpretation of it?   I don't see anyone else
running comparisons of KDevelop vs Visual Studio 2005 on their blogs
for C++ programming and I don't see anyone else running articles that
explain how to migrate from Windows to Linux for those familiar with
Visual Studio.  So, go ahead and killfile me, but you are sticking up
for the wrong side.

0
stork
3/4/2006 12:01:56 AM
Stork replied to:

>Charitably, he also doesn't seem to be able to imagine that someone
>could use his GPLed library for one application, released under the GPL,
>while developing other (proprietary) applications that have nothing to
>do with his library.

No, what I'm saying is, if you use my GPL library for your app, it
ought to be released under the GPL, unless you and I make a seperate
arrangement.  If you don't use my library in other stuff, why the hell
would I care?

0
stork
3/4/2006 12:03:40 AM
On 2006-03-04, stork <stork@storkyak.com> wrote:
> Stork replied to:
>
>>Charitably, he also doesn't seem to be able to imagine that someone
>>could use his GPLed library for one application, released under the GPL,
>>while developing other (proprietary) applications that have nothing to
>>do with his library.
>
> No, what I'm saying is, if you use my GPL library for your app, it
> ought to be released under the GPL, unless you and I make a seperate
> arrangement.  If you don't use my library in other stuff, why the hell
> would I care?

That's not what you _SAID_, though.
0
Jordan
3/4/2006 1:38:37 AM
"stork" <stork@storkyak.com> wrote in message 
news:1141430516.111071.274660@z34g2000cwc.googlegroups.com...

> Stork replied to:

>>Given that stork's B.S. arrived in sync
>>with a massive Microsoft push to FUD the issues of licenses (I'm
>>seeing it many places), I'm thinking this may be more of same.  Time
>>to visit the kill file.

> How can you say that I'm in bed with MS when I'm the one that is
> arguing a pro-open source developer point of view of the GPL and
> defending Trolltech's interpretation of it?

    Because you are arguing for a significantly narrower view of first sale 
and fair use rights. This likely benefits those who wish to develop 
proprietary software and interfaces more than it benefits the *real* open 
source community.

> I don't see anyone else
> running comparisons of KDevelop vs Visual Studio 2005 on their blogs
> for C++ programming and I don't see anyone else running articles that
> explain how to migrate from Windows to Linux for those familiar with
> Visual Studio.  So, go ahead and killfile me, but you are sticking up
> for the wrong side.

    What's the point if Linux isn't free as in freedom?

    DS


0
David
3/4/2006 4:46:48 AM
"stork" <stork@storkyak.com> wrote in message 
news:1141430620.237450.145880@u72g2000cwu.googlegroups.com...

> No, what I'm saying is, if you use my GPL library for your app, it
> ought to be released under the GPL, unless you and I make a seperate
> arrangement.  If you don't use my library in other stuff, why the hell
> would I care?

    What you miss is that the GPL is supposed to be free as in *freedom*, 
not free as in free beer. What you want is a license that *doesn't* give you 
the freedom to tinker. Such licenses exist, but the GPL is not one of them.

    DS


0
David
3/4/2006 4:48:18 AM
Storkyak

>That's not what you _SAID_, though. 

Uh, no.

0
stork
3/4/2006 6:55:49 AM
Stork replied to:

>Because you are arguing for a significantly narrower view of first sale
>and fair use rights. This likely benefits those who wish to develop
>proprietary software and interfaces more than it benefits the *real* open
>source community.

While I appreciate the advantages of open source and the commitment
people make to do so, I don't any need to give up any rights whatsoever
as an artist in order to have an open source society.

I'm arguing that artists should be allowed to choose the manner of
their licensing.  The most compelling thing about Linux is that
membership in the community is voluntary but that the GPL must have
legal teeth to ensure that once you commit to the community, you can't
break the rules.  If we are going to jump off of the cliff of open
source and into the great wide unknown, then we can't have some of us
saying, "well, I'll just make this closed source application on the
side".  We sink or swim together.  If you want to write closed source
software, write it for Windows.  I won't hold it against you.  I can't
say I won't write closedware for Windows again.  Vista does look pretty
cool.  But don't go arguing "freedom" in order to taint the spirit of
keeping all software open source that is the heart of Linux.  If it
means that Linux somehow doesn't "beat" Windows, well, so be it.  I'd
rather have a good open source community than yet another semi-closed
alternative OS - I've already invested in Be and OS/2 and Amiga enough
to not even bother with that route again.  It's the openness of Linux
that makes it special.

And, to be honest, the "free" as in beer aspect of Linux is more of a
hindrance than a help.  I -wish- there was a way to nominally charge
for the distribution of Linux copies so that the various teams would
get some bucks to work on their projects.  I'd like to see the people
at X.org or KDE or Gnome get better funding.  Keep it cheap, I don't
want to pay -too- much, but, even a modest charge is worth it.  That's
why I buy my Linux distros even though I don't have to.

>What's the point if Linux isn't free as in freedom?

It's the ultimate paradox - you can't have a free society without rules
to enforce the freedom.  In a capitalistic society, the way to achieve
this is give the artists increasing power over their intellectual
property so that they can effectively choose who is in the community
and who is not.   That way, if someone doesn't play ball, they can get
the boot.

Computer software IS private property.  Intellectual property IS real
property, and, if I choose to share that property to better a community
somehow, I want that to be my choice, not some automatic socialistic
rule handed down by the government saying that software is not
property.    It is.  You can invest into it, add onto it, build it just
as you would a house.  Linus Torvalds wrote the Linux Kernel. It's his.
 Thanks for the use.  But, if he wanted to charge for it down the road,
he certainly could do so.

0
stork
3/4/2006 7:15:17 AM
On 2006-03-04, stork <stork@storkyak.com> wrote:
> Storkyak
>
>>That's not what you _SAID_, though. 
>
> Uh, no.

"I said, you can use all of this if I can use all of yours."

The only reasonable interpretation of "all of yours" includes stuff that
the other author writes that does not link against your library.
0
Jordan
3/4/2006 8:04:30 AM
Stork replied to:

>What you miss is that the GPL is supposed to be free as in *freedom*,
>not free as in free beer. What you want is a license that *doesn't* give you
>the freedom to tinker.

No, not at all.  By all means, tinker away.  But, if you go and take my
open source stuff, and make a closed source product with it, then,
that's a different story.  In other words, I'm jumping into the open
source community, standing on the backs of all the incredible people
that wrote things like gcc, gas, linux itself, etc, and as part of that
I -should- regardless of the "legality", make my stuff open source GPL
as well.   And, if you or anyone else should stand on my shoulders,
then, they too should do the same.

The GPL is a social compact, I guess, for lack of a better phrase.  I
see it more as a constitution of a community shoehorned into a
copyright, per se.  The one area where I would disagree with the GPL,
is that, I don't really believe too much in the importance of free as
in beer, but free as in open and tinkering, well, that should be open.
If it were feasible, I'd almost like to see a Linux DRM with
micropayments to all the various developers so that we could have
benefits of open source and the rich literary spirit of academic
exploration that goes with it, but, see the authors get paid for it as
well.  But, if it doesn't happen, I could live with it too, and, given
the beauty and depth of what Linux is today, maybe the free as in beer
thing is ok too.

0
stork
3/4/2006 8:12:35 AM
You are consistently failing to understand the difference between what
you want the GPL to be and what it actually is.
0
Jordan
3/4/2006 8:13:57 AM
Stork replied to:

>The only reasonable interpretation of "all of yours" includes stuff that
>the other author writes that does not link against your library.

The context of the conversation made it clear that this was not the
case.

0
stork
3/4/2006 8:14:11 AM
"stork" <stork@storkyak.com> wrote in message 
news:1141456517.602982.277650@i40g2000cwc.googlegroups.com...

> While I appreciate the advantages of open source and the commitment
> people make to do so, I don't any need to give up any rights whatsoever
> as an artist in order to have an open source society.

    You are arguing for a world where the streets are paved with gumdrops 
and it only rains lollypops. It is impossible to give artists all the rights 
you imagine without leading to absurdities. For example, why can't I write a 
poem, drop copies of it from an airplane, and then sue/charge everyone who 
reads it? Why should I have to give up any rights to charge for use just 
because I give copies of my poem to everyone?

    Despite the fact that I've mentioned this absurdity in your reasoning at 
least five times,  you've never addressed it. I guess shame on me for 
treating you like a rational person.

    DS


0
David
3/5/2006 7:17:44 AM
"stork" <stork@storkyak.com> wrote in message 
news:1141459955.242390.259190@u72g2000cwu.googlegroups.com...

> Stork replied to:

>>What you miss is that the GPL is supposed to be free as in *freedom*,
>>not free as in free beer. What you want is a license that *doesn't* give 
>>you
>>the freedom to tinker.

> No, not at all.  By all means, tinker away.  But, if you go and take my
> open source stuff, and make a closed source product with it, then,
> that's a different story.

    I thought you were arguing that making a derivative work *wasn't* use. 
Now it sounds like you are saying that it is. I guess you're being so vague 
nobody can tell what you mean.

    If I buy a library, should I automatically have the right to develop 
applications that use it? Yes or no?

    If someone posts the source code to their library on a billboard, are 
they giving up the right to say you can't use that library in your 
application? Yes or no?

> In other words, I'm jumping into the open
> source community, standing on the backs of all the incredible people
> that wrote things like gcc, gas, linux itself, etc, and as part of that
> I -should- regardless of the "legality", make my stuff open source GPL
> as well.   And, if you or anyone else should stand on my shoulders,
> then, they too should do the same.

    But you are not arguing that they should, you are arguing that they 
*must*.

> The GPL is a social compact, I guess, for lack of a better phrase.  I
> see it more as a constitution of a community shoehorned into a
> copyright, per se.  The one area where I would disagree with the GPL,
> is that, I don't really believe too much in the importance of free as
> in beer, but free as in open and tinkering, well, that should be open.

    So you think that developing derivative works of a library should be 
included in fair use and first sale? Or not?

> If it were feasible, I'd almost like to see a Linux DRM with
> micropayments to all the various developers so that we could have
> benefits of open source and the rich literary spirit of academic
> exploration that goes with it, but, see the authors get paid for it as
> well.  But, if it doesn't happen, I could live with it too, and, given
> the beauty and depth of what Linux is today, maybe the free as in beer
> thing is ok too.

    If you want a proprietary OS, you know where to find them. If you had to 
pay for Linux, it wouldn't be Linux anymore.

    Worse, you want to steal the end result of free cooperation and then 
slam a price tag on it at the end. You totally forget that it was the 
freedom that the developers believed in that made Linux what it is today. 
With that sort of arrangement, many of the key developers likely would not 
have participated and certainly would have participated differently.

    DS


0
David
3/5/2006 7:21:21 AM
"stork" <stork@storkyak.com> wrote in message 
news:1141460051.183023.144980@p10g2000cwp.googlegroups.com...

> Stork replied to:

>>The only reasonable interpretation of "all of yours" includes stuff that
>>the other author writes that does not link against your library.

> The context of the conversation made it clear that this was not the
> case.

    Actually, with your complete lack of understanding of what a derivative 
work is, I doubt you had any clue what you meant it to encompass.

    DS


0
David
3/5/2006 7:21:56 AM
On 2006-03-05, David Schwartz <davids@webmaster.com> wrote:
>
> "stork" <stork@storkyak.com> wrote in message 
> news:1141460051.183023.144980@p10g2000cwp.googlegroups.com...
>
>> Stork replied to:
>
>>>The only reasonable interpretation of "all of yours" includes stuff that
>>>the other author writes that does not link against your library.
>
>> The context of the conversation made it clear that this was not the
>> case.
>
>     Actually, with your complete lack of understanding of what a
> derivative work is, I doubt you had any clue what you meant it to
> encompass.

Especially since at one point he seemed to be saying that since "the
whole network" is part of "the system" or some BS like that, all
programs are derivative work of all other programs across the entire
internet. [He did use the term "network" at one point]. He also said
that the laws about derivative works can't be reasonably applied to
software at all like they can for books. [Clearly they have to be
applied - maybe he MEANT they won't lead to the conclusion "not a DW"
where they would lead to it for books, but he didn't SAY that]
0
Jordan
3/5/2006 9:25:13 PM
"Jordan Abel" <random832@gmail.com> wrote in message 
news:slrne0mm0c.1a75.random832@random.yi.org...

> Especially since at one point he seemed to be saying that since "the
> whole network" is part of "the system" or some BS like that, all
> programs are derivative work of all other programs across the entire
> internet. [He did use the term "network" at one point]. He also said
> that the laws about derivative works can't be reasonably applied to
> software at all like they can for books. [Clearly they have to be
> applied - maybe he MEANT they won't lead to the conclusion "not a DW"
> where they would lead to it for books, but he didn't SAY that]

    It's kind of funny, any time there's some kind of discussion about the 
legal and moral implications of the GPL, someone seems to pop up with some 
irrelevent, obviously incorrect point that sidetracks any attempt at 
legitimate discussions. It happens too often to be coincidence.

    DS


0
David
3/6/2006 7:22:29 AM
"stork" <stork@storkyak.com> wrote in message 
news:1141140376.112689.282490@p10g2000cwp.googlegroups.com...

> Stork replied to:

>>When you place something under the GPL and you
>>sell or transfer it to at least one person, you have effectively made
>>everyone a legal licensee of the work

> Nope, because, as a copyright owner, I have the right to determine who
> can use my work or not.  The GPL is not a transfer of copyright, as you
> would argue that it is.  Besides, your viewpoint is silly.  You would
> argue that because GM might give away a few hundred Pontiac G6's on
> Oprah, that everyone should get it for free.

    By the way, I should have pointed out that the right to determine who 
can use a work or the right to restrict its use is *NOT* one of the rights 
granted under copyright law. See, for example, 
http://www.copyright.gov/circs/circ1.html

    The right to restrict the distribution of derivative works is also *NOT* 
one of the rights granted under copyright law.

    DS


0
David
3/6/2006 9:16:19 AM
On 2006-03-06, David Schwartz <davids@webmaster.com> wrote:
>
>     The right to restrict the distribution of derivative works is also *NOT* 
> one of the rights granted under copyright law.

Though there is a right to restrict the _creation_ of them, which is
almost as good in practice.
0
Jordan
3/6/2006 3:06:54 PM
Stork replied to:

> For example, why can't I write a
>poem, drop copies of it from an airplane, and then sue/charge everyone who
>eads it?

Once again.  If I drop the poems from an airplane, everyone can read
them.  But, if someone copies that work or attempts to sell copies of
it, that's an IP violation.

>Why should I have to give up any rights to charge for use just
>ecause I give copies of my poem to everyone?

You don't.  You gave away 500 poems, and thus, have 500 free users.
But those users cannot go and fire up a printing press to crank them
out.  If I give you the latest Stephen King book, you don't get to just
put it on the internet as fair use.

>Despite the fact that I've mentioned this absurdity in your reasoning at
>east five times,  you've never addressed it.

And as I've repeatedly said, you are drawing false conclusions by using
a confused definition of "use".

0
stork
3/6/2006 3:14:11 PM
Stork replied to:

> If I buy a library, should I automatically have the right to develop
>applications that use it? Yes or no?

Automatically?  No.  You have the rights to the library that you
contracted to when you purchased it.  Obviously, you wouldn't buy the
library if you couldn't develop with it.

>I thought you were arguing that making a derivative work *wasn't* use.
>Now it sounds like you are saying that it is. I guess you're being so vague
>nobody can tell what you mean.

I think you are just spinning things here.

>If someone posts the source code to their library on a billboard, are
>they giving up the right to say you can't use that library in your
>application? Yes or no?

Absolutely not.  If the NFL posts its logo on the bulletin board, does
that mean I should be allowed to start selling NFL t-shirts because I
saw it?  What about Star Wars or Star Trek?

>    But you are not arguing that they should, you are arguing that they
>*must*.

That's right.  They -must-.

>So you think that developing derivative works of a library should be
>included in fair use and first sale? Or not?

No, they should not.

>Worse, you want to steal the end result of free cooperation and then
>slam a price tag on it at the end

That's a fair criticism.  What I really wish is for the Linux system to
get more funding for development.

>many of the key developers

I'm interested in preserving the rights of developers.  If they want to
create a free society then they must have the means to enforce it, but,
membership in that society should not be done by a mandatory federal
law.  There's advantages to the closed system as well.  To wit:  I do
not like some aspects of Windows, but, there are some aspects that I
-do- like.  I just happen to like Linux more.

With that said, I do think the -length- of copyrights and patents
should be reduced.  Protect the authors yes, but, once they are dead,
given a few years, the work should enter the public domain.  Like,  I
think Disney should not have copyrights on Mickey Mouse and company...
its time for them to think up something new.

0
stork
3/6/2006 5:57:47 PM
On 2006-03-06, stork <stork@storkyak.com> wrote:
> Stork replied to:
>
>> If I buy a library, should I automatically have the right to develop
>>applications that use it? Yes or no?
>
> Automatically?  No.  You have the rights to the library that you
> contracted to when you purchased it.  Obviously, you wouldn't buy the
> library if you couldn't develop with it.

So what about a GPL library? Can you develop with it or not? I mean,
legally in real life, not morally in your fantasy world.

>>I thought you were arguing that making a derivative work *wasn't* use.
>>Now it sounds like you are saying that it is. I guess you're being so vague
>>nobody can tell what you mean.
>
> I think you are just spinning things here.

I think you were being deliberately vague so you would be free to spin
things later and in conflicting ways in different subthreads.
0
Jordan
3/6/2006 6:49:16 PM
Stork replied to:

>So what about a GPL library? Can you develop with it or not?

You can develop with it so long as your own application is also GPL.
Or, you can make a seperate arrangement with the author of the library.
 So far there has been no legal challenge to the GPL that directly
contradicts that basic tenent.  If the library is LGPL, or BSD style,
then, you can make a closed source product with it.

>I mean, legally in real life, not morally in your fantasy world.

Sorry, I guess I made the mistake of assuming that morality was not a
fantasy, but I guess with you it is.

>I think you were being deliberately vague so you would be free to spin
>hings later and in conflicting ways in different subthreads.

I've said nothing in conflict.  I've specifically said, at the
beginning, if you want to make a closed source application with
Trolltech stuff, then you have to buy the license.  Otherwise, you can
make a GPL application.  Incidentally, this is the same thing with
Intel's C++ compiler...

0
tbandrow
3/6/2006 7:42:11 PM
Stork replied to:

>By the way, I should have pointed out that the right to determine who
>can use a work or the right to restrict its use is *NOT* one of the rights
>granted under copyright law. See, for example,
>The right to restrict the distribution of derivative works is also *NOT*
>one of the rights granted under copyright law

Dude, that's not what the web site says:

"Section 106 of the 1976 Copyright Act generally gives the owner of
copyright the exclusive right to do and to authorize others to do the
following:

"To reproduce the work in copies or phonorecords;

"To prepare derivative works based upon the work;

In other words, I have the right as the copyright holder to authorize
who may prepare derivative works, and who may distribute them.  You
make the ridiculous argument that I cannot do both at once.  That's
like saying, well, its illegal to rob someone, and its illegal to kill
someone, but, if you go and kill and rob someone, that's not covered.

And, if we read the act itself, we find:

"(b)(1)(A) Notwithstanding the provisions of subsection (a), UNLESS
AUTHORIZED BY THE OWNERS OF COPYRIGHT IN THE SOUND RECORDING OR THE
OWNER OF COPYRIGHT IN A COMPUTER PROGRAM (including any tape, disk, or
other medium embodying such program), and in the case of a sound
recording in the musical works embodied therein, neither the owner of a
particular phonorecord nor any person in possession of a particular
copy of a computer program (including any tape, disk, or other medium
embodying such program), may, for the purposes of direct or indirect
commercial advantage, dispose of, or authorize the disposal of, the
possession of that phonorecord or computer program (including any tape,
disk, or other medium embodying such program) by rental, lease, or
lending, or by any other act or practice in the nature of rental,
lease, or lending. Nothing "

Explicitly, what you are talking about is totally illegal, under that
provision alone!

0
tbandrow
3/6/2006 7:52:34 PM
On 2006-03-06, tbandrow@storkyak.com <tbandrow@storkyak.com> wrote:
> Stork replied to:
>
>>So what about a GPL library? Can you develop with it or not?
>
> You can develop with it so long as your own application is also GPL.
> Or, you can make a seperate arrangement with the author of the library.
>  So far there has been no legal challenge to the GPL that directly
> contradicts that basic tenent.  If the library is LGPL, or BSD style,
> then, you can make a closed source product with it.

So far there has been no legal challenge to the GPL at all - that
doesn't make the FSF's interpretation correct.

>>I mean, legally in real life, not morally in your fantasy world.
>
> Sorry, I guess I made the mistake of assuming that morality was not a
> fantasy, but I guess with you it is.

The moral content you are projecting onto the GPL is your own creation
and may not coincide with what other authors using the GPL really
intend.

>>I think you were being deliberately vague so you would be free to spin
>>hings later and in conflicting ways in different subthreads.
>
> I've said nothing in conflict.  I've specifically said, at the
> beginning, if you want to make a closed source application with
> Trolltech stuff, then you have to buy the license.  Otherwise, you can
> make a GPL application.  Incidentally, this is the same thing with
> Intel's C++ compiler...
>
0
Jordan
3/6/2006 7:55:58 PM
On 2006-03-06, tbandrow@storkyak.com <tbandrow@storkyak.com> wrote:
> Stork replied to:
>
>>By the way, I should have pointed out that the right to determine who
>>can use a work or the right to restrict its use is *NOT* one of the rights
>>granted under copyright law. See, for example,
>>The right to restrict the distribution of derivative works is also *NOT*
>>one of the rights granted under copyright law
>
> Dude, that's not what the web site says:
>
> "Section 106 of the 1976 Copyright Act generally gives the owner of
> copyright the exclusive right to do and to authorize others to do the
> following:
>
> "To reproduce the work in copies or phonorecords;
>
> "To prepare derivative works based upon the work;
>
> In other words, I have the right as the copyright holder to authorize
> who may prepare derivative works, and who may distribute them.  You
> make the ridiculous argument that I cannot do both at once.  That's
> like saying, well, its illegal to rob someone, and its illegal to kill
> someone, but, if you go and kill and rob someone, that's not covered.

The copyright to the derivative work, once created, belongs to its
creator, not to you. Thus it makes no sense for you to be able to
directly restrict the distribution of those derivative works. Arguably,
you can make such a restriction as a condition of a license of the
ability to create one, so this entire discussion is splitting hairs. But
if, for example, you authorize someone to create a derivative work with
only the _expectation_ that you can then control its distribution,
without actually putting that in the contract, you have no control.

You _CAN_ do both at once. You can both restrict the creation of
derivative works and restrict the distribution of your own work. You can
even restrict the same person from doing either. But if you allow the
creation of a derivative work you can NOT, under copyright law, restrict
its distribution. Only its owner may do that.

IANAL but neither are you.
0
Jordan
3/6/2006 8:00:53 PM
stork replied to:

>So far there has been no legal challenge to the GPL at all - that
>doesn't make the FSF's interpretation correct.
>The moral content you are projecting onto the GPL is your own creation
>and may not coincide with what other authors using the GPL really
>intend.

My position is both logical and grants the authors maximum rights.  If
they waive those rights, that is there perogative, but is their
decision, not yours.

0
tbandrow
3/6/2006 8:02:50 PM
On 2006-03-06, tbandrow@storkyak.com <tbandrow@storkyak.com> wrote:
> stork replied to:
>
>>So far there has been no legal challenge to the GPL at all - that
>>doesn't make the FSF's interpretation correct.
>>The moral content you are projecting onto the GPL is your own creation
>>and may not coincide with what other authors using the GPL really
>>intend.
>
> My position is both logical and grants the authors maximum rights.  If
> they waive those rights, that is there perogative, but is their
> decision, not yours.

But it's not what the GPL says, and, at one point in this thread you
were arguing a position even more extreme than the FSF's own. [Or maybe
you got tripped up over the word 'use' - that whole "all computer
programs on the whole network are interconnected" line, though, seems to
imply otherwise]
0
Jordan
3/6/2006 8:05:01 PM
Stork replied to:

>Arguably, you can make such a restriction as a condition of a license of the
>ability to create one, so this entire discussion is splitting hairs
>But if you allow the creation of a derivative work you can NOT, under copyright law, restrict
>its distribution. Only its owner may do that.

These are academic distinctions because the copyright holder can
extract whatever terms he or she wants in order to permit you to use or
distribute the work.

The trick with the GPL, which I admit, may not hold up in courts, is in
the nature of its agreement.

The only legally relevant question in all of this is whether or not a
work published under the GPL constitutes a public display.  As David is
so fond of saying, software published under a GPL is like being dropped
out of a plane or put up on a billboard, in which case, one might
reasonably argue that such software might be copyrightable as a public
performance or display.  You couldn't go into a concert and video tape
a performance and sell it, because that's an infringement.

A GPL'd work, I argue, is the same thing.  But if it were not the same
thing, then, putting something into GPL would effectively be putting it
into the public domain as there was license acceptance by the end user
to receive the work.  I don't know if the GPL's automatic copyright
revocation would hold.  As a practical matter, I do not see the powers
that be ever challenging that because to weaken the GPL would also be
to weaken Microsoft's own software licensing and distribution schemes.
For example, if someone used a Microsoft library with their application
and it wound up on my machine, then what's to prohibit me from
violating all of the MS licensing terms?  If the library is a public
display, I'm not entitled to it, but, if it is not, then its a gift
from the gods and I can do with it as I will.

So, IANAL, but, as far as I can see, if software is a public display,
like a concert or a play in a park, then the GPL stands, and if not,
its screwed and has to be replaced with a more shrink wrapped EULA.

0
tbandrow
3/6/2006 8:17:05 PM
Stork replied to:

>IANAL but neither are you

True, but, here's a common sense test.  Can you resell something you
record off of TV or the radio, or a concert you saw in the park?  No.
Can you resell a piece of software that is GPLd onto your hard drive?
No.

0
tbandrow
3/6/2006 8:23:57 PM
"Jordan Abel" <random832@gmail.com> wrote in message 
news:slrne0ok74.1eg8.random832@random.yi.org...

> On 2006-03-06, David Schwartz <davids@webmaster.com> wrote:

>>     The right to restrict the distribution of derivative works is also 
>> *NOT*
>> one of the rights granted under copyright law.

> Though there is a right to restrict the _creation_ of them, which is
> almost as good in practice.

    No, not at all. For example, suppose I get the right to create a 
derivative work somehow, say through first sale or fair use. If you had an 
independent right to restrict the distribution of derivative works, I could 
not distribute that derivative work after I created. Since you don't, I may 
be able to.

    It gets really complicated, unfortunately. But unless you want to get 
into ugly nitty gritty legal details, you'll have to take my word for it. ;)

    DS


0
David
3/6/2006 8:47:24 PM
"stork" <stork@storkyak.com> wrote in message 
news:1141658050.992153.297870@j33g2000cwa.googlegroups.com...

> Stork replied to:

>> For example, why can't I write a
>>poem, drop copies of it from an airplane, and then sue/charge everyone who
>>eads it?

> Once again.  If I drop the poems from an airplane, everyone can read
> them.  But, if someone copies that work or attempts to sell copies of
> it, that's an IP violation.

    Right, however, you can't restrict *use* or anything that's necessary 
for use. You can't restrict transfers, that is, you can't stop me from 
giving my copy of the poem to someone else.

>>Why should I have to give up any rights to charge for use just
>>ecause I give copies of my poem to everyone?
>
> You don't.  You gave away 500 poems, and thus, have 500 free users.
> But those users cannot go and fire up a printing press to crank them
> out.  If I give you the latest Stephen King book, you don't get to just
> put it on the internet as fair use.

    But that's not the case here. The case here is that you give away an 
unlimited number of copies and authorize third parties to similarly give 
away unlimited numbers of copies. At that point, everyone is a licensee of 
your work.

>>Despite the fact that I've mentioned this absurdity in your reasoning at
>>east five times,  you've never addressed it.

> And as I've repeatedly said, you are drawing false conclusions by using
> a confused definition of "use".

    I have been so totally precise about what I mean by "use" that I can't 
imagine how you could say that it is confused. "Use" means the ordinary, 
reasonable, expected way that a work would be employed, including anything 
necessary to do that. For a software library, this would include developing 
applications that use that library.

    DS


0
David
3/6/2006 8:49:32 PM
"stork" <stork@storkyak.com> wrote in message 
news:1141667866.955760.105100@u72g2000cwu.googlegroups.com...

> Stork replied to:

>> If I buy a library, should I automatically have the right to develop
>>applications that use it? Yes or no?

> Automatically?  No.  You have the rights to the library that you
> contracted to when you purchased it.  Obviously, you wouldn't buy the
> library if you couldn't develop with it.

    When you legally acquire a work, without a EULA, signed contract, 
shrink-wrap or similar *agreeement* (not a license), you automatically 
acquire the right to use the work. To say "No", you must either take the 
position that developing an application that uses a library is not ordinary 
use of that library or that one should not acquire the right to use a work 
if one legally acquires the work. Which way will it be?

    DS


0
David
3/6/2006 8:50:48 PM
<tbandrow@storkyak.com> wrote in message 
news:1141674131.648951.31370@z34g2000cwc.googlegroups.com...

> Stork replied to:

>>So what about a GPL library? Can you develop with it or not?

> You can develop with it so long as your own application is also GPL.
> Or, you can make a seperate arrangement with the author of the library.
> So far there has been no legal challenge to the GPL that directly
> contradicts that basic tenent.  If the library is LGPL, or BSD style,
> then, you can make a closed source product with it.

    This is not a basic tenet of the GPL. The GPL only attempts to restrict 
things that are *not* normal use. Read the second half of section zero 
again. Or is developing applications that use a library not normal use? If 
that's your position, what would be "use" of a library?

    DS


0
David
3/6/2006 8:52:09 PM
<tbandrow@storkyak.com> wrote in message 
news:1141674754.161107.300540@j33g2000cwa.googlegroups.com...
> Stork replied to:
>
>>By the way, I should have pointed out that the right to determine who
>>can use a work or the right to restrict its use is *NOT* one of the rights
>>granted under copyright law. See, for example,
>>The right to restrict the distribution of derivative works is also *NOT*
>>one of the rights granted under copyright law
>
> Dude, that's not what the web site says:
>
> "Section 106 of the 1976 Copyright Act generally gives the owner of
> copyright the exclusive right to do and to authorize others to do the
> following:
>
> "To reproduce the work in copies or phonorecords;
>
> "To prepare derivative works based upon the work;
>
> In other words, I have the right as the copyright holder to authorize
> who may prepare derivative works, and who may distribute them.

    You have the right to authorize who may prepare derivative works but 
*NOT* who can distribute them. This makes a *huge* difference.

> You
> make the ridiculous argument that I cannot do both at once.  That's
> like saying, well, its illegal to rob someone, and its illegal to kill
> someone, but, if you go and kill and rob someone, that's not covered.

    Huh?

> And, if we read the act itself, we find:
>
> "(b)(1)(A) Notwithstanding the provisions of subsection (a), UNLESS
> AUTHORIZED BY THE OWNERS OF COPYRIGHT IN THE SOUND RECORDING OR THE
> OWNER OF COPYRIGHT IN A COMPUTER PROGRAM (including any tape, disk, or
> other medium embodying such program), and in the case of a sound
> recording in the musical works embodied therein, neither the owner of a
> particular phonorecord nor any person in possession of a particular
> copy of a computer program (including any tape, disk, or other medium
> embodying such program), may, for the purposes of direct or indirect
> commercial advantage, dispose of, or authorize the disposal of, the
> possession of that phonorecord or computer program (including any tape,
> disk, or other medium embodying such program) by rental, lease, or
> lending, or by any other act or practice in the nature of rental,
> lease, or lending. Nothing "

    That provision is about the original work, not derivative works.

> Explicitly, what you are talking about is totally illegal, under that
> provision alone!

    Who was talking about rental, leasing or lending?

    DS


0
David
3/6/2006 8:53:36 PM
"Jordan Abel" <random832@gmail.com> wrote in message 
news:slrne0p5ec.2e6v.random832@random.yi.org...

> The copyright to the derivative work, once created, belongs to its
> creator, not to you. Thus it makes no sense for you to be able to
> directly restrict the distribution of those derivative works.

    Exactly.

> Arguably,
> you can make such a restriction as a condition of a license of the
> ability to create one, so this entire discussion is splitting hairs.

    No, no, no! Supppose you create a derivative work in the course of 
normal use.

> But if you allow the
> creation of a derivative work you can NOT, under copyright law, restrict
> its distribution.

    Alas, it's not that simple. For example, suppose you buy a copy of a 
music CD. You decide to add some extra drum sounds to the work. Assume, for 
the sake of argument, this is deemed to be fair use. It would be totally 
unreasonable to argue that you can now sell copies of this derivative work, 
simply because you created it legally.

    We are getting far afield here and into very technical waters. However, 
no court has ever held that a lawfully-created derivative work may not be 
distributed to licensees of the original work with the permission of the 
creator of the derivative work in the absence of an agreement. And that's 
what's important in the GPL case, because everyone may use the original 
work.

    DS


0
David
3/6/2006 8:56:53 PM
<tbandrow@storkyak.com> wrote in message 
news:1141675370.753511.272380@p10g2000cwp.googlegroups.com...

> stork replied to:

>>So far there has been no legal challenge to the GPL at all - that
>>doesn't make the FSF's interpretation correct.
>>The moral content you are projecting onto the GPL is your own creation
>>and may not coincide with what other authors using the GPL really
>>intend.

> My position is both logical and grants the authors maximum rights.  If
> they waive those rights, that is there perogative, but is their
> decision, not yours.

    But your position is totally illogical, since it grants authors the 
right to post a copy of their work on a billboard and then sue everyone who 
reads it.

    DS


0
David
3/6/2006 8:57:23 PM
<tbandrow@storkyak.com> wrote in message 
news:1141676225.503593.100060@u72g2000cwu.googlegroups.com...

> Stork replied to:

>>Arguably, you can make such a restriction as a condition of a license of 
>>the
>>ability to create one, so this entire discussion is splitting hairs
>>But if you allow the creation of a derivative work you can NOT, under 
>>copyright law, restrict
>>its distribution. Only its owner may do that.

> These are academic distinctions because the copyright holder can
> extract whatever terms he or she wants in order to permit you to use or
> distribute the work.

    Not if he gives the work away. Then you get certain rights under the 
law, and the copyright holder is left with only those specific rights the 
law grants him.

> The trick with the GPL, which I admit, may not hold up in courts, is in
> the nature of its agreement.

> The only legally relevant question in all of this is whether or not a
> work published under the GPL constitutes a public display.  As David is
> so fond of saying, software published under a GPL is like being dropped
> out of a plane or put up on a billboard, in which case, one might
> reasonably argue that such software might be copyrightable as a public
> performance or display.  You couldn't go into a concert and video tape
> a performance and sell it, because that's an infringement.

    Huh? Whether or not it's a public display, it is available to anyone who 
wants it free of any restriction other than those found in copyright law. 
Everyone can get a copy without agreeing to anything and then has their 
rights under first sale. (Assuming at least one person who has the work is 
willing to distribute it, of course.)

> A GPL'd work, I argue, is the same thing.  But if it were not the same
> thing, then, putting something into GPL would effectively be putting it
> into the public domain as there was license acceptance by the end user
> to receive the work.

    The difference between putting something under the GPL and putting 
something into the public domain is people who want to distribute a GPL'd 
work (or create deriviative works other than in the course of ordinary use) 
must comply with the GPL's requirements.

> I don't know if the GPL's automatic copyright
> revocation would hold.  As a practical matter, I do not see the powers
> that be ever challenging that because to weaken the GPL would also be
> to weaken Microsoft's own software licensing and distribution schemes.
> For example, if someone used a Microsoft library with their application
> and it wound up on my machine, then what's to prohibit me from
> violating all of the MS licensing terms?  If the library is a public
> display, I'm not entitled to it, but, if it is not, then its a gift
> from the gods and I can do with it as I will.

    If someone redistributed a Microsoft library to you as part of a 
product, I would argue that you are entitled to use that library, even if 
that includes creating applications of your own that use it. That's 
assuming, of course, there's no EULA, shrink-wrap agreement, click-through, 
or so on.

> So, IANAL, but, as far as I can see, if software is a public display,
> like a concert or a play in a park, then the GPL stands, and if not,
> its screwed and has to be replaced with a more shrink wrapped EULA.

    I don't see what the public display issue has to do with anything.

    DS


0
David
3/6/2006 9:00:53 PM
<tbandrow@storkyak.com> wrote in message 
news:1141676637.333666.229170@z34g2000cwc.googlegroups.com...

> Stork replied to:

>>IANAL but neither are you

> True, but, here's a common sense test.  Can you resell something you
> record off of TV or the radio, or a concert you saw in the park?  No.
> Can you resell a piece of software that is GPLd onto your hard drive?
> No.

    The GPL *grants* you the right to sell copies of GPL'd software. 
Otherwise every company that sells Linux CDs would be violating copyright.

    DS


0
David
3/6/2006 9:01:43 PM
On 2006-03-06, tbandrow@storkyak.com <tbandrow@storkyak.com> wrote:
> Stork replied to:
>
>>Arguably, you can make such a restriction as a condition of a license of the
>>ability to create one, so this entire discussion is splitting hairs
>>But if you allow the creation of a derivative work you can NOT, under copyright law, restrict
>>its distribution. Only its owner may do that.
>
> These are academic distinctions because the copyright holder can
> extract whatever terms he or she wants in order to permit you to use or
> distribute the work.

Yes, but if the copyright holder forgets to do so because of ignorance
of the law [i.e. he or she thinks that copyright law already gives him
or her that control], I doubt the courts will be sympathetic.
0
Jordan
3/6/2006 9:04:25 PM
On 2006-03-06, David Schwartz <davids@webmaster.com> wrote:
>
> "Jordan Abel" <random832@gmail.com> wrote in message 
> news:slrne0ok74.1eg8.random832@random.yi.org...
>
>> On 2006-03-06, David Schwartz <davids@webmaster.com> wrote:
>
>>>     The right to restrict the distribution of derivative works is also 
>>> *NOT*
>>> one of the rights granted under copyright law.
>
>> Though there is a right to restrict the _creation_ of them, which is
>> almost as good in practice.
>
>     No, not at all. For example, suppose I get the right to create a 
> derivative work somehow, say through first sale or fair use. If you had an 
> independent right to restrict the distribution of derivative works, I could 
> not distribute that derivative work after I created. Since you don't, I may 
> be able to.

The right to create derivative works does NOT come with first sale. And,
in the fair use case, you'd have to defend the particular derivative
work you made as being fair use.

The main issue here is whether it's a derivative work at all. [which, as
far as I can tell, you and I agree that it is]

>     It gets really complicated, unfortunately. But unless you want to get 
> into ugly nitty gritty legal details, you'll have to take my word for it. ;)
>
>     DS
>
>
0
Jordan
3/6/2006 9:07:15 PM
On 2006-03-06, David Schwartz <davids@webmaster.com> wrote:
>
> "stork" <stork@storkyak.com> wrote in message 
> news:1141658050.992153.297870@j33g2000cwa.googlegroups.com...
>
>> Stork replied to:
>
>>> For example, why can't I write a
>>>poem, drop copies of it from an airplane, and then sue/charge everyone who
>>>eads it?
>
>> Once again.  If I drop the poems from an airplane, everyone can read
>> them.  But, if someone copies that work or attempts to sell copies of
>> it, that's an IP violation.
>
>     Right, however, you can't restrict *use* or anything that's necessary 
> for use. You can't restrict transfers, that is, you can't stop me from 
> giving my copy of the poem to someone else.
>
>>>Why should I have to give up any rights to charge for use just
>>>ecause I give copies of my poem to everyone?
>>
>> You don't.  You gave away 500 poems, and thus, have 500 free users.
>> But those users cannot go and fire up a printing press to crank them
>> out.  If I give you the latest Stephen King book, you don't get to just
>> put it on the internet as fair use.
>
>     But that's not the case here. The case here is that you give away an 
> unlimited number of copies and authorize third parties to similarly give 
> away unlimited numbers of copies. At that point, everyone is a licensee of 
> your work.

well... not exactly. Suppose that there exists a set of people who
didn't receive an original copy, and who no-one who did receive an
original copy [nor anyone who received later-generation copies] wants to
give a copy. Also, suppose that no-one puts the poem up on the internet
for download. Effectively, that set of people are not licensees of the
work.
0
Jordan
3/6/2006 9:11:18 PM
Stork replies to:

>But your position is totally illogical, since it grants authors the
>right to post a copy of their work on a billboard and then sue everyone who
>reads it.

No it doesn't, any more than it would allow you to go and video tape a
TV show then sell it on the internet.

0
tbandrow
3/6/2006 9:14:09 PM
"Jordan Abel" <random832@gmail.com> wrote in message 
news:slrne0p9ar.2evf.random832@random.yi.org...

>>     No, not at all. For example, suppose I get the right to create a
>> derivative work somehow, say through first sale or fair use. If you had 
>> an
>> independent right to restrict the distribution of derivative works, I 
>> could
>> not distribute that derivative work after I created. Since you don't, I 
>> may
>> be able to.

> The right to create derivative works does NOT come with first sale. And,
> in the fair use case, you'd have to defend the particular derivative
> work you made as being fair use.

    It does if it's part of ordinary use. How do you use the source code to 
a library? You compile it. You develop applications that use it. You run 
those applications. So all of that is ordinary use, even if it creates, in 
the process, things that might be derivative works.

> The main issue here is whether it's a derivative work at all. [which, as
> far as I can tell, you and I agree that it is]

    In some cases yes, in some cases no. In most cases, I would argue that 
an applications that links to a library is likely not a derivative work of 
that library because "linking" is not a creative process, it's legally mere 
aggregation.

    Again, my position is that except for those exceptions codified in law 
(such as translation, which was presumably intended to be always creative, 
machine translation being beyond imagination when that was written), a 
derivative work must *creatively* take portions of the original work. 
Linking, compressing, archiving together or other *automated* combinations 
form aggregates, not derivatives.

    It's kind of hard to say though. Most of the law defining what is or is 
not a derivative work is for two purposes neither of which is this one. 
Sometimes it's to determine whether it was or wasn't legal to *create* that 
work, given that it wasn't ordinary use, which is not this case because the 
creation occurs in the process of ordinary use. Sometimes it's to determine 
whether the supposed derivative work is sufficiently creative to justify 
giving any rights to the person who created the derivative work, which is 
not this case either because we don't care how much new material is in the 
work but how much old material.

    There is very little case law I know of looking at whether a work is 
derivative for the purpose of determining if it can or can't be distributed 
to licensees of the original work. In the very few cases I know of (none of 
which were directly this issue though) it was held that there is no 
independent right to restrict the distribution of derivative works; however, 
most of those cases were polluted by "necessary step" arguments that don't 
apply here.

    Perhaps nobody knows what courts will ultimately hold. As far as I know, 
there is nothing definitive. My bet will be courts will hold that automatic 
combinations do not create derivative works and that lawfully created 
derivative works may be distributed to licensee of the original work with 
permission of the person who created the derivative. (But as this is also 
exactly what I *want* courts to hold, that may pollute my objectivity.)

    DS


0
David
3/6/2006 9:24:28 PM
"Jordan Abel" <random832@gmail.com> wrote in message 
news:slrne0p9ie.2evf.random832@random.yi.org...

> well... not exactly. Suppose that there exists a set of people who
> didn't receive an original copy, and who no-one who did receive an
> original copy [nor anyone who received later-generation copies] wants to
> give a copy. Also, suppose that no-one puts the poem up on the internet
> for download. Effectively, that set of people are not licensees of the
> work.

    Correct. The GPL does not require you to give the public access to a 
work. However, you cannot prevent a recipient from providing public access.

    In all of the cases we were talking about, there is someone willing to 
give anyone who wants it a copy of the *original* work. So it doesn't really 
matter.

    DS


0
David
3/6/2006 9:25:28 PM
<tbandrow@storkyak.com> wrote in message 
news:1141679649.897601.8680@j33g2000cwa.googlegroups.com...

> Stork replies to:

>>But your position is totally illogical, since it grants authors the
>>right to post a copy of their work on a billboard and then sue everyone 
>>who
>>reads it.

> No it doesn't, any more than it would allow you to go and video tape a
> TV show then sell it on the internet.

    I don't understand how this is a reply to my criticism. If "it" is your 
position, I never said your position would let people sell copies, I said it 
would give authors rights that make no sense, not that it would take away 
rights they should have.

    DS


0
David
3/6/2006 9:26:14 PM
Stork replied to:

>The GPL *grants* you the right to sell copies of GPL'd software.
>Otherwise every company that sells Linux CDs would be violating copyright.

You aren't selling the software, you are charging for a distribution.
The GPL is pretty clear on that.

Yes or No.  Under current copyright law, do you think you can take a
Linux source, create your own closed source product with it, and sell
it?

Yes or No?

0
tbandrow
3/7/2006 2:02:42 AM
>I never said your position would let people sell copies, I said it
>would give authors rights that make no sens

Rights that do not make sense to you, because you continually misapply
them with FUD like counter arguments.

>not that it would take away
rights they should have.

"should have".  According to you.  It's always the people that don't
create that want to strip the rights from people that do.

0
tbandrow
3/7/2006 2:04:02 AM
Stork replied to:

>  It gets really complicated, unfortunately. But unless you want to get
>into ugly nitty gritty legal details, you'll have to take my word for it. ;)

No we don't.

0
tbandrow
3/7/2006 2:05:24 AM
Stork replied to:

>You have the right to authorize who may prepare derivative works but
>*NOT* who can distribute them. This makes a *huge* difference.

Flat out wrong.  If we were talking about books or movies, I would
agree with you.  The devil is in the practice.  You have the right to
authorize who may prepare derivitive works.  You have the right to
decide who may distribute your original work.  In software, a person
has to distribute your software often times to create their derived
work, so, in practical terms, you DO have the right to decide who can
distribute your derivitive work.

>Who was talking about rental, leasing or lending?

The law is intended to cover a transfer to third parties for trade, of
which rental, leasing, or lending is a part.  IT's intended to protect
against copying without the author's intent.  I think it's reasonable
that a pro-private property SCOTUS would agree - I mean, do you expect
the likes of Scalia, Roberts, Thomas and Alito to strike down anything
that defines additional private property?

Again, answer this:

Yes or No. Does copyright law grant you the right to take a GPL
product, make a closed source product out of it, and sell it, without
having to adhere to the terms of the GPL from your work that uses GPL
work?

Yes or No.  Let's get back to the original question.  Do you have to
get a seperate license or special permission from any author, such as
Trolltech, in order to create a closed source product?

0
tbandrow
3/7/2006 2:14:11 AM
Stork replied to:

>     Right, however, you can't restrict *use* or anything that's necessary
> for use. You can't restrict transfers, that is, you can't stop me from
> giving my copy of the poem to someone else.

Except in the case of computer software if it is a for profit transfer,
or rent, or lease, etc.

>     But that's not the case here. The case here is that you give away an
> unlimited number of copies and authorize third parties to similarly give
> away unlimited numbers of copies. At that point, everyone is a licensee of
> your work.

I didn't give away an unlimited amount of copies.  I only authorized
those people to receive copies if they agreed to make open source
solutions themselves with it.

>     I have been so totally precise about what I mean by "use" that I can't
> imagine how you could say that it is confused. "Use" means the ordinary,
> reasonable, expected way that a work would be employed, including anything
> necessary to do that. For a software library, this would include developing
> applications that use that library.
>

Sure, so long as you adhere to the terms of use of the library, as
prescribed by the copyright holder, and in the case of GPL holders, you
either pay them via a separate license, like Trolltech, or, you make
your own work open source!

0
stork
3/7/2006 3:58:00 AM
Stork replied to:

> I don't see what the public display issue has to do with anything

Public display is everything about this argument.  A building is
copyrightable, as is a sporting event, or a concert, or a play in the
park, or, more specifically, a sign on a billboard.  Just because you
display something publicly, does not give the public the right to any
more than to see the display.

If what you argued held true, you would be able to, under your own
non-standard definition of "first use", be able to record TV shows off
of TV and resell them.  Downloading something over an internet,
broadcasting it over the airwaves, seeing a show in a park, are all
public displays.  You don't have any rights to that display, other than
to see it.

0
stork
3/7/2006 4:03:19 AM
Stork replied to:

> The way TrollTech gets to tell me what I can and can't do is by copyright and
>copyright only applies to derived works.

Nope, they look at your application redistributing their stuff, then
they haul you into court, and say, "jeez, this guy has our source code,
he has our actual library", and they will side one of two ways.   They
will either rule in favor of Trolltech, which will further greater
industry and capital formation in the United States, or, they will rule
in your favor, establishing legal precedence to resell TV broadcasts,
redistribute songs broadcast on the radio, let people reuse images on
billboards willy nilly, let alone go against all of what IBM,
Microsoft, Sun, and Apple all view about computer software, and trash
nearly 500 billion dollars of the US economy.  I think you lose in that
case.  In fact, I don't even see how you could possibly win. 90% of the
time, the courts will err on the side of private property, regardless
of whether they are Democrat or Republicans.

0
stork
3/7/2006 4:07:27 AM
Stork replied to:

> It doesn't matter. TrollTech can refer to "code written by people who
>have ever heard of TrollTech" and it would have no legal force

What you can't see is that if you go and say that anyone can create a
closed source product from an open source one, legally, you will have
killed open source.  Open source depends on strong intellectual
property rights for the author.  It's already illegal under the DMCA to
reverse engineer many things, and another round to tighten that up is
working its way through the congress.  So, you make it possible to
subvert open source, while at the same time stacking the deck in favor
of closed source systems, you will have killed Linux, and your freedom
will be reduced to deciding which of two screen savers that come with
Vista that you like best, subject to the click through license that
Microsoft provides.

0
stork
3/7/2006 4:13:28 AM
On 2006-03-07, stork <stork@storkyak.com> wrote:
> Stork replied to:

Please attribute PROPERLY.

>> I don't see what the public display issue has to do with anything
>
> Public display is everything about this argument.  A building is
> copyrightable, as is a sporting event, or a concert, or a play in the
> park, or, more specifically, a sign on a billboard.  Just because you
> display something publicly, does not give the public the right to any
> more than to see the display.
>
> If what you argued held true, you would be able to, under your own
> non-standard definition of "first use", be able to record TV shows off
> of TV and resell them.  Downloading something over an internet,
> broadcasting it over the airwaves, seeing a show in a park, are all
> public displays.  You don't have any rights to that display, other than
> to see it.

You have the right to distribute GPL'ed software because the license
grants that right. You have the right to create a program using it if
it is a library because (he says) that's part of "normal use" which is
protected by copyright law, the same as loading a program up into main
RAM to execute. These are two separate rights and you're using a
ridiculous argument that seems to be against the former [obviously a
messed-up argument] when your issue seems to be with the latter claim.
0
Jordan
3/7/2006 5:11:18 AM
On 2006-03-07, tbandrow@storkyak.com <tbandrow@storkyak.com> wrote:
> Flat out wrong.  If we were talking about books or movies, I would
> agree with you.  The devil is in the practice.  You have the right to
> authorize who may prepare derivitive works.  You have the right to
> decide who may distribute your original work.  In software, a person
> has to distribute your software often times to create their derived
> work, so, in practical terms, you DO have the right to decide who can
> distribute your derivitive work.

No, because you have granted an unconditional and irrevocable license to
dsitribute your original work, and you can't terminate that license
based on the creation of a legitimate derivative work. The ONLY
questions here is whether A) a program created with a library is a
derivative work at all [i claim not] or B) if it _is_ a derivative work,
whether you can prohibit someone from making such works when your own
product is a library intended for that purpose [other person in this
thread claims not].

> Again, answer this:
>
> Yes or No. Does copyright law grant you the right to take a GPL
> product, make a closed source product out of it, and sell it, without
> having to adhere to the terms of the GPL from your work that uses GPL
> work?

"Make a closed source product out of it" is a loaded term. If I make a
shell, and I use readline to... well, read lines, and I don't modify any
readline code in this process, have I _REALLY_ "made a closed source
product out of" readline?  Keep in mind that the header files and link
tables involved are uncopyrightable.

> Yes or No.  Let's get back to the original question.  Do you have to
> get a seperate license or special permission from any author, such as
> Trolltech, in order to create a closed source product?
0
Jordan
3/7/2006 5:16:20 AM
On 2006-03-07, stork <stork@storkyak.com> wrote:
> Stork replied to:
>
>> The way TrollTech gets to tell me what I can and can't do is by copyright and
>>copyright only applies to derived works.
>
> Nope, they look at your application redistributing their stuff, then
> they haul you into court, and say, "jeez, this guy has our source code,
> he has our actual library", and they will side one of two ways.   They
> will either rule in favor of Trolltech, which will further greater
> industry and capital formation in the United States, or, they will rule
> in your favor, establishing legal precedence to resell TV broadcasts,
> redistribute songs broadcast on the radio, let people reuse images on
> billboards willy nilly, 

Such a ruling would NOT establish such a precedent. A TV broadcast is
not under the GPL. A song is not under the GPL. The GPL grants specific
priveliges to the recepient that are the CORE of this argument.

> let alone go against all of what IBM, Microsoft, Sun, and Apple all
> view about computer software, and trash nearly 500 billion dollars of
> the US economy.

Actually, if such a ruling were made, Microsoft wins out. And besides,
none of those vendors have ever really presumed to be able to restrict
developers from writing applications using their APIs.

> I think you lose in that case.  In fact, I don't even see how you
> could possibly win. 90% of the time, the courts will err on the side
> of private property, regardless of whether they are Democrat or
> Republicans.
0
Jordan
3/7/2006 5:19:16 AM
Stork replied to:

>These are two separate rights and you're using a ridiculous argument that seems to be >against the former

Not at all.  But let's cut to the chase:

Yes or no.  Do you feel you have the right to create closed source
applications out of a GPL'd work?

0
stork
3/7/2006 11:03:56 AM
Stork replied to:

>"Make a closed source product out of it" is a loaded term

Just answer the goddamned question.  Yes or no.  Can you make a closed
source shrink wrapped licensed product out of a GPL library or code,
and can you redistribute that library with your shrink wrapped
solution?

>Keep in mind that the header files and link tables involved are uncopyrightable

I don't agree with that.  Header files are definitely copywritable.
Link tables, maybe / maybe not.

0
stork
3/7/2006 11:16:52 AM
Stork replied to:

>Such a ruling would NOT establish such a precedent. A TV broadcast is
>not under the GPL. A song is not under the GPL. The GPL grants specific
>priveliges to the recepient that are the CORE of this argument.

And it requires some conditions of the recipient as well.  Before we go
further, I want to make sure you are not in the same boat as David.  Do
you feel that you have the right to create a closed source application
based on a GPL'd work, and sell that application via a normal
commercial license?  That's David's ultimate destination, and I just
think that's f--- up.

If so, what legalism gives you the right to subvert it?

Is it the GPL itself, or like David, can you subvert the GPL innately
under what he considers to be copyright law, using a "I don't have to
pay my taxes because the IRS is illegal" like chain of reasoning.

0
stork
3/7/2006 11:45:43 AM
Stork replied to:

>If you *need* to use a
>particular chunk of code to get something done, and there is no other way to
>do that same thing, then it is *not* copyrightable material

It's not self refuting at all!  There is more than one way to get
something done.  Use another library, or write your own.  Your
exclusionary radius is too small.

0
tbandrow
3/7/2006 3:20:41 PM
Stork replied to:

>You are ignoring the rules of rational debate. You cannot disagree with
>a reasoned argument because you don't like what I'm going to do with it
>later.

Your reasoning is flawed, therefor, you are not debating rationally!

>Pot. Kettle. Black. 

Absolutely.

0
tbandrow
3/7/2006 3:21:54 PM
-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1

tbandrow@storkyak.com writes:

> Stork replied to:
>
>>You are ignoring the rules of rational debate. You cannot disagree with
>>a reasoned argument because you don't like what I'm going to do with it
>>later.
>
> Your reasoning is flawed, therefor, you are not debating rationally!
>
>>Pot. Kettle. Black. 
>
> Absolutely.

This is unproductive and off-topic.  Could you please take this to
private mail?


Thanks,
Roger

- -- 
Roger Leigh
                Printing on GNU/Linux?  http://gutenprint.sourceforge.net/
                Debian GNU/Linux        http://www.debian.org/
                GPG Public Key: 0x25BFB848.  Please sign and encrypt your mail.
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.4.2 (GNU/Linux)
Comment: Processed by Mailcrypt 3.5.8+ <http://mailcrypt.sourceforge.net/>

iD8DBQFEDbasVcFcaSW/uEgRAhpeAJ40SAbNpYlSaXGSkvEDIHei4P4qRgCfa0pd
fm1Ps5AlujhuWPMadPM26cM=
=QUD9
-----END PGP SIGNATURE-----
0
Roger
3/7/2006 4:37:03 PM
On 2006-03-07, stork <stork@storkyak.com> wrote:
> Stork replied to:
>
>>"Make a closed source product out of it" is a loaded term
>
> Just answer the goddamned question.  Yes or no.  Can you make a closed
> source shrink wrapped licensed product out of a GPL library or code,
> and can you redistribute that library with your shrink wrapped
> solution?

But that's NOT the question. The question is if you can make a closed
source product that USES a GPL library. "out of" implies your product is
based on the library code itself.

My argument is that YES, you CAN make a closed-source, or otherwise
non-gpl-compatible [the GPL doesn't differentiate between "closed source
shrink wrapped licensed product", vs, say, BSD with advertising clause,
or even the least restrictive of the creative commons licenses] piece of
software that uses an API defined by a GPL library.

For a real-world example, consider binary kernel modules for linux.

Or, hell, even closed-source PROGRAMS for linux. [glibc may be LGPL, but
the kernel itself is GPL]

And, if you can do that (even if you don't accept that, you can still
accept that given that step the next step is perfectly legal), and you
do that, then YES you can redistribute the library. Because once the
right to make the closed-source program itself is established,
distributing the library with it is "mere aggregation". It's no
different in that case from distributing the library by itself, or with
a closed-source program that does NOT use it.

>>Keep in mind that the header files and link tables involved are
>>uncopyrightable
>
> I don't agree with that.  Header files are definitely copywritable.

What copyrightable information do these contain _THAT GETS INCLUDED INTO
THE PROGRAM_? Let's look at _all_ the information that can be included
in a header file, and rule out what doesn't get included into the
program. Comments clearly aren't included into the binary program -
neither are macros that define to nothing. I doubt you would argue that
macros that define to a constant, other than perhaps those that expand
to a string literal, are copyrightable. That leaves declarations: The
declarations are instructions to tell the compiler how to do something,
they're neither useful for nor included in the output of the compiler.
And last, but not least, function-like macros that expand to an
expression. That is perhaps the only kind of header file content that
can be argued to be copyrightable. However, it is likely that it is the
only code that can accomplish a given purpose, and is [as explained
elsewhere in this thread] uncopyrightable for that reason

(My point is, sure, you CAN put anything in a header file, but the kind
of things that get put in files we call "header files" tend not to be
things that are copyrightable)

> Link tables, maybe / maybe not.
0
Jordan
3/7/2006 6:44:55 PM
On 2006-03-07, stork <stork@storkyak.com> wrote:
> Stork replied to:
>
>>Such a ruling would NOT establish such a precedent. A TV broadcast is
>>not under the GPL. A song is not under the GPL. The GPL grants specific
>>priveliges to the recepient that are the CORE of this argument.
>
> And it requires some conditions of the recipient as well.  Before we go
> further, I want to make sure you are not in the same boat as David.  Do
> you feel that you have the right to create a closed source application
> based on a GPL'd work,

No. I assert that a program using the exported API from a GPL'd library
is NOT "based on a GPL'd work". That you don't see a difference says
more about you than it does about reality.

> and sell that application via a normal commercial license?  That's
> David's ultimate destination,

No, it is not. You are misrepresenting normal use of a library as being
equivalent to making a closed-source version of the library itself.

> and I just think that's f--- up.

Because you have blinders on that make you unable to see the CLEAR moral
difference between, writing a program FOR something and writing a
program BASED ON that something.
0
Jordan
3/7/2006 6:47:50 PM
On 2006-03-07, tbandrow@storkyak.com <tbandrow@storkyak.com> wrote:
> Stork replied to:
>
>>If you *need* to use a
>>particular chunk of code to get something done, and there is no other way to
>>do that same thing, then it is *not* copyrightable material
>
> It's not self refuting at all!  There is more than one way to get
> something done.  Use another library, or write your own.  Your
> exclusionary radius is too small.

It's the same one used in the law itself. The courts didn't tell
Accolade to make their own game console instead of doing what needed to
be done to make Sega Genesis games.
0
Jordan
3/7/2006 6:48:54 PM
Stork replied to:

>The courts didn't tell Accolade to make their own game console instead of doing what needed to
>be done to make Sega Genesis games.

That's not the same thing at all.

0
tbandrow
3/7/2006 8:26:44 PM
Stork replied to:

>No. I assert that a program using the exported API from a GPL'd library
>is NOT "based on a GPL'd work". That you don't see a difference says
>more about you than it does about reality

The difference is that you are splitting hairs.  The end result is that
you are receiving the benefit of that library.  Do you feel that you
are entitled to distribute that library itself with your closed source
application?

>No, it is not. You are misrepresenting normal use of a library as being
>equivalent to making a closed-source version of the library itself.

It's not normal use.

>Because you have blinders on that make you unable to see the CLEAR >moral difference between, writing a program FOR something and writing a
>program BASED ON that something.

It's the same thing!  You just can't think straight past your own smoke
screen and see it for what it is.  You are receiving the benefit of
that library.  Whether you are cutting and pasting source, or making
API calls, you are specializing that library or set of libraries to a
particular task.

0
tbandrow
3/7/2006 8:30:15 PM
On 2006-03-07, tbandrow@storkyak.com <tbandrow@storkyak.com> wrote:
> Stork replied to:
>
>>No. I assert that a program using the exported API from a GPL'd library
>>is NOT "based on a GPL'd work". That you don't see a difference says
>>more about you than it does about reality
>
> The difference is that you are splitting hairs.

There _IS_ a difference, and YOU are refusing to recognize a legitimate
distinction.

Linux kernel binary modules do it just fine, why can't other programs?

> The end result is that you are receiving the benefit of that library.

So? Who says that you have to open-source your programs to "receive the
benefit of" a library? What does "receiving the benefit of" even mean?
It's not a term in law, as far as I know (IANAL)

> Do you feel that you are entitled to distribute that library itself
> with your closed source application?

It is UNDISPUTED that i am entitled to distribute the library with a
non-GPL application that does not link to it. You have NOT attempted to
justify the claim that having the non-GPL application link to it changes
anything.

>>No, it is not. You are misrepresenting normal use of a library as being
>>equivalent to making a closed-source version of the library itself.
>
> It's not normal use.

This is in dispute, please justify the claim rather than just stating
your denial as if it were an accepted fact.

>>Because you have blinders on that make you unable to see the CLEAR
>>moral difference between, writing a program FOR something and writing
>>a program BASED ON that something.
>
> It's the same thing!  You just can't think straight past your own smoke
> screen and see it for what it is.  You are receiving the benefit of
> that library.  Whether you are cutting and pasting source, or making
> API calls, you are specializing that library or set of libraries to a
> particular task.

IT IS NOT THE SAME THING! NOBODY EXCEPT YOU CLAIMS IT IS THE SAME THING.
IF IT WERE THE SAME THING, THE LGPL COULD NOT MAKE THE DISTINCTION THAT
IT MAKES BETWEEN THE TWO CASES!
0
Jordan
3/7/2006 8:46:00 PM
Jordan Abel wrote:

> On 2006-03-07, tbandrow@storkyak.com <tbandrow@storkyak.com> wrote:
>> Stork replied to:
>>
>>>No. I assert that a program using the exported API from a GPL'd library
>>>is NOT "based on a GPL'd work". That you don't see a difference says
>>>more about you than it does about reality
>>
>> The difference is that you are splitting hairs.
> 
> There _IS_ a difference, and YOU are refusing to recognize a legitimate
> distinction.
> 
> Linux kernel binary modules do it just fine, why can't other programs?
> 

Because for the linux kernel special exceptions are made
Just read for once the GPL, the kernel license and some other stuff

You /can/ read, can you?

< snip >
-- 
Microsoft: The company that made email dangerous
And web browsing. And viewing pictures. And...

0
Peter
3/7/2006 9:01:15 PM
<tbandrow@storkyak.com> wrote in message 
news:1141696962.686998.227310@j33g2000cwa.googlegroups.com...

> Yes or No.  Under current copyright law, do you think you can take a
> Linux source, create your own closed source product with it, and sell
> it?
>
> Yes or No?

    Yes or no, depending upon what you mean by "take a Linux source".

    If you mean do I think I can make a binary of the Linux kernel, with or 
without changes, and distribute it without making the source code available. 
No.

    If you mean do I think I can make some program that uses the Linux 
kernel and distribute it without making the source code available (even if I 
also distribute the Linux kernel, assuming I distribute the source to the 
kernel). Yes. Of course.

    DS


0
David
3/7/2006 9:01:58 PM
<tbandrow@storkyak.com> wrote in message 
news:1141697124.942061.6890@z34g2000cwc.googlegroups.com...

> Stork replied to:

>>  It gets really complicated, unfortunately. But unless you want to get
>>into ugly nitty gritty legal details, you'll have to take my word for it. 
>>;)

> No we don't.

    It is dishonest to cut this remark out of its context.

    DS


0
David
3/7/2006 9:02:40 PM
"stork" <stork@storkyak.com> wrote in message 
news:1141703880.777360.210040@e56g2000cwe.googlegroups.com...

> Stork replied to:

>>     Right, however, you can't restrict *use* or anything that's necessary
>> for use. You can't restrict transfers, that is, you can't stop me from
>> giving my copy of the poem to someone else.

> Except in the case of computer software if it is a for profit transfer,
> or rent, or lease, etc.

    You can restrict rentals and leasings, but you cannot restrict a 
for-profit transfer. If you legally acquire a piece of software (and there's 
no EULA, shrink-wrap, click-through or signed contract), you can sell it.

>>     But that's not the case here. The case here is that you give away an
>> unlimited number of copies and authorize third parties to similarly give
>> away unlimited numbers of copies. At that point, everyone is a licensee 
>> of
>> your work.

> I didn't give away an unlimited amount of copies.  I only authorized
> those people to receive copies if they agreed to make open source
> solutions themselves with it.

    This is the mistake you keep making. You think that you must agree to 
the GPL as a condition of receiving the work. But you don't. The GPL does 
not authorize *receiving* copies but making or sending them. Until you 
understand this, you will keep uttering nonsense.

>>     I have been so totally precise about what I mean by "use" that I 
>> can't
>> imagine how you could say that it is confused. "Use" means the ordinary,
>> reasonable, expected way that a work would be employed, including 
>> anything
>> necessary to do that. For a software library, this would include 
>> developing
>> applications that use that library.

> Sure, so long as you adhere to the terms of use of the library, as
> prescribed by the copyright holder, and in the case of GPL holders, you
> either pay them via a separate license, like Trolltech, or, you make
> your own work open source!

    I'm sorry, I don't know what to say other than that you need to 
understand the law in order to be able to say something that makes any kind 
of sense. I've refuted this claim many, many times now and you still keep 
repeating it without any indication that you even understand my criticism of 
it.

    DS


0
David
3/7/2006 9:06:00 PM
On 2006-03-07, Peter K�hlmann <peter.koehlmann@t-online.de> wrote:
> Jordan Abel wrote:
>
>> On 2006-03-07, tbandrow@storkyak.com <tbandrow@storkyak.com> wrote:
>>> Stork replied to:
>>>
>>>>No. I assert that a program using the exported API from a GPL'd library
>>>>is NOT "based on a GPL'd work". That you don't see a difference says
>>>>more about you than it does about reality
>>>
>>> The difference is that you are splitting hairs.
>> 
>> There _IS_ a difference, and YOU are refusing to recognize a legitimate
>> distinction.
>> 
>> Linux kernel binary modules do it just fine, why can't other programs?
>> 
>
> Because for the linux kernel special exceptions are made

That is not a special exception. It is a declared interpretation of the
GPL. An example of a special exception would be the extra clause on
libgcc.

> Just read for once the GPL, the kernel license and some other stuff
>
> You /can/ read, can you?
>
> < snip >
0
Jordan
3/7/2006 9:08:31 PM
"stork" <stork@storkyak.com> wrote in message 
news:1141729436.392513.208800@j33g2000cwa.googlegroups.com...

> Yes or no.  Do you feel you have the right to create closed source
> applications out of a GPL'd work?

    It depends what you mean by "out of". If you mean can I create a closed 
source version of a GPL'd application or library, no. If you mean can I 
create a closed source product that uses a GPL'd application or library, yes 
(depending upon how much of the GPL'd work it actually contains).

    DS


0
David
3/7/2006 9:09:05 PM
On 2006-03-07, stork <stork@storkyak.com> wrote:
> Sure, so long as you adhere to the terms of use of the library,

The GPL does not prescribe any terms of use.
0
Jordan
3/7/2006 9:09:51 PM
"stork" <stork@storkyak.com> wrote in message 
news:1141730212.071293.158400@i40g2000cwc.googlegroups.com...

> Stork replied to:

>>"Make a closed source product out of it" is a loaded term

> Just answer the goddamned question.  Yes or no.  Can you make a closed
> source shrink wrapped licensed product out of a GPL library or code,
> and can you redistribute that library with your shrink wrapped
> solution?

    Yes. You are simply using the library and everyone who recieves your 
product is using the library. Use is not restricted under the GPL.

>>Keep in mind that the header files and link tables involved are 
>>uncopyrightable
>
> I don't agree with that.  Header files are definitely copywritable.
> Link tables, maybe / maybe not.

    Did you read the Lexmark decision?

    DS


0
David
3/7/2006 9:09:59 PM
On 2006-03-07, David Schwartz <davids@webmaster.com> wrote:
>
> "stork" <stork@storkyak.com> wrote in message 
> news:1141729436.392513.208800@j33g2000cwa.googlegroups.com...
>
>> Yes or no.  Do you feel you have the right to create closed source
>> applications out of a GPL'd work?
>
>     It depends what you mean by "out of".

He's equivocating, so he means both.

> If you mean can I create a closed source version of a GPL'd
> application or library, no. If you mean can I create a closed source
> product that uses a GPL'd application or library, yes (depending upon
> how much of the GPL'd work it actually contains).
0
Jordan
3/7/2006 9:10:23 PM
"stork" <stork@storkyak.com> wrote in message 
news:1141729954.049351.140630@i40g2000cwc.googlegroups.com...

> And it requires some conditions of the recipient as well.

    No, it does not. There are *NO* conditions you must agree to in order to 
receive a GPL'd work. The GPL is *NOT* an EULA, shrink-wrap agreement, 
contract, or click-through.

    DS


0
David
3/7/2006 9:10:57 PM
<tbandrow@storkyak.com> wrote in message 
news:1141744841.682794.191130@p10g2000cwp.googlegroups.com...

> Stork replied to:

>>If you *need* to use a
>>particular chunk of code to get something done, and there is no other way 
>>to
>>do that same thing, then it is *not* copyrightable material

> It's not self refuting at all!  There is more than one way to get
> something done.

    It is not whether the is more way to get "something" done, it is whether 
there is more than one way to get the *same* thing done.

> Use another library, or write your own.  Your
> exclusionary radius is too small.

    These are ways to express different ideas, not other ways to express the 
same idea. Did you read the Lexmark case?

    DS


0
David
3/7/2006 9:11:47 PM
<tbandrow@storkyak.com> wrote in message 
news:1141763415.204293.169420@p10g2000cwp.googlegroups.com...

> Stork replied to:

>>No. I assert that a program using the exported API from a GPL'd library
>>is NOT "based on a GPL'd work". That you don't see a difference says
>>more about you than it does about reality

> The difference is that you are splitting hairs.  The end result is that
> you are receiving the benefit of that library.  Do you feel that you
> are entitled to distribute that library itself with your closed source
> application?

    Please read section 0 of the GPL again. You are *supposed* to be able to 
get the full benefit of *using* the library without having to do anything 
special. With a GPL'd work, you get all the rights you would normally get if 
you bought a library without having to do anything special.

>>No, it is not. You are misrepresenting normal use of a library as being
>>equivalent to making a closed-source version of the library itself.

> It's not normal use.

    What is normal use of a library then?

>>Because you have blinders on that make you unable to see the CLEAR >moral 
>>difference between, writing a program FOR something and writing a
>>program BASED ON that something.

> It's the same thing!  You just can't think straight past your own smoke
> screen and see it for what it is.

    Your head is firmly up your ass here.

    Writing an application that uses a library is the normal and expected 
use of a library. It's pretty much the *only* thing you can do with a 
library.

> You are receiving the benefit of
> that library.  Whether you are cutting and pasting source, or making
> API calls, you are specializing that library or set of libraries to a
> particular task.

    If I let you borrow my car, can you switch engines? Driving the car to 
Newark and switching the engine are both specializing the use of my car to a 
particular task. The difference is that one is the ordinary and expected use 
of the car and the other is not.

    If you are right that they are equivalent, and one is ordinary use, then 
so is the other. So this actually hurts your overall argument. (And if 
neither is ordinary use of a library provided in source code form, please 
explain what is.)

    DS


0
David
3/7/2006 9:16:17 PM
-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1

Jordan Abel <random832@gmail.com> writes:

> On 2006-03-07, tbandrow@storkyak.com <tbandrow@storkyak.com> wrote:
>> Stork replied to:
>>
>>>No. I assert that a program using the exported API from a GPL'd library
>>>is NOT "based on a GPL'd work". That you don't see a difference says
>>>more about you than it does about reality
>>
>> The difference is that you are splitting hairs.
>
> There _IS_ a difference, and YOU are refusing to recognize a legitimate
> distinction.
>
> Linux kernel binary modules do it just fine, why can't other programs?

Binary kernel modules can't "do it just fine".  Kernel modules are
derived works, and hence subject to the GPL, in all but a few gray
areas, such as where the code is not dependent upon Linux interfaces
(e.g. AFS).  See, for example:

  http://kerneltrap.org/node/1735
  http://www.google.com/search?q=linux+binary+modules+legal


Regards,
Roger

- -- 
Roger Leigh
                Printing on GNU/Linux?  http://gutenprint.sourceforge.net/
                Debian GNU/Linux        http://www.debian.org/
                GPG Public Key: 0x25BFB848.  Please sign and encrypt your mail.
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.4.2 (GNU/Linux)
Comment: Processed by Mailcrypt 3.5.8+ <http://mailcrypt.sourceforge.net/>

iD8DBQFEDfhfVcFcaSW/uEgRAnzwAJ4h4mzPN5KfB8c9Lw4SHdOav2qmugCfQifC
ZDwI7rhecsH2FTKkBsEl+MM=
=BdRI
-----END PGP SIGNATURE-----
0
Roger
3/7/2006 9:17:32 PM
On 2006-03-07, Roger Leigh <${rleigh}@invalid.whinlatter.ukfsn.org.invalid> wrote:
>
> Jordan Abel <random832@gmail.com> writes:
>
>> On 2006-03-07, tbandrow@storkyak.com <tbandrow@storkyak.com> wrote:
>>> Stork replied to:
>>>
>>>>No. I assert that a program using the exported API from a GPL'd library
>>>>is NOT "based on a GPL'd work". That you don't see a difference says
>>>>more about you than it does about reality
>>>
>>> The difference is that you are splitting hairs.
>>
>> There _IS_ a difference, and YOU are refusing to recognize a legitimate
>> distinction.
>>
>> Linux kernel binary modules do it just fine, why can't other programs?
>
> Binary kernel modules can't "do it just fine".

Your claim is contrary to reality.

> Kernel modules are derived works,

Many are, but not by the simple fact that they are linked in, any more
than a user program is.

> and hence subject to the GPL, in all but a few gray areas, such as
> where the code is not dependent upon Linux interfaces (e.g. AFS).
> See, for example:
>
>   http://kerneltrap.org/node/1735
>   http://www.google.com/search?q=linux+binary+modules+legal
0
Jordan
3/7/2006 9:30:57 PM
"Roger Leigh" <${rleigh}@invalid.whinlatter.ukfsn.org.invalid> wrote in 
message news:87pskykl8z.fsf@hardknott.home.whinlatter.ukfsn.org...

> Binary kernel modules can't "do it just fine".  Kernel modules are
> derived works, and hence subject to the GPL, in all but a few gray
> areas, such as where the code is not dependent upon Linux interfaces
> (e.g. AFS).  See, for example:

    Have you read the Lexmark v. Static Control case? If using the kernel 
header files is the only way to make a kernel module (module changing names 
or orders which doesn't matter), then it's not copyrightable content. 
There's no one section I can cite that perfectly captures the entire 
argument, but this one will at least get you looking in the right direction:

For similar reasons, when external factors constrain the choice of 
expressive vehicle, the doctrine
of "sc�nes � faire"-"scenes," in other words, "that must be done"-precludes 
copyright protection. See

Twentieth Century Fox Film, 361 F.3d at 319-20; see generally Nimmer � 
13.03[B][4]. In the literary

context, the doctrine means that certain phrases that are "standard, stock, 
.. . . or that necessarily follow from

a common theme or setting" may not obtain copyright protection. Gates 
Rubber, 9 F.3d at 838. In the

computer-software context, the doctrine means that the elements of a program 
dictated by practical

realities-e.g., by hardware standards and mechanical specifications, 
software standards and compatibility

requirements, computer manufacturer design standards, target industry 
practices, and standard computer

programming practices-may not obtain protection. Id. (citing case examples); 
see Sega Enters., 977 F.2d

at 1524 ("To the extent that a work is functional or factual, it may be 
copied."); Brown Bag Software v.

Symantec Corp., 960 F.2d 1465, 1473 (9th Cir. 1992) (affirming district 
court's finding that "[p]laintiffs

may not claim copyright protection of an . . . expression that is, if not 
standard, then commonplace in the

computer software industry"). As "an industry-wide goal," programming 
"[e]fficiency" represents an

external constraint that figures prominently in the copyrightability of 
computer programs. Altai, 982 F.2d

at 708.

    DS


0
David
3/7/2006 9:37:41 PM
-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1

Jordan Abel <random832@gmail.com> writes:

> On 2006-03-07, Roger Leigh <${rleigh}@invalid.whinlatter.ukfsn.org.invalid> wrote:
>>
>> Jordan Abel <random832@gmail.com> writes:

>>> Linux kernel binary modules do it just fine, why can't other programs?
>>
>> Binary kernel modules can't "do it just fine".
>
> Your claim is contrary to reality.

You are adding a lot of noise to this group, and precious little
information.  Please, think before posting, and do some background
reading (do read the kerneltrap link).  This is not a new question,
and has been asked many times over the last decade; you'll find it all
in the group archives over at groups.google.com.  There are also
hundreds of web pages available which discuss the issues at length,
which you would do well to look at.

This rather boring debate here does nothing to change the legal
issues, whatever your viewpoint.

Do: read the licences, the many resources on the subject, and get a
professional legal opinion if you have need of it.
Don't: Carry on the debate here.  This group is about application
development, not licensing flamewars.  We've seen it all before.


Regards,
Roger

- -- 
Roger Leigh
                Printing on GNU/Linux?  http://gutenprint.sourceforge.net/
                Debian GNU/Linux        http://www.debian.org/
                GPG Public Key: 0x25BFB848.  Please sign and encrypt your mail.
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.4.2 (GNU/Linux)
Comment: Processed by Mailcrypt 3.5.8+ <http://mailcrypt.sourceforge.net/>

iD8DBQFEDgE0VcFcaSW/uEgRAv+WAJ4ttax8GAHviEDngkAtt/bSw/CAfgCgyubB
KUE1nNplpokEWGtd2d1+1LM=
=wTut
-----END PGP SIGNATURE-----
0
Roger
3/7/2006 9:55:03 PM
Jordan Abel wrote:

> On 2006-03-07, Peter Köhlmann <peter.koehlmann@t-online.de> wrote:
>> Jordan Abel wrote:
>>
>>> On 2006-03-07, tbandrow@storkyak.com <tbandrow@storkyak.com> wrote:
>>>> Stork replied to:
>>>>
>>>>>No. I assert that a program using the exported API from a GPL'd library
>>>>>is NOT "based on a GPL'd work". That you don't see a difference says
>>>>>more about you than it does about reality
>>>>
>>>> The difference is that you are splitting hairs.
>>> 
>>> There _IS_ a difference, and YOU are refusing to recognize a legitimate
>>> distinction.
>>> 
>>> Linux kernel binary modules do it just fine, why can't other programs?
>>> 
>>
>> Because for the linux kernel special exceptions are made
> 
> That is not a special exception. It is a declared interpretation of the
> GPL. An example of a special exception would be the extra clause on
> libgcc.
> 

You should tell Linus Torvalds about that. He would certainly be somewhat
surprised
He stated that closed source *can* link against the kernel, although the
kernel is GPLed. It is a special exception for the linux kernel. Linux
Torvalds made it. He is in a position to do that

But closed source can *not* legally link against (for example) readline
Which is also GPLed. Everything linking against readline has to be GPLed
itself. If you don't agree, fine. Don't link against it. Write your own
implementation of readline

>> Just read for once the GPL, the kernel license and some other stuff
>>
>> You /can/ read, can you?
>>

Well, that question is superfluous by now. 
You can't read

And this "discussion" is, frankly, quite off-topic in this group
-- 
Ninety percent of the time things will turn out worse than you expect.
The other 10 percent of the time you had no right to expect so much.

0
Peter
3/7/2006 10:32:26 PM
Jordan Abel wrote:

> On 2006-03-07, stork <stork@storkyak.com> wrote:
>> Sure, so long as you adhere to the terms of use of the library,
> 
> The GPL does not prescribe any terms of use.

It does not. Only you want to redefine "use" to "linking to function calls /
API" of GPLed libraries.

Read the GPL. /If/ you can read
-- 
Just out of curiosity does this actually mean something or have some
of the few remaining bits of your brain just evaporated?

0
Peter
3/7/2006 10:34:55 PM
David Schwartz wrote:

> 
> "stork" <stork@storkyak.com> wrote in message
> news:1141730212.071293.158400@i40g2000cwc.googlegroups.com...
> 
>> Stork replied to:
> 
>>>"Make a closed source product out of it" is a loaded term
> 
>> Just answer the goddamned question.  Yes or no.  Can you make a closed
>> source shrink wrapped licensed product out of a GPL library or code,
>> and can you redistribute that library with your shrink wrapped
>> solution?
> 
>     Yes. You are simply using the library and everyone who recieves your
> product is using the library. Use is not restricted under the GPL.
> 

This is complete, utter bullshit
"Using" a library does in include "linking against it"

< snip >
-- 
I refuse to have a battle of wits with an unarmed person.

0
Peter
3/7/2006 10:39:52 PM
I honestly didn't mean to contribute more to this flame war.  I can
however not let the piece of misinformation below go by unanswered.

Peter K�hlmann <peter.koehlmann@t-online.de> writes:

> But closed source can *not* legally link against (for example) readline
> Which is also GPLed. Everything linking against readline has to be GPLed
> itself. If you don't agree, fine. Don't link against it. Write your own
> implementation of readline

Readline is an exceptionally bad example in this context.  There is a
BSD licensed source and binary compatible library called editline.  It
can work as a drop-in replacement for GNU readline without any need to
modify or even recompile any apps that use it.  Now tell me again,
which library are those apps "derived" from?

-- 
M�ns Rullg�rd
mru@inprovide.com
0
iso
3/7/2006 10:51:39 PM
"Peter K�hlmann" <peter.koehlmann@t-online.de> wrote in message 
news:dul1rf$hoq$02$2@news.t-online.com...

> It does not. Only you want to redefine "use" to "linking to function calls 
> /
> API" of GPLed libraries.

    What would "use" of a library supplied in the form of source code be 
then?

    DS


0
David
3/7/2006 11:22:03 PM
David Schwartz wrote:

> 
> "Peter K�hlmann" <peter.koehlmann@t-online.de> wrote in message
> news:dul1rf$hoq$02$2@news.t-online.com...
> 
>> It does not. Only you want to redefine "use" to "linking to function
>> calls /
>> API" of GPLed libraries.
> 
>     What would "use" of a library supplied in the form of source code be
> then?
> 
>     DS

Something different of what you think it might be

Hint: It is for people who actually have a brain. And are able to read.
*And* have reading comprehension. 
So you need not apply
-- 
For every complex problem, there is a solution that is simple, neat,
and wrong.                -- H. L. Mencken

0
Peter
3/7/2006 11:50:09 PM
On 2006-03-07, Roger Leigh <${rleigh}@invalid.whinlatter.ukfsn.org.invalid> wrote:
>
> Jordan Abel <random832@gmail.com> writes:
>
>> On 2006-03-07, Roger Leigh <${rleigh}@invalid.whinlatter.ukfsn.org.invalid> wrote:
>>>
>>> Jordan Abel <random832@gmail.com> writes:
>
>>>> Linux kernel binary modules do it just fine, why can't other programs?
>>>
>>> Binary kernel modules can't "do it just fine".
>>
>> Your claim is contrary to reality.
>
> You are adding a lot of noise to this group, and precious little
> information.

All I was saying is that the fact that they do exist fundamentally
contradicts your claim that they cannot.

> This rather boring debate here does nothing to change the legal
> issues, whatever your viewpoint.
[...]
> Don't: Carry on the debate here.  This group is about application
> development, not licensing flamewars.  We've seen it all before.

So kill the thread - it's just one thread.

Fine, i'll stop.
0
Jordan
3/8/2006 12:09:35 AM
On 2006-03-07, Peter K�hlmann <peter.koehlmann@t-online.de> wrote:
> Jordan Abel wrote:
>
>> On 2006-03-07, Peter K�hlmann <peter.koehlmann@t-online.de> wrote:
>>> Jordan Abel wrote:
>>>
>>>> On 2006-03-07, tbandrow@storkyak.com <tbandrow@storkyak.com> wrote:
>>>>> Stork replied to:
>>>>>
>>>>>>No. I assert that a program using the exported API from a GPL'd library
>>>>>>is NOT "based on a GPL'd work". That you don't see a difference says
>>>>>>more about you than it does about reality
>>>>>
>>>>> The difference is that you are splitting hairs.
>>>> 
>>>> There _IS_ a difference, and YOU are refusing to recognize a legitimate
>>>> distinction.
>>>> 
>>>> Linux kernel binary modules do it just fine, why can't other programs?
>>>> 
>>>
>>> Because for the linux kernel special exceptions are made
>> 
>> That is not a special exception. It is a declared interpretation of the
>> GPL. An example of a special exception would be the extra clause on
>> libgcc.
>
> You should tell Linus Torvalds about that. He would certainly be somewhat
> surprised He stated that closed source *can* link against the kernel,

He believes it can be. So do I. I never said his 'declared
interpretation' was any way incorrect.

> although the kernel is GPLed. It is a special exception for the linux
> kernel. Linux Torvalds made it. He is in a position to do that

I don't think he actually _is_ in a position to do that on behalf of all
the other contributors, if that were what he really did. I also hadn't
heard of an actual exception PERMITTING "derivative works", rather than
just a statement of what he interpreted to not be a derivative work. Quote
please?

> But closed source can *not* legally link against (for example) readline
> Which is also GPLed. 

FSF claims so. It's certainly not the first controversial claim they
have made.

> Everything linking against readline has to be GPLed itself. If you
> don't agree, fine. 

There's a difference between not agreeing with the policy and not
agreeing with their legal interpretation. Just because the GPL is "the
literary work of Richard Stallman" doesn't mean he gets to decide what
it means.

> And this "discussion" is, frankly, quite off-topic in this group

And it was winding down. You could have easily killfiled the thread, or
its three [until today] main participants.
0
Jordan
3/8/2006 12:13:40 AM
On 2006-03-07, Peter K�hlmann <peter.koehlmann@t-online.de> wrote:
> Jordan Abel wrote:
>
>> On 2006-03-07, stork <stork@storkyak.com> wrote:
>>> Sure, so long as you adhere to the terms of use of the library,
>> 
>> The GPL does not prescribe any terms of use.
>
> It does not. Only you want to redefine "use" to "linking to function calls /
> API" of GPLed libraries.

There's at least one person who does want to redefine that, but that's
not me. What I'm saying is that it isn't use or a derived work or
anything at all.

But, what do YOU think use of a library would include?

Thought experiment: If you have the [uncopyrightable] link tables and
function prototypes in hand, I suspect you could, in theory, make a
binary that links against a library that you have never seen even a
binary copy of. What's that a derivative work of?

> Read the GPL. /If/ you can read
0
Jordan
3/8/2006 12:16:51 AM
On 2006-03-07, Peter K�hlmann <peter.koehlmann@t-online.de> wrote:
> David Schwartz wrote:
>
>> 
>> "stork" <stork@storkyak.com> wrote in message
>> news:1141730212.071293.158400@i40g2000cwc.googlegroups.com...
>> 
>>> Stork replied to:
>> 
>>>>"Make a closed source product out of it" is a loaded term
>> 
>>> Just answer the goddamned question.  Yes or no.  Can you make a closed
>>> source shrink wrapped licensed product out of a GPL library or code,
>>> and can you redistribute that library with your shrink wrapped
>>> solution?
>> 
>>     Yes. You are simply using the library and everyone who recieves your
>> product is using the library. Use is not restricted under the GPL.
>> 
>
> This is complete, utter bullshit
> "Using" a library does in include "linking against it"

It's undisputed that you can distribute your non-gpl application in
source form and instruct your end-users to "link against it".  Clearly,
then, "use" includes linking against it. The only question here is
whether that linked binary is a derivative work.
0
Jordan
3/8/2006 12:18:43 AM
Stork replied to:

> This is the mistake you keep making. You think that you must agree to
>the GPL as a condition of receiving the work. But you don't

I'm not saying that AT ALL.  I'm saying that the distribution of a
computer work onto a public computer network is a performance or a
display just like a rock concert or a TV show, and that, as the
recipient of that show, you basically only have the right to view it.
Any additional rights you might want, you have to contact me to get,
or, you can choose to gain additional rights by abiding by the GPL.

>I'm sorry, I don't know what to say other than that you need to
>understand the law in order to be able to say something that makes any kind
>of sense.

I honestly don't think you understand the law, and certainly my IP
attorney thinks your argument has severe flaws.

0
stork
3/8/2006 1:50:00 AM
On 2006-03-08, stork <stork@storkyak.com> wrote:
> Stork replied to:
>
>> This is the mistake you keep making. You think that you must
>> agree to the GPL as a condition of receiving the work. But you
>> don't
>
> I'm not saying that AT ALL.  I'm saying that the distribution of a
> computer work onto a public computer network is a performance or a
> display just like a rock concert or a TV show, and that, as the
> recipient of that show, you basically only have the right to view
> it.  Any additional rights you might want, you have to contact me
> to get, or, you can choose to gain additional rights by abiding by
> the GPL.
>
>> I'm sorry, I don't know what to say other than that you need to
>> understand the law in order to be able to say something that
>> makes any kind of sense.
>
> I honestly don't think you understand the law, and certainly my IP
> attorney thinks your argument has severe flaws.

    The simple fact is that the question of whether a program that
uses an API provided by a library is a derived work of that library,
has not been decided in a court of law. Until then, there are many
complicating factors, paramount among which is the fact that it is
possible to make an application binary that will successfully dynam-
ically link against a library without ever being in possession of
that library. There is also the fact that the parts of a library
that actually need to be incorporated into the application binary
are often uncopyrightable, the fact that it is undisputed that the
end user may link a non-gpl program against a gpl library [or vice
versa], the fact that the FSF has officially acknowledged a distinc-
tion between modifying a library and making a program that uses its
API, and many others.

    When you talk about "the spirit of the GPL" you are talking
about what you want the law to be, and indeed what you want the spi-
rit of the GPL itself to be - not necessarily what either actually
is.
0
Jordan
3/8/2006 2:02:31 AM
Stork replied to:

>Writing an application that uses a library is the normal and expected
>use of a library. It's pretty much the *only* thing you can do with a
>library.

All I'm doing is applying the iTunes model of DRM to software.  The
whole world is headed towards that.  Vista will have it built in and I
bet Linux will have it built in as well. Torvalds already said that he
can live with a DRM based kernal module - he just wants everyone to use
Linux.

So...a user on that machine can use your library, but, he can't grant a
third party the right to use it, only the copyright holder can.  So, I
have every right to use whatever library is on my machine, that's fair
use.  But I can't transfer it.  It's like how iTunes works.  I can
download a song and  play it as much as I want, but no one else is
allowed to touch that song by me.  If I want to transfer it, I go to
Apple, and say hey, can I transfer, and then they authorize it.

For an open source system, you add to that DRM a mechanism where you
supply your source to the copyright authority, they actually build your
application for distribution, and then, that proves to them that your
application is indeed open source, and then you can distribute that
application and library as needed, and everyone is free and open.

You know, I think I should actually patent that...

0
stork
3/8/2006 2:14:46 AM
On 2006-03-08, stork <stork@storkyak.com> wrote:
> Stork replied to:
>
>> Writing an application that uses a library is the normal and
>> expected use of a library. It's pretty much the *only* thing you
>> can do with a library.
>
> All I'm doing is applying the iTunes model of DRM to software.
> The whole world is headed towards that. Vista will have it built
> in and I bet Linux will have it built in as well. Torvalds already
> said that he can live with a DRM based kernal module - he just
> wants everyone to use Linux.

Quoting out of context is one thing - quoting with sufficient con-
text to support an idea that the context shows the quote does not
support is just moronic.

> So...a user on that machine can use your library, but, he can't
> grant a third party the right to use it, only the copyright holder
> can. 

That is NOT how the GPL works and it's 100% against its spirit.

> So, I have every right to use whatever library is on my machine,
> that's fair use. But I can't transfer it. It's like how iTunes
> works. I can download a song and play it as much as I want, but no
> one else is allowed to touch that song by me. If I want to trans-
> fer it, I go to Apple, and say hey, can I transfer, and then they
> authorize it.
>
> For an open source system, you add to that DRM a mechanism where
> you supply your source to the copyright authority, they actually
> build your application for distribution, and then, that proves to
> them that your application is indeed open source, and then you can
> distribute that application and library as needed, and everyone is
> free and open.
>
> You know, I think I should actually patent that...

That line, more than the rest of this entire thread, shows that you
really don't get it.
0
Jordan
3/8/2006 2:22:30 AM
"stork" <stork@storkyak.com> wrote in message 
news:1141782600.270841.201350@i40g2000cwc.googlegroups.com...

> I'm not saying that AT ALL.  I'm saying that the distribution of a
> computer work onto a public computer network is a performance or a
> display just like a rock concert or a TV show, and that, as the
> recipient of that show, you basically only have the right to view it.
> Any additional rights you might want, you have to contact me to get,
> or, you can choose to gain additional rights by abiding by the GPL.

    The right to view a show is a fancy way of saying the right to the 
ordinary, reasonable expected use of a work. What do you think is analogous 
to viewing a show if the work is a library supplied in source code form? 
What would be the normal expected use if compiling it and develop 
applications that link to it isn't?

    You have never answered this question, and I suspect you never will.

    DS 


0
David
3/8/2006 3:02:04 AM
Stork replied to:

>That line, more than the rest of this entire thread, shows that you
>really don't get it.

You are the one that cannot see that open source and free software are
diametrically opposed.

In order for software to be open, then you have to restrict the freedom
to use open software in non-free systems.  Otherwise everything will
wrapped up by proprietary systems and then the openness will be
gradually whittled away until the whole thing is closed and secret.
The GPL does try to prevent that, but you and David argue that the GPL
cannot legally do that, and even worse, reserve for yourselves the
right to directly subvert open source by using them in closed source,
proprietary products.

The guiding light of your argument is that that software, ultimately,
is not property, and I, along with -most- other software authors,
believe that it is.  Ultimately, you reserve for yourselves the right
to strip authors of their writes to their work, to be able to create
your own closed systems, and, yet you claim to be in favor of open
source?  You are right, I don't get how you can make that claim,
because, really, your position sounds a lot more like rationalized
theft of the works of the thousands of motivated people without even
having the decency to join them in their open source stand.

Computer source code, free or not, should be open source.  And, it is
more important to have an open source code system that remains the
property of the authors, then it is to have closed source libraries
that are essentially publicly domain.  In my view of the world,
everyone gets the benefit of truly shared knowledge, and your view of
the world promises little more than a Walmart savings on software.

You are right, I don't get it at all.

0
stork
3/10/2006 3:15:12 PM
>That line, more than the rest of this entire thread, shows that you
>really don't get it.

Linus Torvalds "Just to explain the fundamental issue: To me, the GPL
really boils down to "I give out code, I want you to do the same."  "
in a Forbes interview today.

Does he?

0
stork
3/10/2006 8:23:13 PM
On 2006-03-10, stork <tbandrow@mightyware.com> wrote:
>>That line, more than the rest of this entire thread, shows that you
>>really don't get it.
>
> Linus Torvalds "Just to explain the fundamental issue: To me, the GPL
> really boils down to "I give out code, I want you to do the same."  "
> in a Forbes interview today.
>
> Does he?

The line I was talking about, which you snipped, implied there is no
right AT ALL for someone to redistribute GPLed code. That is
_explicitly_ one of the goals of the GPL.
0
Jordan
3/10/2006 9:38:56 PM
"stork" <tbandrow@mightyware.com> wrote in message 
news:1142003712.185981.22590@p10g2000cwp.googlegroups.com...

> The guiding light of your argument is that that software, ultimately,
> is not property, and I, along with -most- other software authors,
> believe that it is.

    Nonsense. I am a strong advocate for intellectual property rights.

> Ultimately, you reserve for yourselves the right
> to strip authors of their writes to their work, to be able to create
> your own closed systems, and, yet you claim to be in favor of open
> source?

    The authors strip themselves of the right to control the *use* of their 
work when they decided to give away unlimited copies of that software for no 
fee and with no restrictive contract or agreement.

    You have total right of control over your car, until you give it away 
for free with no contractual restrictions. Then you have much less control. 
Does my arguing that mean that I'm arguing that your car is any less your 
property *before* you give it away?

> You are right, I don't get how you can make that claim,
> because, really, your position sounds a lot more like rationalized
> theft of the works of the thousands of motivated people without even
> having the decency to join them in their open source stand.

    You don't have to steal what is given to you for free without 
contractual restriction.

    I have taken an awful lot of time to explain to you what the law is, why 
it's that way, and what the consequences are. You don't like the end result, 
but you are unwilling to come up with any sort of coherent dispute of any of 
the steps on the way.

    Basically, it just comes down to you wishing that the universe was 
different than it is and that somehow what you think is ideal could work. 
But we've shown that if we changed the world to work the way you want it to, 
you get completely absurd results. You have still never addressed the 
argument that if we changed the law so that receiving a work for free didn't 
give you the right to the normal use of that work someone could drop copies 
of a poem from an airplane and sue everyone who read it.

    You want a world where the roads are paved with gumdrops and it only 
rains lollypops. And when people show that the way you think the world 
should work leads to absurdities, you complain that somehow you didn't mean 
that, you actually want things to work out right -- somehow.

    Sorry. It just doesn't work that way. If you want to keep control over 
something, don't give it away to anyone without some kind of agreement from 
them. It's really that simple.

    DS


0
David
3/10/2006 10:37:22 PM
Stork replied to:

> Nonsense. I am a strong advocate for intellectual property rights

Yeah, and the good old CCCP wanted prosperity for everyone too.  Anyone
with half a brain can cut through your half stoned Orwellian nonsense
just by reading your own glowing affirmation of intellectual property
rights: "You want a world where the roads are paved with gumdrops and
it only rains lollypops"  You can't say that you favor intellectual
property rights when you think they are absurd!

>I have taken an awful lot of time to explain to you what the law is, why
>it's that way, and what the consequences are. You don't like the end result,
>but you are unwilling to come up with any sort of coherent dispute of any of
>the steps on the way

I would seem more coherent to you if you were not consumed with an
almost pornographic lust to steal software.  Keeping intellectual
property around you is like keeping kids around Michael Jackson... you
just can't see straight enough about this stuff because you are just
overwhelmed with an unpurgable, grinding, syrup of a burned up coffee
soul condemned to software theft.

Once you accept your own failings, you will see that "absurd" outcome
of my view is that you are not allowed to steal someone's work!
Property is to be respected. Keep your hands out of that parisioner's
pants, padre!

>You have still never addressed the
>argument that if we changed the law so that receiving a work for free didn't
>give you the right to the normal use of that work someone could drop copies
>of a poem from an airplane and sue everyone who read it.

Normal use of library would be for you to write your own application
against it and using it.  Making copies of your application with the
library and selling it is not normal use.

Unfortunately, you use the deceptive trick of using a false parrallel.
You argue that the software is not so different from dropping poems out
of an airplane, which might be the case, but at the same time you
ignore the obvious parallel in broadcast television or radio, where
fair use allows you to record shows to watch later, but, not to
redistribute.  Then, you make the hilarious argument that you could get
sued for reading this airplane note, as if, the act of taking the paper
to a copier and printing 42,000 of them is the same as reading it.
Hello, I drop a note of software to you.  Boom, you go and use it.  Ok.
 Then you go and make 50,000 copies of it. Not Ok.  Get it?

>?orry. It just doesn't work that way. If you want to keep control over
>something, don't give it away to anyone without some kind of agreement from
>them. It's really that simple.

No, what's simple is: You are the one that wants to steal someone's
work and make money with it, despite the fact that said work comes
wrapped up with an appeal by the author to not do so.  If you don't
agree with the ideology, don't use it.

Keep your hands off other people's work, you frakking thief!  At least,
if you want to be a thief, admit it.    If you don't like to be
labelled as someone who steals, don't do it.  Listen OJ, we just don't
believe that Nicole tripped over the knife and cut her own tits off.

0
tbandrow
3/11/2006 6:22:40 PM
Stork replied to:

>It's undisputed that you can distribute your non-gpl application in
>source form and instruct your end-users to "link against it".

Yep.

>Clearly,
>then, "use" includes linking against it.

Nope.

>The only question here is
>whether that linked binary is a derivative work.

Well, there's a lot more than that.  Nice way to try and frame this
argument around legitimized software theft, but, you are still wrong.

0
tbandrow
3/11/2006 7:10:47 PM
Stork replied to:

>Did you read the Lexmark decision?

Lexmark is not a precendent for what we are talking about.

0
tbandrow
3/11/2006 7:11:33 PM
On 2006-03-11, tbandrow@storkyak.com <tbandrow@storkyak.com> wrote:
> Stork replied to:
>
>>It's undisputed that you can distribute your non-gpl application in
>>source form and instruct your end-users to "link against it".
>
> Yep.
>
>>Clearly, then, "use" includes linking against it.
>
> Nope.

Then where does the right for end-users to link a non-gpl application
against a gpl library come from? If it's illegal, it's always illegal.

My second statement follows DIRECTLY from the first statement.
0
Jordan
3/11/2006 9:12:39 PM
<tbandrow@storkyak.com> wrote in message 
news:1142101360.626132.186970@j52g2000cwj.googlegroups.com...

> Normal use of library would be for you to write your own application
> against it and using it.  Making copies of your application with the
> library and selling it is not normal use.

    Thank you, we now seem to agree that making an application that uses a 
library is normal use of that library. That's at least one step in the right 
direction.

    DS


0
David
3/11/2006 9:38:24 PM
<tbandrow@storkyak.com> wrote in message 
news:1142104293.221259.58030@j52g2000cwj.googlegroups.com...

> Stork replied to:

>>Did you read the Lexmark decision?

> Lexmark is not a precendent for what we are talking about.

    I asked a yes or no question. Either you did or you didn't, which is it?

    The Lexmark case is about you own every practical way to do a particular 
thing, you cease to be allowed to use copyright to do it. (Lexmark had a 
copyrighted Toner Loading Program in their print cartridges which Static 
Controls 'stole' to make compatible print cartridges. The court held that, 
among other things, even though the TLP would otherwise have been 
copyrightable, since it was the only practical way to make a cartridge work 
with certain Lexmark printers, copyright was not applicable.)

    That is exactly the issue here - can you use copyright to claim 
ownership of *every* way to make an application use a particular library? 
And the answer is -- of course not. Copyright can only allow you to own one 
particular way of doing something.

    DS


0
David
3/11/2006 9:40:08 PM
David Schwartz wrote:

> 
> <tbandrow@storkyak.com> wrote in message
> news:1142101360.626132.186970@j52g2000cwj.googlegroups.com...
> 
>> Normal use of library would be for you to write your own application
>> against it and using it.  Making copies of your application with the
>> library and selling it is not normal use.
> 
>     Thank you, we now seem to agree that making an application that uses a
> library is normal use of that library. That's at least one step in the
> right direction.
> 

"Dishonestly putting words in other peoples mouth" seems about the only
thing one would agree you are doing
-- 
Like being presumed a thief and a liar before using a product?
If so, use M$ XP

0
Peter
3/11/2006 9:41:47 PM
Jordan Abel wrote:

> On 2006-03-11, tbandrow@storkyak.com <tbandrow@storkyak.com> wrote:
>> Stork replied to:
>>
>>>It's undisputed that you can distribute your non-gpl application in
>>>source form and instruct your end-users to "link against it".
>>
>> Yep.
>>
>>>Clearly, then, "use" includes linking against it.
>>
>> Nope.
> 
> Then where does the right for end-users to link a non-gpl application
> against a gpl library come from? If it's illegal, it's always illegal.
> 
> My second statement follows DIRECTLY from the first statement.

Well, now explain what this end user "distributes"
-- 
Real programmers don't comment their code.  It was hard to write, 
it should be hard to understand.

0
Peter
3/11/2006 9:43:06 PM
Stork replied to:

> I asked a yes or no question. Either you did or you didn't, which is it?

I did read the decision, and, its not all about what you say it is.

It's a narrow ruling that says you are allowed to make after market
toner cartridges for a photocopier.  The toner loading program was but
a handful of bytes and a lookup table, and it was an embedded
application at that.  Lexmark's lawyers tried to stuff in some
protected phrase and copywrited it.  It was but one or two words and
the whole toner loading program, again, was but a handful of bytes.
It's not even clear that the toner loading computer hardware was
actually a turing machine.

Thus, there are so many differences between Lexmark's tiny embedded
toner program and the idea of using general purpose software on a
general purpose computer that this ruling is completely inapplicable.
Had the ruling had more meat to it, I'm sure that the Supreme Court
would have actually considered it, but they refused to do so.

So keep waving around Lexmark, but, it's only good for photocopiers and
tiny embedded devices, not, interfaces between general purpose systems
where there is much greater degree of freedom.

0
tbandrow
3/11/2006 10:24:44 PM
Stork replied to:

>The Lexmark case is about you own every practical way to do a particular
>thing

Your argument also falls apart on the "owning every practical way to do
a particular thing test".  It simply doesn't apply in the general
purpose software world.  I would use the same legal argument that
Dunkin Donuts used when it beat Amy Joy on the anti-trust rap.  Dunkin
Donuts went around and said, you can buy donuts from supermarkets,
convenience stores, and we don't have a monopoly on donuts at all.  Amy
Joy lost and went belly up.

In the software world, you would say, "I need to do XYZ and thus I have
the right to take Stork's library and redistribute it without any
contractual approval from him", and I would say "well, you can do XYZ
with A library, B library and C library, and in fact you could also do
it on Windows or Macintosh or any other system, and, because I'm only a
solo developer, it's even feasible for your firm to do it yourself,
thus, respecting my copyright in no way is a restraint of trade" -
which is what Lexmark was really all about.

Then, I could go on and argue further that the GPL that you abrogated
demanded that you create an open source work, which, is a really
reasonable request, does not impact your customers at all, and allows
everyone to benefit from your own expertise and knowledge.  On two key
points, your Lexmark argument goes up in smoke.

0
tbandrow
3/11/2006 10:47:48 PM
On 2006-03-11, tbandrow@storkyak.com <tbandrow@storkyak.com> wrote:
> Normal use of library would be for you to write your own application
> against it and using it.  Making copies of your application with the
> library and selling it is not normal use.

No. _that_ part is what's granted by the GPL [regarding the library. the
GPL has no standing at all to restrict or allow making copies of the
application]

And stop saying "Stork replied to:"!
0
Jordan
3/11/2006 11:32:30 PM
On 2006-03-11, tbandrow@storkyak.com <tbandrow@storkyak.com> wrote:
> Stork replied to:
>
>> I asked a yes or no question. Either you did or you didn't, which is it?
>
> I did read the decision, and, its not all about what you say it is.
>
> It's a narrow ruling that says you are allowed to make after market
> toner cartridges for a photocopier.  The toner loading program was but
> a handful of bytes and a lookup table,

That's about the size of what's actually included in a
dynamically-linked program from a library.
0
Jordan
3/11/2006 11:33:25 PM
"Peter K�hlmann" <peter.koehlmann@t-online.de> wrote in message 
news:duvg7t$fu3$00$1@news.t-online.com...

> David Schwartz wrote:

>> <tbandrow@storkyak.com> wrote in message
>> news:1142101360.626132.186970@j52g2000cwj.googlegroups.com...

>>> Normal use of library would be for you to write your own application
>>> against it and using it.  Making copies of your application with the
>>> library and selling it is not normal use.

>>     Thank you, we now seem to agree that making an application that uses 
>> a
>> library is normal use of that library. That's at least one step in the
>> right direction.

> "Dishonestly putting words in other peoples mouth" seems about the only
> thing one would agree you are doing

    He has never stated what his position is on what normal use of a library 
would be until that paragraph. I've asked him several times to do so.

    Tell me, what do you think normal use of a library would be?

    DS


0
David
3/12/2006 3:01:42 AM
Stork replied to:

>That's about the size of what's actually included in a
>dynamically-linked program from a library

I disagree on two counts:

Linker information is much much larger, particularly if you are using
C++ and have mangled method and function names and parameter lists.
Even in C, you still have to ask the library - that is use it - , at
run time, to provide you additional content about itself.  The dynamic
is much larger, the exchange more detailed, the communication more
rich.

A linker specification can be a creative work, simply because
interfaces to libraries have the hallmark of human design in their own
right.  We do a lot of work to separate a library's external interface
from its implementation - there's stuff deliberately hidden from
library users, names of public functions, classes and methods are
carefully thought out to make the use of the library as easy as
possible, and the mere existence of this analysis strongly suggests
that the interface to the library is a creative work, and to copy it
would be infringement in its own right.  The mechanical nature of the
linker is not an issue.  Photocopiers are mechanical too and you can't
go copying books willy nilly.

0
stork
3/13/2006 9:28:37 PM
On 2006-03-13, stork <tbandrow@mightyware.com> wrote:
> Stork replied to:
>
>>That's about the size of what's actually included in a
>>dynamically-linked program from a library
>
> I disagree on two counts:
>
> Linker information is much much larger, particularly if you are using
> C++ and have mangled method and function names and parameter lists.
> Even in C, you still have to ask the library - that is use it - , at
> run time, to provide you additional content about itself.  The dynamic
> is much larger, the exchange more detailed, the communication more
> rich.

It is possible, in principle, to make a program that is correctly
dynamically linked to a library without ever having the library.

> A linker specification can be a creative work, simply because
> interfaces to libraries have the hallmark of human design in their own
> right.

No they don't. Library interfaces are purely functional. If this were
not the case, we would have seen this argument in play when Compaq
cloned IBM's bios [i.e. writing another library to provide the same
interface]

> We do a lot of work to separate a library's external interface from
> its implementation - there's stuff deliberately hidden from library
> users, names of public functions, classes and methods are carefully
> thought out to make the use of the library as easy as possible, and
> the mere existence of this analysis strongly suggests that the
> interface to the library is a creative work, and to copy it would be
> infringement in its own right.  The mechanical nature of the linker is
> not an issue.  Photocopiers are mechanical too and you can't go
> copying books willy nilly.

If you _do_ copy a book, you don't then have a derived work of the book.
The mechanical nature of the linker means that the GPL's "viral" quality
cannot apply.
0
Jordan
3/13/2006 10:32:46 PM
On 2006-03-21, David Schwartz <davids@webmaster.com> wrote
>
> tbandrow@storkyak.com> wrote in messag
> news:1142941503.916688.204890@u72g2000cwu.googlegroups.com..
>
> Stork replied to Dave'
>
> Microsoft is wrong, they're generally accepted to be wrong, an
thei
> position has never been tested in court. Using software is a righ
> granted by 17 USC 117, and you don't automatically accept a contrac
b
> doing so
>
> I didn't say that. I don't think Microsoft is wrong and I don'
thin
> they're generally accepted to be wrong
>
Eh? You said they're wrong about THIS, given that microsoft claims yo
automatically accept by using [not by copying] software, and you hav
SPECIFICALLY claimed otherwise. Or did you lose track

>
> You have no right to us
> software without acceptance of an agreement, and use constitute
> acceptance of that agreement.  The more and more I think about it
th
> more and more this makes sense
>
> You don't need any special right to use something that you lawfull
> possess
>
Microsoft claims you do. Yet you still say you don't think they'r
wrong? stop undermining your own argument.  The issue we're talkin
about is Microsoft's claim that By using this software [not "b
clickin
'i accept'"] you accept the license agreement at <URL>

0
jmabel
3/21/2006 10:06:07 PM
Let me start by reminding everyone that this chain of arguments relates
*only* to EULAs, shrink-wrap agreements, click-throughs and the like.
It has nothing to do with copyright licenses like the GPL. The two
types of things have totally different strengths and weaknessess.

> Eh? You said they're wrong about THIS, given that microsoft claims you
> automatically accept by using [not by copying] software, and you have
> SPECIFICALLY claimed otherwise. Or did you lose track?

Microsoft does not, to my knowledge, rely on just a notice that by
using the software you accept an EULA. They would be very foolish to do
so because if the notice ever became separated from the software, they
would have no way to argue the EULA had been accepted. If you can find
an example of this, please do tell.

What Microsoft does, in my experience, is use a shotgun approach to
make sure that something sticks. For example, before you download a
product with an EULA, you are required to indicate your acceptance. An
EULA notice is placed on the box. The media may be sealed with a notice
that breaking the seal indicates acceptance of the EULA. There may be a
click-through during installation. The EULA will be installed along
with the product. You may need to indicate your acceptance the first
time you run the program.

Some of these techniques are almost certainly going to stand up in
court and some of them are much less likely to. A mere notice that by
using the software you agree to terms is almost certain *not* to stand
up unless somehow agreeing to the notice was made a necessary
precondition of using the software. Courts have already rejected such
notices when there was no oppurtunity to review the notice prior to
purchasing and opening the program and the software was non-returnable
after being opened.

Note that all of these techniques have weaknesses. For example, courts
have already observed that a click-through to download a program won't
work if I lend you my computer, you download something, and leave it on
my desktop. Similarly, a physical seal on media won't work if someone
breaks the seal, loses the media, someone else finds it, turns it in to
the police, nobody claims it, and the CD becomes the finder's property.

These types of agreements, because they are in fact agreements, will
always have 'innocent third party' weaknesses like this. This is
because the agreement seeks to take away something you would otherwise
have. Copyright licenses, like the GPL, do not have this weakness
because they only seek to give you something you otherwise would not
have.

Note that, as far as I know, none of this is controversial. I would be
surprised if the FSF (or Microsoft) disagreed with anything I said
here, except perhaps any implications as to whether things *should* be
the way they are. (Although if you see that in what I'm saying, it's
unintentional.)

DS

0
davids
3/22/2006 8:08:52 AM
Stork replied to DS:

>Let me start by reminding everyone that this chain of arguments relates
>*only* to EULAs, shrink-wrap agreements, click-throughs and the like.
>It has nothing to do with copyright licenses like the GPL.

You see, that's the trick.  If the court were to eventually find that
"use" was acceptance of a license, then, the GPL would actually be the
same as a normal EULA or a shrink wrap.  That's where my argument about
GPL enforcability hinged.  If you use it, you accept the contract that
represents the terms of its use.

I speculate that such an IP regime is where we will ultimately go,
because of the 'innocent third party' argument.  However, there will be
limits placed on what such a license could enable the software to do,
and how reasonable is it for a person to know the agreement, limits on
banning reverse engineering, publishing benchmarks, etc.

This isn't really a question for the courts, its really a problem for
the Congress to instruct the courts with.

0
stork
3/22/2006 5:06:20 PM
On 2006-03-22, stork <tbandrow@mightyware.com> wrote:
> Stork replied to DS:
>
>>Let me start by reminding everyone that this chain of arguments relates
>>*only* to EULAs, shrink-wrap agreements, click-throughs and the like.
>>It has nothing to do with copyright licenses like the GPL.
>
> You see, that's the trick.  If the court were to eventually find that
> "use" was acceptance of a license, then, the GPL would actually be the
> same as a normal EULA or a shrink wrap.

The FSF, though, claims that the GPL is a different KIND of license,
that explicitly isn't necessary for "use".
0
Jordan
3/22/2006 5:08:27 PM
Stork replied to:

>The FSF, though, claims that the GPL is a different KIND of license,
>that explicitly isn't necessary for "use".

Well, that's their spin anyway, but doesn't the GPL have something
like, "well, this isn't a EULA or for use, but, um, it is the only
rights you have because we believe copyright law excludes everything
else".

Isn't that, in a nutshell, you and Dave's sticking point on the GPL
anyway?  That, in reality, copyright law says some things about first
use that the GPL cannot claim are excluded.  Under those circumstances,
the only way that the FSF could argue the GPL would be if in fact, it
was a contract after all, wherein, the user is actually waiving some
rights in order to use the software.

Ultimately, the FSF could wind up in the bizarre situation of having to
legally argue that that use of software is not a right unless
explicitly granted by the author, in order for GPL licenses to hold.  I
do think it is a real possibility that the only way a purely public
license such as the GPL could specifically work, without strong
property rights and a bashing of first use, is that the Congress is
going to have to carve out a specific language for public licenses to
work among non-profits or researchers or both, and also create specific
waivers to allow patent infringement in the open source world.

Thoughts?

0
tbandrow
3/23/2006 3:47:52 AM
> You see, that's the trick.  If the court were to eventually find that
> "use" was acceptance of a license, then, the GPL would actually be the
> same as a normal EULA or a shrink wrap.  That's where my argument about
> GPL enforcability hinged.  If you use it, you accept the contract that
> represents the terms of its use.

How could a court validating an agreement's term that mere use
indicated acceptance have any bearing on an agreement that has no such
term?! You are way out there. You're either a troll or you're beyond
the reach of reason.

In any event, I already explained in a previous post why courts would
not hold this to be the case in general, but only in very specific
instances. It would prove much less reliable protection than the
mechanism the GPL uses, which is ironclad.

Consider:

Your claim: By doing (thing that you already have the legal right to
do) you agree to X.
The GPL: I grant you the right to do (thing that copyright law
expressly prohibits you from doing) if and only if you agree to X.

See the difference?

DS

0
davids
3/24/2006 8:58:57 PM
Stork replied to DS:

>See the difference?

Yes, but the loophole in the GPL is "that which copyright law expressly
prohibits you from doing".  In our case, I think we've agreed that
there are a lot of gray areas as to what "copyright law expressly
prohibits you from doing" exactly means.  There's certainly little
settled law on this in terms of software, and as such, it would be
preferable for the Congress, which is elected to describe it anyway,
and instruct the courts properly, rather than leave it up to an
unelected panel of judges.

0
tbandrow
3/25/2006 7:29:57 PM
<tbandrow@storkyak.com> wrote in message 
news:1143314997.446741.97180@u72g2000cwu.googlegroups.com...

> Yes, but the loophole in the GPL is "that which copyright law expressly
> prohibits you from doing".

    Exactly. That is the huge difference between the power and scope of the 
GPL and that of an EULA. The GPL is ironclad in the sense that you cannot 
wrangle out of it by failing to agree to it (there is no 'innocent third 
party' defense). However, the GPL can only be invoked if you do something 
that you had no other way to do, and that means it must be outside the scope 
of fair use, scenes a faire, first sale, lack of protectable expression, and 
so on.

> In our case, I think we've agreed that
> there are a lot of gray areas as to what "copyright law expressly
> prohibits you from doing" exactly means.  There's certainly little
> settled law on this in terms of software, and as such, it would be
> preferable for the Congress, which is elected to describe it anyway,
> and instruct the courts properly, rather than leave it up to an
> unelected panel of judges.

    That may or may not make a difference. Either way, it's not going to be 
possible to describe the full scope of possible situations before those 
situations occur. IMO, courts have done a remarkably good job in extending 
conventional intellectual property law into new areas. The biggest screwups, 
IMO, have been when the legislature has tried to create new law specific to 
new technologies, such as the DMCA, and inadvertently messed up delicate 
balances.

    One thing that definitely won't work is extending copyrights to have the 
power of patents (protect functional elements rather than just creative 
expression) but without any of the balancing limitations (public disclosure, 
more limited time, review requirements, and so on).

    DS


0
David
3/26/2006 12:18:59 AM
tbandrow@storkyak.com wrote:

> Stork replied to:
>
> >The FSF, though, claims that the GPL is a different KIND of license,
> >that explicitly isn't necessary for "use".

> Well, that's their spin anyway, but doesn't the GPL have something
> like, "well, this isn't a EULA or for use, but, um, it is the only
> rights you have because we believe copyright law excludes everything
> else".

The GPL *is* different because it doesn't try to take any rights away
from you and you don't have to agree to it. It's in a totally different
category from an EULA.

> Isn't that, in a nutshell, you and Dave's sticking point on the GPL
> anyway?  That, in reality, copyright law says some things about first
> use that the GPL cannot claim are excluded.  Under those circumstances,
> the only way that the FSF could argue the GPL would be if in fact, it
> was a contract after all, wherein, the user is actually waiving some
> rights in order to use the software.

There is almost no way they could argue that because the GPL just
doesn't work that way. It never says "by doing X, you are bound to
comply with the GPL", it just says, "if you want permission to do X
from me, you must do Y". This difference is *huge* (because with the
latter, if you get permission some other way, you are not bound).

> Ultimately, the FSF could wind up in the bizarre situation of having to
> legally argue that that use of software is not a right unless
> explicitly granted by the author, in order for GPL licenses to hold.

Except that would fail the giggle test. That would mean I could drop a
million copies of a poem from an airplane and sue everyone who read it
for royalties.

> I
> do think it is a real possibility that the only way a purely public
> license such as the GPL could specifically work, without strong
> property rights and a bashing of first use, is that the Congress is
> going to have to carve out a specific language for public licenses to
> work among non-profits or researchers or both, and also create specific
> waivers to allow patent infringement in the open source world.

The thing is, the licenses *do* work. They just don't give authors as
much control over they might want. But this is appropriate. You should
not be allowed to give something away for free and still retain such a
high level of control over it unless people agree to give it to you.

The same people arguing for such ludicrously strong copyright are
arguing *against* software patents.

DS

0
davids
3/26/2006 5:37:51 AM
Reply: