f



PledgeWare: Dolphin 7 and 64 bit Poll Results

Hi folks,

It's been a good few weeks since I put out the call for pledges to 
support upgrading the Dolphin VM to 64-bit and to release a new version 
of the IDE as Dolphin 7. So, I thought you might like to know the results.

The response has been very good and it certainly shows that a fair 
number of you out there are using Dolphin for some quite sophisticated 
"heavy lifting" applications. I was surprised by the number of people 
willing to pledge $1000 or over for the 64-bit version. However, the aim 
of the poll wasn't to find the maximum price but rather a price at which 
we could generate sufficient revenue to support the work needed.

 From the results pooled so far it seems that an upgrade fee of slightly 
under $500 (say, $450) will be enough to justify the likely development 
effort.

The aim now is to begin work on the new VM/IDE and, at some point in the 
future, I will write to all of you who pledged $500 or more to ask you 
to reconfirm a pledge for $450 in a slightly more formal fashion (just 
to make sure we still have sufficient interest). When we have reached 
the stage where a beta version seems plausible I will invite committed 
pledgers (only) into the programme.

So, can I just say thank you to all the people who responded to the poll 
and for the enthusiasm and continued support of the Dolphin community in 
general.

Best regards,

Andy Bower
Object Arts Ltd
0
bower (2)
12/18/2009 12:02:45 PM
comp.lang.smalltalk.dolphin 3769 articles. 0 followers. Post Follow

37 Replies
627 Views

Similar Articles

[PageSpeed] 30

Andy,

Very cool news indeed !!!

Regards,
Bruno
0
Bruno
12/18/2009 12:17:00 PM
It seems that the Dolphin community is pretty alive !!!

Regards,
Bruno
0
Bruno
12/18/2009 12:23:21 PM
Andy,
Really VERY GOOD NEWS!!!

Go Dolphin!

Regards,
Guillermo Sapaya

On 18 dic, 09:02, Andy Bower <bo...@SnipThisObject-arts.com> wrote:
> Hi folks,
>
> It's been a good few weeks since I put out the call for pledges to
> support upgrading the Dolphin VM to 64-bit and to release a new version
> of the IDE as Dolphin 7. So, I thought you might like to know the results=
..
>
> The response has been very good and it certainly shows that a fair
> number of you out there are using Dolphin for some quite sophisticated
> "heavy lifting" applications. I was surprised by the number of people
> willing to pledge $1000 or over for the 64-bit version. However, the aim
> of the poll wasn't to find the maximum price but rather a price at which
> we could generate sufficient revenue to support the work needed.
>
> =A0From the results pooled so far it seems that an upgrade fee of slightl=
y
> under $500 (say, $450) will be enough to justify the likely development
> effort.
>
> The aim now is to begin work on the new VM/IDE and, at some point in the
> future, I will write to all of you who pledged $500 or more to ask you
> to reconfirm a pledge for $450 in a slightly more formal fashion (just
> to make sure we still have sufficient interest). When we have reached
> the stage where a beta version seems plausible I will invite committed
> pledgers (only) into the programme.
>
> So, can I just say thank you to all the people who responded to the poll
> and for the enthusiasm and continued support of the Dolphin community in
> general.
>
> Best regards,
>
> Andy Bower
> Object Arts Ltd

0
Guiye
12/18/2009 1:34:01 PM
Hi!


Very exciting news! Thank you and happy Christmas and New Year for
all.

Cheers
  Sebastian Calvo


On 18 dic, 09:23, Bruno Buzzi Brasesco <bruno.brase...@gmail.com>
wrote:
> It seems that the Dolphin community is pretty alive !!!
>
> Regards,
> Bruno

0
GallegO
12/18/2009 1:36:37 PM
Andy,

what a great XMAS gift!!! Thank you very much for you efforts.

Best Regards,

Udo


Andy Bower schrieb:
> Hi folks,
> 
> It's been a good few weeks since I put out the call for pledges to 
> support upgrading the Dolphin VM to 64-bit and to release a new version 
> of the IDE as Dolphin 7. So, I thought you might like to know the results.
> 
> The response has been very good and it certainly shows that a fair 
> number of you out there are using Dolphin for some quite sophisticated 
> "heavy lifting" applications. I was surprised by the number of people 
> willing to pledge $1000 or over for the 64-bit version. However, the aim 
> of the poll wasn't to find the maximum price but rather a price at which 
> we could generate sufficient revenue to support the work needed.
> 
>  From the results pooled so far it seems that an upgrade fee of slightly 
> under $500 (say, $450) will be enough to justify the likely development 
> effort.
> 
> The aim now is to begin work on the new VM/IDE and, at some point in the 
> future, I will write to all of you who pledged $500 or more to ask you 
> to reconfirm a pledge for $450 in a slightly more formal fashion (just 
> to make sure we still have sufficient interest). When we have reached 
> the stage where a beta version seems plausible I will invite committed 
> pledgers (only) into the programme.
> 
> So, can I just say thank you to all the people who responded to the poll 
> and for the enthusiasm and continued support of the Dolphin community in 
> general.
> 
> Best regards,
> 
> Andy Bower
> Object Arts Ltd
0
Udo
12/18/2009 1:45:12 PM
This is one of the top Smalltalk news in 2009.

Do you already have an schedule for a beta release or something like
that?

Here at InfOil we'll drink a couple of beers celebrating this :D

Regards!





0
Esteban
12/18/2009 3:11:16 PM
Andy Bower schrieb:

> When we have reached the stage where a beta version seems plausible 
> I will invite committed pledgers (only) into the programme.

Today is day 1 of "priviledged community processes" :)

Have fun, Guido Stepken
0
Guido
12/18/2009 4:35:12 PM
Andy Bower wrote:

> The aim now is to begin work on the new VM/IDE and, at some point in the 
> future, I will write to all of you who pledged $500 or more to ask you 
> to reconfirm a pledge for $450 in a slightly more formal fashion (just 
> to make sure we still have sufficient interest).  
   I think Dolphin under your diligent effort will proceed to a new era
on Windows 7 and makes a shining future, the way to go forever and
independently. please count me in as an interested hobbyist and a
pledged fidelity member.
0
Eno
12/19/2009 12:37:54 AM
Awesome, thank you!

Travis
0
Travis
12/19/2009 6:19:58 AM
Andy,

verry good news! Thanks a lot!

Cheers,
Klaus
0
klausk
12/19/2009 2:13:23 PM
I'm in for the $450 upgrade to Dolphin 7.

I recently bought Dolphin 6.

I really like it.

It's worth it over the community edition.  I really like the Idea
Browser and the Systerm Browser.

I really like the single file .exe generation.  That is really extra
super great.  It's critical to me.

I have found historicaloptionsdata.com and so I want to be able to
move a lot of data fast.

Please keep Omnibase in Dolphin 7.  Please add the 3D animation
stuff.  If you could make an interface to a large free SQL database
that would be cool.  Could you include the Java interface?

Can you do tail call optimization?

Please make it so that the Dolphin 6 code doesn't get hopelessly
broken by 7.

Please make zero input BlockClosures really fast if possible.
AbstractGenerators use them.

I have gotten all my Generator stuff to file into Dolphin 6.  I would
like to publish it on sourceforge.net under the MIT license for
Dolphin.  But right now it is under GPL.  I have to change it in the
comments.  Ok F it.  I will publish it now under GPL and later under
MIT.  You can combine generators and FiniteStateMachines and
pushDownAutomata parser compilers.  I am using the pushDownAutomata
Generator stuff to convert a random number sequence into a tree
structure for market analysis.

Please look at picoLARC on sourceforge.net for ideas about how method
dispatch can be done via an Array lookup for Object level methods.
Also unique methods in leaf Classes can be found without any lookup.
The main idea is that polymorphic methods are grouped together into
genericFunctions so if there is only one method in the genericFunction
then there is no lookup needed.  If one of the versions of a method M
is in the Object Class then you can use an Array lookup to find any of
the different versions of M at runtime.  But in order to do this you
have to be able to put a Method into a variable at runtime like in
Scheme.  Perhaps Smalltalk just can't do that. So you look in the
Method variable and get the genericFunction and then get the Class of
the first input which is the receiver.  The genericFunction has all
the different versions of that Method in it.  You use the Class as the
key to look in the genericFunction to find the FunctionEval that
implements the Method version.  Instead of using the method selector
as the key to look into a Class to find a Method you use the Class as
the key to look into the genericFunction to find the Method.  If there
is only one monomorphic version of a Method in the system then no
lookup is required.  This only happens in leaf Classes.  If a
polymorphic Method has a version in the Object Class then the
genericFunction has all the Classes in it so an Array lookup will do
it.  What?  That doesn't seem right.  Oh well I can't remember the
details.  It was something like that. Yes.  No it does make sense.
Because if a Method version is in the Object Class then all instances
respond to that message so all Classes are in the genericFunction.  So
an Array lookup is what is needed.  Each Class gets an id number which
is the key to the Array lookup.  So if you compile a link to the
genericFunction into every Smalltalk message send, then you should be
able to do this also in Smalltalk.  Like it is being done in
picoLARC.  And then you can tell programmers that if they stick a
version of a method into the Object class or make it a unique leaf
Class method then the method dispatch will be speeded up.  If they
have a leaf Class with nothing but unique private methods in it then
no method lookups will be needed by those methods.  And that Class can
be faster internally.  A method that was unique in a leaf Class could
be simple recursion tail call optimized even in Smalltalk.  You could
signal an optimization warning if the programmer makes another version
in some other Class.  Also these genericFunctions are all hidden.  So
you don't have to add any special syntax or new concepts to the
Smalltalk language in order to make use of these ideas.  There are no
GenericFunction browsers in picoLARC.  And no way to define
GenericFunctionEvals programatically.  It is all done automatically.

And bla bla bla bla bla.

It means that I am glad Dolphin is moving forwards.  But I don't care
if it dies because I will still keep on using it even if it does.
Because enough source code is included that it can't really ever die.

And I want in on the $450 upgrade deal.

What do I get?

0
picoVerse
12/21/2009 8:13:20 AM
I greatly applaud the efforts to keep
Dolphin alive.

The question remains. Why isn't Dolphin
the big success it deserves to be? Why
is it still uncertain how long Dolphin
will be around?


It seems to me by now that it is impossible
for any small company to create a successfull
ongoing development platform without making
at least the main core of it open source.
C# could not survive without Mono.
Even Java had to go open source.

I hope you can prove me wrong.

Programmers and investors know a "community
edition" is not the same thing as "open source".
Open source means never having to worry about
whether the company supporting it dies OR
decides to kill the product, or is bought.
Like what happened with Smalltalk-V.


Why does it take so long to come up with new
versions of Dolphin? I think it is because
there are too few programmers working on that.

My guess is there are probably less than 10
programmers working on the next Dolphin version.
Am I correct?  If Dolphin was open source, there
could be hundreds of programmers working on it.

I just found an article on the web by a long time
Sun employee Jeremy Allison,  which opened my
eyes a bit in this matter:

  http://blogs.zdnet.com/BTL/?p=31418

He's saying although Sun OS was/is the best,
Linux ran past it because Linus let the code
run free. Sun tried jealously keep all the code
to itself, too long.

Oracle is thriving and Red Hat is thriving.
Sun Microsystems is no more.

Anyway, best success, keep Dolphin swimming

-Panu Viljamaa

0
Panu
3/10/2010 7:19:14 AM
Panu <panu@nospam.com> writes:

> I greatly applaud the efforts to keep
> Dolphin alive.
>
> The question remains. Why isn't Dolphin
> the big success it deserves to be? Why
> is it still uncertain how long Dolphin
> will be around?
>
>
> It seems to me by now that it is impossible
> for any small company to create a successfull
> ongoing development platform without making
> at least the main core of it open source.
> C# could not survive without Mono.
This is not true. If you want to do "easier" windows programming
you will use C# or VB.NET you do not need mono for it. You won't tell us
that Mono with it's market share from below 10% is the "reason" of
success.




> Even Java had to go open source.
>
> I hope you can prove me wrong.
Sure easy. Is .NET from M$ open source? 


>
>
> Why does it take so long to come up with new
> versions of Dolphin? I think it is because
> there are too few programmers working on that.
It's not the number alone, but also how much time the programmmers
can spend on it. If people would be Dolphin Smalltalk  like hell, this
would not be a problem. ObjectArts could pay other Smalltalk developers
for helping improving Dolphin.
>
> My guess is there are probably less than 10
> programmers working on the next Dolphin version.
> Am I correct?  If Dolphin was open source, there
> could be hundreds of programmers working on it.
You do not believe that seriously. I suggest you try to figure out how
many "core" programmers do work on certain "open source software". I
suggest you try to figure that out for gcc. 

>
> He's saying although Sun OS was/is the best,
> Linux ran past it because Linus let the code
> run free. Sun tried jealously keep all the code
> to itself, too long.
Counter argument above, You claim Java is a success because it's open
source, however the publisher of it has been bought from Oracle. So in
the end the Open Source Java has driven Sun into the arms or Oracle.
>
> Oracle is thriving
Orcale surely is an example of open source software. I'm sure they are
very willing giving you the sources of their Database

> and Red Hat is thriving.
Oh, and that's not because of services but because of selling licences
of RedHat Linux?

> Sun Microsystems is no more.
Ah well Java is open source, Sun Microsystems has gone. Maybe the Open
source Java was a bit too expensive?

Regards
Friedrich



-- 
Please remove just-for-news- to reply via e-mail.
0
Friedrich
3/10/2010 7:42:22 AM
Panu wrote:
> I greatly applaud the efforts to keep
> Dolphin alive.
> 
> The question remains. Why isn't Dolphin
> the big success it deserves to be? Why
> is it still uncertain how long Dolphin
> will be around?

The market is not always as visionary as one might like.  I continue to 
think that OA would do well to add support for a cross-platform GUI 
framework and to get a sibling Smalltalk with a basic GUI running on 
Linux and Mac OS.  Andy has expressed various concerns, including that 
it would be difficult/impossible to sell to a Linux audience given that 
they expect everything free.  *I* would happily fork over for a portable 
OA Smalltalk, but I can guarantee no particular market size.



> It seems to me by now that it is impossible
> for any small company to create a successfull
> ongoing development platform without making
> at least the main core of it open source.
> C# could not survive without Mono.

With respect, rubbish!!!  C# is new and "everybody is using it."  For 
many people, there appear to be very few other points to consider in 
choosing a platform.  Mono is an also-ran that will forever be reverse 
engineering a moving target.  MS will exploit it as far as it allows 
them to claim to have cross-platform support, but you will see nothing 
concrete from them in terms of helping to make things work on 
non-Windows platforms.

In my mind, C# was MS' response to being slapped around when they tried 
to add features to Java to create vendor and platform lock; when they 
were told to play hands-off of Java, they made C# (aka Java flat minor).



> Even Java had to go open source.

I'm not sure Java deserves an "even" in this context.  Nothing could 
live up to the hype it received in its glossy-ad peak, and there was a 
guarantee of an eventual backlash when people learned the truth. 
Ancient history.



> I hope you can prove me wrong.

I share your desire to see Dolphin live long and prosper.


> Programmers and investors know a "community
> edition" is not the same thing as "open source".
> Open source means never having to worry about
> whether the company supporting it dies OR
> decides to kill the product, or is bought.
> Like what happened with Smalltalk-V.

Never is a long time.  Any non-trivial product has a bus factor, whether 
  one realizes it or not.


> Why does it take so long to come up with new
> versions of Dolphin? I think it is because
> there are too few programmers working on that.

There is not enough of a market to drive it; that is sad but true.



> My guess is there are probably less than 10
> programmers working on the next Dolphin version.
> Am I correct?  If Dolphin was open source, there
> could be hundreds of programmers working on it.

Open source makes financial sense "only" if the product in question is a 
means to an end.  In other words, OA would have to want Dolphin for 
other purposes, in which case they would gain an improved and expanded 
Dolphin to use to make money doing things with Dolphin.  With their 
focus on creating Dolphin as an end product, opening the sources would 
offer no gain and would close their revenue stream, which is a form of 
fiscal suicide.




> I just found an article on the web by a long time
> Sun employee Jeremy Allison,  which opened my
> eyes a bit in this matter:
> 
>  http://blogs.zdnet.com/BTL/?p=31418
> 
> He's saying although Sun OS was/is the best,
> Linux ran past it because Linus let the code
> run free. Sun tried jealously keep all the code
> to itself, too long.

Linux is doing well because it has gotten to be really quite good, 
working on Damn Good(TM) status.  Start with bare hardware and time 
yourself getting Windows installed, patched and ready to run vs. 
installing Linux from a Live CD of your choice and updating it; there 
was a time when Linux was indeed difficult to use, but that has changed. 
  I'm typing this on a Gnome based system that I very much enjoy using. 
  Some of the heavy Unix systems are apparently taking a beating from 
the competition, but that makes sense as the first place that a 
unix-like open system would score its first points.



> Oracle is thriving and Red Hat is thriving.
> Sun Microsystems is no more.

I got an ear-full just today from an MS junkie (they can do no wrong in 
his view) about how Red Hat is toast and Oracle will be next.  I don't 
pretend to know.

Bill
0
Wilhelm
3/11/2010 1:57:37 AM
Friedrich Dominicus wrote:
> Panu <panu@nospam.com> writes: 
>> C# could not survive without Mono.
> This is not true. If you want to do "easier" windows programming
> you will use C# or VB.NET you do not need mono for it. You won't tell us
> that Mono with it's market share from below 10% is the "reason" of
> success.

I think you missed my point which is Microsoft had to
allow an open source implementation of .NET to exist
in order to give C# a better chance at competing with
Java. There could be a .NET without Mono, but it
wouldn't be as popular.


>> Even Java had to go open source.
>> I hope you can prove me wrong.
> Sure easy. Is .NET from M$ open source? 

No, but .NET Mono is. MS seems to have
made a conscious decision to allow an
open-source version of .NET to exist.
Why? I think because they thought it
would improve .NET's chances of survival.


>>  If Dolphin was open source, there
>> could be hundreds of programmers working on it.
> You do not believe that seriously. 

I do believe there could be hundreds of
programmers working to improve Dolphin and
its libraries.


> I suggest you try to figure out how
> many "core" programmers do work on certain "open source software". I
> suggest you try to figure that out for gcc. 

If you have that figured out why not
share the information, with us.


> 
>> He's saying although Sun OS was/is the best,
>> Linux ran past it because Linus let the code
>> run free. Sun tried jealously keep all the code
>> to itself, too long.
> Counter argument above, You claim Java is a success because it's open
> source, however the publisher of it has been bought from Oracle. So in
> the end the Open Source Java has driven Sun into the arms or Oracle.


If you read the article I linked to, the point
is more about Solaris and the rest of Sun software
being closed-sourced too long. The point is about
Linux winning market-share over Solaris which was
a better product, but proprietary.

( http://blogs.zdnet.com/BTL/?p=31418 )

Java was not open source from the beginning, but
it always was "free". That's why it is in a strong
position today.

Now that Solaris IS open source, I believe
it has a chance at survival.


>> Oracle is thriving
> Orcale surely is an example of open source software. I'm sure they are
> very willing giving you the sources of their Database
> 


I'm sure there's good business to be made in
proprietary software.  My point in particular
was that open source is probably the best way
to create  programming "platforms". Platforms
need to be open source, to gain maximum adoption.

One reason is that unlike business applications
platforms are products for programmers. Programmers
will prefer to create and modify their own platforms,
to best suit their needs - because they can. The
most prevalent platforms will be open source because:

1) They are cost-effective

2) Programmers can make their own modifications to them.
     (But so can you with Dolphin, so ...)

3) Programmers can share their modifications with others.
     (Not sure if Dolphin's licensing allows that)

4) When there's lots and lots of programmers using and
    improving a specific platform, it makes sense
    for you to use it too. Snowball effect kicks in.

5) The platform will survive the death of any number
    of companies supporting it. Not so with proprietary
    "single source" platforms.

The opposite of "open source" is "single source".

The best way for small companies in the platforms-
business to succeed is to open-source their core
offerings I believe. Eclipse is thriving. Dolphin
not so much - even though it is a better IDE
at the moment.


The world of programming platforms is all
about natural selection i.e. survival of
the "fittest". Open source is more likely
to survive because it can more easily
adapt, to "fit" to changing circumstances
- because anybody can modify it.



>> and Red Hat is thriving.
> Oh, and that's not because of services but because of selling licences
> of RedHat Linux?
> 

You are probably right about the services,
but the reason they are thriving is that Linux
is thriving, and Linux is thriving because
it is and always was open source.

Regards
Panu Viljamaa

0
Panu
3/11/2010 2:47:49 AM
Wilhelm K. Schwab wrote:

> 
> Open source makes financial sense "only" if the product in question is a 
> means to an end.  In other words, OA would have to want Dolphin for 
> other purposes, in which case they would gain an improved and expanded 
> Dolphin to use to make money doing things with Dolphin.  With their 
> focus on creating Dolphin as an end product, opening the sources would 
> offer no gain and would close their revenue stream, which is a form of 
> fiscal suicide.


Those are good points. I don't pretend to
know their business better than they do.

I just know personally I'm very reluctant
to invest time writing a program for
a platform that can easily become a dead
end, because it is "single source".

It seems the industry as a whole feels
the same way and therefore Smalltalk is a
niche player so far.


IF Dolphin was open-source I think OA could
make more money out of support contracts
with big companies, who would be willing to
choose Dolphin because they knew it would
survive anything that happens to OA.


Consider the case currently:  If I am IBM
I don't want to bet my future on Dolphin
because Oracle might any day BUY OA. Likewise
Oracle would not want to become dependent
on OA's products because IBM might buy it
any day.

If Dolphin was open source, BOTH Oracle
and IBM could trust it, and OA would get
many more paying support customers.

OA might further use their superior knowledge
of Dolphin to build superior commercial
applications, or programming tools on top
of the free open-source platform.


OA would have more time and resources to
create such applications if the community
would take part in the work of creating
the superior platform.


-Panu Viljamaa
0
Panu
3/11/2010 3:58:30 AM
Panu wrote:
> Friedrich Dominicus wrote:
>> Panu <panu@nospam.com> writes:
>>> C# could not survive without Mono.
>> This is not true. If you want to do "easier" windows programming
>> you will use C# or VB.NET you do not need mono for it. You won't tell us
>> that Mono with it's market share from below 10% is the "reason" of
>> success.
> 
> I think you missed my point which is Microsoft had to
> allow an open source implementation of .NET to exist
> in order to give C# a better chance at competing with
> Java. There could be a .NET without Mono, but it
> wouldn't be as popular.

Again, "everybody's using it..." is all the hype-compliant developer 
needs to justify a decision.  People drinking the coolaid do not care 
one bit about an open alternative.  Opening another Smalltalk won't get 
them to change to it.


>>> Even Java had to go open source.
>>> I hope you can prove me wrong.
>> Sure easy. Is .NET from M$ open source? 
> 
> No, but .NET Mono is. MS seems to have
> made a conscious decision to allow an
> open-source version of .NET to exist.
> Why? I think because they thought it
> would improve .NET's chances of survival.

I doubt it: it allows them to make lip-service to being cross-platform. 
  Beyond that they don't give a flying whatever.



> The opposite of "open source" is "single source".

Do you program with logic like that?  Your assertion will be news to 
Cincom, Object Connect and Instantiations, in addition to Object Arts.

Bill


0
Wilhelm
3/11/2010 5:15:32 AM
Panu wrote:

> It seems the industry as a whole feels
> the same way and therefore Smalltalk is a
> niche player so far.

The key word is "feel."  I have known many people who dismiss Smalltalk, 
  none of whom appeared to have any dynamic language experience.



> Consider the case currently:  If I am IBM
> I don't want to bet my future on Dolphin
> because Oracle might any day BUY OA. Likewise
> Oracle would not want to become dependent
> on OA's products because IBM might buy it
> any day.
> 
> If Dolphin was open source, BOTH Oracle
> and IBM could trust it, and OA would get
> many more paying support customers.

Did not IBM dump VA Smalltalk off to Instantiations?  You are dreaming.



> OA might further use their superior knowledge
> of Dolphin to build superior commercial
> applications, or programming tools on top
> of the free open-source platform.

You are not the first to suggest that model, but OA was interested in 
being a tool vendor rather than a consulting firm.

Bill
0
Wilhelm
3/11/2010 5:23:29 AM
"Wilhelm K. Schwab" <bschwab@anest.ufl.edu> writes:

> Panu wrote:
>> I greatly applaud the efforts to keep
>> Dolphin alive.
>>
>> The question remains. Why isn't Dolphin
>> the big success it deserves to be? Why
>> is it still uncertain how long Dolphin
>> will be around?
>
> The market is not always as visionary as one might like.  I continue
> to think that OA would do well to add support for a cross-platform GUI
> framework and to get a sibling Smalltalk with a basic GUI running on
> Linux and Mac OS.  Andy has expressed various concerns, including that
> it would be difficult/impossible to sell to a Linux audience given
> that they expect everything free.

Definitly, we once had Qide a port of our C-IDE it was quite nice to
have some 10 years ago. But the first question you here is just "isn't
it open source".... 


>  *I* would happily fork over for a
> portable OA Smalltalk, but I can guarantee no particular market size.
I can guarantee that you get some downloads and then the beforementioned
question.

>
>
>
>> It seems to me by now that it is impossible
>> for any small company to create a successfull
>> ongoing development platform without making
>> at least the main core of it open source.
>> C# could not survive without Mono.
>
> With respect, rubbish!!!  C# is new and "everybody is using it."  For
> many people, there appear to be very few other points to consider in
> choosing a platform.  Mono is an also-ran that will forever be reverse
> engineering a moving target.  MS will exploit it as far as it allows
> them to claim to have cross-platform support, but you will see nothing
> concrete from them in terms of helping to make things work on
> non-Windows platforms.
It seems MS has given some kind of support. But I do not remember where
I read about it.
>
> In my mind, C# was MS' response to being slapped around when they
> tried to add features to Java to create vendor and platform lock; when
> they were told to play hands-off of Java, they made C# (aka Java flat
> minor).
Well for me it seems that was a wise decisoin for MS. So they can bind
the people to windows, and that's the only thing they are intersted in. 

>
>
>
>> Even Java had to go open source.
>
> I'm not sure Java deserves an "even" in this context.  Nothing could
> live up to the hype it received in its glossy-ad peak, and there was a
> guarantee of an eventual backlash when people learned the
> truth. Ancient history.
Well the even just means they did not sell enough of it and so they
tried to hit back M$, with going "open-source" But again it would be
very intersting to know how many programmers do work on Java by Sun now
Oracle...

>
>
>
>> I hope you can prove me wrong.
>
> I share your desire to see Dolphin live long and prosper.
I bought it and it's very well integrated into the Windows world. I
tried connecting to Access and automate things in Excel with it and that
has worked. It's a pitty for me that it's Windows only, but if I'd be
ObjectArts I would not spend a millisecond porting it to Linux (maybe
peoplee on Macs are willing to pay for it, I don't know). You never ever
will get a proper pay back from users on Linux. 
>
>> Why does it take so long to come up with new
>> versions of Dolphin? I think it is because
>> there are too few programmers working on that.
>
> There is not enough of a market to drive it; that is sad but true.
Exactly what I wrote
>
>
>
>> My guess is there are probably less than 10
>> programmers working on the next Dolphin version.
>> Am I correct?  If Dolphin was open source, there
>> could be hundreds of programmers working on it.
>
> Open source makes financial sense "only" if the product in question is
> a means to an end.
Exactly. 



>  In other words, OA would have to want Dolphin for
> other purposes, in which case they would gain an improved and expanded
> Dolphin to use to make money doing things with Dolphin.
Isn't Smalltalk/X an example for that?



>  With their
> focus on creating Dolphin as an end product, opening the sources would
> offer no gain and would close their revenue stream, which is a form of
> fiscal suicide.
Definitly. 


> I'm typing this on a Gnome based system that I very much
> enjoy using. Some of the heavy Unix systems are apparently taking a
> beating from the competition, but that makes sense as the first place
> that a unix-like open system would score its first points.
If SCO would not have been that stupid and expensive, nobody would talk
about Linux. Because SCO Unix would be the "Unix" for the people. But
they found it very attractive to be extremly expensive and so they left
the market open for MS and their DOS....

There was a time, that SCO Unices had everything. A stable OS, even a
useful GUI and tons of rother things....



>
> I got an ear-full just today from an MS junkie (they can do no wrong
> in his view) about how Red Hat is toast and Oracle will be next.  I
> don't pretend to know.
Well Sun has gone but havn't they had this "great" open-source Java?

Regards
Friedrich
-- 
Please remove just-for-news- to reply via e-mail.
0
Friedrich
3/11/2010 6:50:23 AM
Except... that's pretty much not how it goes with large companies.  We
(Cincom) have many contracts with large companies, and whether the
code is OSS doesn't come up.  They cover themselves on that issue by
making sure that a code escrow arrangement has been made.  The only
OSS issue that arises is one of license.  Most large companies
distrust the GPL and the LGPL.  I'm not going to get into why, or
whether it's rational; it just is.  They're ok with the MIT, BSD,
Apache, (and a raft of others).

I agree with what other people have said about an OSS approach for
Dolphin, or for that matter, for Cincom's products.  OSS is a means to
an end, not an end in and of itself.  If that end doesn't conform with
the business model, it probably isn't going to work for that
company...

On Mar 10, 10:58=A0pm, Panu <p...@nospam.com> wrote:
>
> Consider the case currently: =A0If I am IBM
> I don't want to bet my future on Dolphin
> because Oracle might any day BUY OA. Likewise
> Oracle would not want to become dependent
> on OA's products because IBM might buy it
> any day.
>
> If Dolphin was open source, BOTH Oracle
> and IBM could trust it, and OA would get
> many more paying support customers.
>
> -Panu Viljamaa

0
jarober
3/11/2010 11:23:36 AM
Wilhelm K. Schwab wrote:
> 
> The key word is "feel."  I have known many people who dismiss Smalltalk, 
>  none of whom appeared to have any dynamic language experience.

Totally agree. But "feeling" is a reality,
based on which buyers make decisions.


>...  Did not IBM dump VA Smalltalk off to Instantiations?  You are dreaming.

IBM came to the conclusion they did not want
to be a Smalltalk VENDOR, for whatever reason.
Maybe one reason was that IBM's customers thought
VisualAge was too proprietary, high-priced product,
so they preferred  Java as an "essentially
non-proprietary" and "free" platform.

Why didn't IBM turn VA into open-source then?
Maybe they had invested too much in Java already,
and VA didn't run on enough platforms, perhaps.
And existing customers would have felt betrayed
after having payed thousands in licenses for a
product that was now given away for free.


>> OA might further use their superior knowledge
>> of Dolphin to build superior commercial
>> applications, or programming tools on top
>> of the free open-source platform.
> 
> You are not the first to suggest that model, but OA was interested in 
> being a tool vendor rather than a consulting firm.

I think it makes sense to be a tool-vendor as long as
your tools run on top of de facto standard libraries,
AND IF you can make tools that are significantly better
than what is available in open source.

Dolphin's tools run on top of Dolphin's libraries
which are are not de facto standard. If those
base libraries were made open source, that would
change the equation somewhat, I think.


Smalltalk was *always* "open" source in the sense
that you could see and edit its source-code
as you wished. Lisp-machines (Symbolics etc.)
had the same mind-set. Their creators had the
insight that it was essential you should have
access to all the source all the time.

What they or their financiers missed is the
following: Not only is it important that
you can modify the source-code of your
libraries easily, it is also important
you can SHARE your modifications with
fellow programmers, who can then improve
upon your improvements, and so on.

That is one reason why I think Lisp machines
failed, and why Smalltalk still is a niche.


Maybe this points to that OA could
structure its licenses in a way that
parts of their platform would be open
source, parts of it commercial.

-Panu Viljamaa

0
Panu
3/12/2010 3:26:11 AM
Friedrich Dominicus wrote:

> Well Sun has gone but havn't 
 > they had this "great" open-source Java?

Whether Java is great or not I think
we both agree on.

But it is great in the sense that it
is still around and has spawned a lot
of efforts to make something better
than itself, running on the same VM
and being able to use the same growing
library-base. (Groovy, Scala, Closure, etc.)

Regards
-Panu Viljamaa

0
Panu
3/12/2010 4:08:26 AM
Panu,

> The world of programming platforms is all
> about natural selection i.e. survival of
> the "fittest". Open source is more likely
> to survive because it can more easily
> adapt, to "fit" to changing circumstances
> - because anybody can modify it.

I don't really want to get embroiled in the Open Source argument again 
but I'm not convinced that opening Dolphin would make it any more of a 
success than it already is.

As an example, take the two already Open Source Smalltalks; Squeak and 
GNU. Although Squeak almost certainly does have a user base larger than 
Dolphins I would not categorise it as a runaway success.  As an 
indication of this, in the Google Summer of Code thread recently started 
on c.l.s, it was suggested that the reason that Squeak was not accepted 
into GSOC last year was because the community was too small. After 13 
years this is hardly an indication that Open Source is a panacea for 
success (no offence intended to the Squeak community). In the case of 
GNU, I don't have any real statistics, but I wouldn't say that this 
community was growing wildly either. What makes you think that adding a 
third Open Source Smalltalk would significantly change things?

There are a lot of reasons why Open Source can be a good idea from a 
customer/user's point of view, and you pointed these out in your posts. 
  However, there are also managerial and design issues involved in 
getting a disparate community to work cohesively on a project. Squeak 
has famously been plagued with these sorts of issues, as witnessed now 
by the breakaway Pharao project (which is a great idea IMO).

It seems to me that there are two sorts of OS project; ones which are 
funded and managed by establishments with alternate sources of income 
and those that are run purely as community projects. In the first 
category comes Java (Sun, IBM) and Eclipse (IBM). These tend to be the 
more successful ones. But even this doesn't guarantee success -- Squeak 
was supported initially by large companies, Apple and Disney, and still 
wasn't able to reach critical mass.

The community projects are on the whole much less successful and liable 
to come and go. For whatever reasons, very few make it big (if any). 
Whatever happened to the Bistro Open Source Smalltalk from 2002, for 
example?

In the past, our community has done more than it's fair share of "navel 
gazing"; analyzing why Smalltalk has never reached critical mass and 
taken off like the other more mainstream languages. Blame has been laid 
at many doors - dynamic binding, no source files, everything is an 
object, keyword syntax, weird image idea, lack of large vendors and, inn 
this thread, lack of Open Source. But I think the problem is more 
fundamental than that - it is simply that the rest of the world (i.e. 
outside those who "get" Smalltalk) just doesn't perceive that there is 
problem with what they currently have.

Alan Kay has does several talks on "The Computer Revolution Hasn't 
Happened Yet". Interestingly, there are at least 3 different versions 
each with a substantially different focus. Perhaps the best is the 1997 
OOPSLA talk but take a look at this one from his ACM Turing Prize 
lecture of 2004:

http://awards.acm.org/images%5Cawards%5C140%5Cvstream%5C2003%5CAlanKay%5CAK768kFull.mov

Get to about a quarter of the way in when he starts talking about 
different classes of motivation and then "forest fires" and this gives a 
clue as to why Smalltalk hasn't made it in the mainstream (sadly the 
graphics he uses aren't seen on the video so it's a bit difficult to 
follow).

Until the 80% of outer motivated/practical programmers actually  "get" 
the ideas in Smalltalk by coming to realize that they are things that 
they want or need *now*, we are going to carry on getting nowhere fast.

The rest of the talk is good too, and relevant, so don't stop there.

Best regards,

Andy Bower
0
Andy
3/12/2010 11:26:58 AM
Am 12.03.2010 12:26, schrieb Andy Bower:
> Panu,
>
>> The world of programming platforms is all
>> about natural selection i.e. survival of
>> the "fittest". Open source is more likely
>> to survive because it can more easily
>> adapt, to "fit" to changing circumstances
>> - because anybody can modify it.
>
> I don't really want to get embroiled in the Open Source argument again
> but I'm not convinced that opening Dolphin would make it any more of a
> success than it already is.

Ah. But nobody can predict, what will happen. What about e.g. BLENDER? 
Blender became really successful after opening the code.

I think, Pharo Smalltalk can become be your biggest competitor, since it 
runs on all platforms and has a very free license.

Perhaps we will wittness renaissance of Smalltalk Programming Language soon?

Have fun, Guido Stepken

0
Guido
3/12/2010 6:31:43 PM
Guido Stepken wrote:

> I think, Pharo Smalltalk can become be your biggest competitor, since it 
> runs on all platforms and has a very free license.
> 
> Perhaps we will wittness renaissance of Smalltalk Programming Language 
> soon?


I agree. I hope Dolphin will unify its base-
libraries with Pharo.  After that I could
more trustingly program in the Dolphin IDE
without fear that someone will do a Lucy
of C. Brown series on me.

 From a developer's standpoint the issue is
how do I know which parts of the Dolphin
base-library are standard / portable, which
not.

-Panu Viljamaa
0
Panu
3/17/2010 3:59:35 AM
Andy Bower wrote:

> I don't really want to get embroiled in the Open Source argument again 
> but I'm not convinced that opening Dolphin would make it any more of a 
> success than it already is.
>

Good points. It is hard to say why
one thing succeeds and not another.
It may be because of timing.

Why isn't Squeak a better success?
One possibility: No serious corporate
sponsors behind it. With Dolphin
this could be different IF parts
of it were made Open Source AND
sponsored by OA.

Just speculation on my part, but makes
sense, to me.


-Panu Viljamaa





0
Panu
3/17/2010 4:15:43 AM
Panu,

>> I think, Pharo Smalltalk can become be your biggest competitor, since 
>> it runs on all platforms and has a very free license.
>>
>> Perhaps we will wittness renaissance of Smalltalk Programming Language 
>> soon?
> 
> I agree. I hope Dolphin will unify its base-
> libraries with Pharo.  After that I could
> more trustingly program in the Dolphin IDE
> without fear that someone will do a Lucy
> of C. Brown series on me.

What do you mean by unify the base libraries? Which bits of Dolphin are 
valuable enough for you to consider paying for if they were available 
under Pharo? The VM; the MVP framework; COM support; Windows native 
widget support; the IDE browsers and tools?  You see, all of these 
things adds a little piece to the puzzle that overall makes up Dolphin 
and to move them all to Pharo would be a huge undertaking.

Incidentally, if anyone would like to license the Dolphin code to port 
to Pharo (or wherever) on a commercial basis then please let me know.

> 
>  From a developer's standpoint the issue is
> how do I know which parts of the Dolphin
> base-library are standard / portable, which
> not.

By portable, do you mean which classes in the Dolphin code can you 
legally take and run in another IDE such as Pharo/VW etc?

If so, the Dolphin License allows classes in the base image that have 
been marked as "redistributable" (nearly all apart from the IDE tools) 
to be distributed in "object code form" as part of an application. It 
does not say that the source code can't be first moved to Pharo/VW/VAST 
and then compiled and redistributed from there. So I guess that we 
(Object Arts) would not object to you moving the Dolphin codebase into 
another IDE providing a) you have a valid license for the code in the 
first place and b) you don't share the ported source code with anyone else.

Apart from that, none of the Dolphin base image code can be classed as 
open/free source.

Best regards

Andy Bower
0
Andy
3/17/2010 11:29:11 AM
> I think, Pharo Smalltalk can become be your biggest competitor, since it
> runs on all platforms and has a very free license.

I don't see how Pharo could become a competitor for shareware
developers because, as far as I know, one can't make an exe from it.

Don
0
Don
3/17/2010 11:38:59 AM
Am 17.03.2010 12:38, schrieb Don:
>> I think, Pharo Smalltalk can become be your biggest competitor, since it
>> runs on all platforms and has a very free license.
>
> I don't see how Pharo could become a competitor for shareware
> developers because, as far as I know, one can't make an exe from it.
>
> Don

Well, Eliot Miranda, a former developer of VisualWorks Smalltalk is 
working on that: http://www.mirandabanda.org/cogblog/

Should be easy to add a image encryption too.

Have fun, Guido Stepken
0
Guido
3/17/2010 4:17:57 PM
Em 17/3/2010 09:38, Don escreveu:
>> I think, Pharo Smalltalk can become be your biggest competitor, since it
>> runs on all platforms and has a very free license.
>
> I don't see how Pharo could become a competitor for shareware
> developers because, as far as I know, one can't make an exe from it.
>
Don,

Once you finish your application and remove all classes you don't 
want/need, you create an almost 'native' application renaming the VM 
binary to the name of your program.

-- 
Cesar Rabak
GNU/Linux User 52247.
Get counted: http://counter.li.org/
0
Cesar
3/18/2010 2:28:03 AM
Panu <panu@nospam.com> writes:

> Andy Bower wrote:
>
>> I don't really want to get embroiled in the Open Source argument
>> again but I'm not convinced that opening Dolphin would make it any
>> more of a success than it already is.
>>
>
> Good points. It is hard to say why
> one thing succeeds and not another.
> It may be because of timing.
>
> Why isn't Squeak a better success?
Oh that is easy, because of Morphic would be high on the list.
There is not one useful book available for it and it's programmming
model does not fit anything in every other language around (maybe there
are some exceptons I'd like to learn), I just can compare it to CLIM
which I'd argue hardly anyone can use. But it once was the base for all
Common Lisp GUI stuff.

> One possibility: No serious corporate
> sponsors behind it.
Second possibility: too many big egos clashing.



>  With Dolphin
> this could be different IF parts
> of it were made Open Source AND
> sponsored by OA.
Well and who will pay the OA programmers?

>
> Just speculation on my part, but makes
> sense, to me.
Not to me, I disagree

Friedrich

-- 
Please remove just-for-news- to reply via e-mail.
0
Friedrich
3/18/2010 6:48:28 AM
> Once you finish your application and remove all classes you don't
> want/need, you create an almost 'native' application renaming the VM
> binary to the name of your program.
>
> --
> Cesar Rabak
> GNU/Linux User 52247.
> Get counted:http://counter.li.org/

Thanks Cesar. I might experiment with that when I have time.

Don
0
Don
3/18/2010 11:32:16 AM
Don wrote:
>> I think, Pharo Smalltalk can become be your biggest competitor, since it
>> runs on all platforms and has a very free license.
> 
> I don't see how Pharo could become a competitor for shareware
> developers because, as far as I know, one can't make an exe from it.
> 
> Don

That depends on what you mean by "an .exe" - there is nothing to stop 
deployment to something that an end-user sees as a shortcut.  The vm is 
the executable, and it can be pointed to an image that has development 
tools suppressed.  The result is not a single executable file, but 
neither is anything on Windows, as one would be hard pressed to find 
something that does not link to the user, etc. libraries.

See Sophie for a partial answer, though Sophie as distributed has a way 
to exit or quit (forget which) into the Squeak IDE.  Doing that after 
running Pharo for a while is illuminating!

Bill
0
Wilhelm
3/19/2010 2:10:57 AM
Friedrich Dominicus wrote:
> Panu <panu@nospam.com> writes:
> 
>> Andy Bower wrote:
>>
>>> I don't really want to get embroiled in the Open Source argument
>>> again but I'm not convinced that opening Dolphin would make it any
>>> more of a success than it already is.
>>>
>> Good points. It is hard to say why
>> one thing succeeds and not another.
>> It may be because of timing.
>>
>> Why isn't Squeak a better success?
> Oh that is easy, because of Morphic would be high on the list.
> There is not one useful book available for it and it's programmming
> model does not fit anything in every other language around (maybe there
> are some exceptons I'd like to learn), I just can compare it to CLIM
> which I'd argue hardly anyone can use. But it once was the base for all
> Common Lisp GUI stuff.

Morphic itself is not really the problem, but the HORRENDOUS user 
interface of Squeak certainly is a factor.  Look at Zurgle (which the 
Squeak community simply ignored) and Polymorph (which they also tried to 
ignore - Pharo has embraced it), and Morphic starts to look (and feel!!) 
more capable.  I'm not defending it, only pointing out that the true 
deficiencies are in how Morphic is used in Squeak.


>> One possibility: No serious corporate
>> sponsors behind it.
> Second possibility: too many big egos clashing.

:)



>>  With Dolphin
>> this could be different IF parts
>> of it were made Open Source AND
>> sponsored by OA.
> Well and who will pay the OA programmers?

Painfully true.


>> Just speculation on my part, but makes
>> sense, to me.
> Not to me, I disagree

Ditto.

Bill
0
Wilhelm
3/19/2010 2:16:11 AM
In personal view: morphic or not - it does not matter that much, the
result of the GUI is pretty worse and that stopped my from working
with Squeak at all - I feel no fun, when playing around with that GUI.

Even when playing with Croquet I have these problems ....

>
> Morphic itself is not really the problem, but the HORRENDOUS user
> interface of Squeak certainly is a factor. =A0
0
Marten
3/19/2010 8:04:54 AM
Marten wrote:
> In personal view: morphic or not - it does not matter that much, the
> result of the GUI is pretty worse and that stopped my from working
> with Squeak at all - I feel no fun, when playing around with that GUI.
> 
> Even when playing with Croquet I have these problems ....
> 
>> Morphic itself is not really the problem, but the HORRENDOUS user
>> interface of Squeak certainly is a factor.  

No argument: Squeak's IDE is awful, but Morphic itself is perfectly 
capable of supporting a "simulation" of a solid GUI.  Have you looked at 
Pharo?  For historical perspective, dig up a copy of Zurgle (circa 
Squeak 3.6 IIRC) and see what was available and *completely* ignored.

Bill

0
Wilhelm
3/19/2010 12:26:34 PM
"Wilhelm K. Schwab" <bschwab@anest.ufl.edu> writes:

> Marten wrote:
>> In personal view: morphic or not - it does not matter that much, the
>> result of the GUI is pretty worse and that stopped my from working
>> with Squeak at all - I feel no fun, when playing around with that GUI.
>>
>> Even when playing with Croquet I have these problems ....
>>
>>> Morphic itself is not really the problem, but the HORRENDOUS user
>>> interface of Squeak certainly is a factor.  
>
> No argument: Squeak's IDE is awful, but Morphic itself is perfectly
> capable of supporting a "simulation" of a solid GUI.  Have you looked
> at Pharo?  For historical perspective, dig up a copy of Zurgle (circa
> Squeak 3.6 IIRC) and see what was available and *completely* ignored.
that is true, but you have to know how to program such stuff. AFAIKT the
only GUI Builder is something with Bob in it's name and it does not run
in current squeaks or Pharos

Regards
Friedrich

-- 
Please remove just-for-news- to reply via e-mail.
0
Friedrich
3/20/2010 4:51:24 AM
Reply: