Re: [Pharo-project] [smalltalk-research] Re: [Esug-list] Google Summer Of Code 2010 news!!!
On Sun, Mar 7, 2010 at 3:24 PM, Gilad Bracha <gbracha@gmail.com> wrote:
I'm all for it, and hope that John or Eliot can mentor.
I would like to mentor something in this area, but I think the basic goal of a cross-platform system is not achievable in GSoC this summer. A specification of a non-blocking architecture and a reference implementation may be possible. Cross-platform is too big for a summer project and will require input from dialects and vendors. What affects my thinking is the following: 1. I designed and implemented the VisualWorks THAPI system, a threaded extension to the VisualWorks FFI. 2. I have a functional prototype of a multi-threaded FFI for the Cog VM based on David Simmons' VMs. 1's architecture is poor, being complex, slow and not very general (allows only FFI calls to be threaded). 2's architecture is rather simple, rather fast and very general (allows threading to be much more widely used, e.g. calls within plugins can be made threaded by surrounding them with two calls, disownVM and ownVM). 3. I have looked at marshalling semantics in the context of the VW FFI, Andreas' Squeak FFI & Alien, and am still of the opinion, one shared by David, - that the generation of marshalling stubs has to be lifted up into Smalltalk based on simple ABI compilers that somehow (e.g. through RTL) communicate down to a JIT that generates and caches marshalling code (i.e. that VM attempts to interpret simplified call specs, as happens in the VM and Squeak FFIs are in general doomed to poor performance and bugs) - that type coercion code also has to be lifted (e.g. character conversions on strings, null termination, type coercion) - that one be able to depend on a pinning garbage collector to deal with the issues of passing objects through the FFI (although there are workaround hacks that can serve temporarily) I don't see the point of spending a GSoC trying to implement an obsolete (unthreaded, lowest-common-denominator) cross-platform FFI; it won't get used. I do see benefit in an open-source reference implementation that provides some of the above (at least threading/non-blocking). If people agree that an open-source reference implementation is worth pursuing, then I will happily mentor. What the starting point is will depend on to what extent Cog has been open sourced (Teleplace may choose to open source single-threaded Cog initially, keeping back the threaded FFI for a while, it may not open source Cog at all; we'll see :) ). There are other starting points (current Squeak VMs, GST, Newspeak). But the outline above is a move towards better cross-platform commonality, not less, because it's an architecture in which the image does more (type coercion, compiling call signatures to RTL sequences) and the VM (except for the threading ;) ) does less. And so this direction does lead towards better cross-platform facilities. best Eliot
Eliot
What the starting point is will depend on to what extent Cog has been open sourced (Teleplace may choose to open source single-threaded Cog initially, keeping back the threaded FFI for a while, it may not open source Cog at all; we'll see :) ).
May be I the only one to notice the :) which I have problem to understand since for me it announces that COG may not be open-source. What would help teleplace to believe that open-sourcing it would be a win-win situation? A bigger community of contributors ? Making sure that the truck factor does not touch them in case elliot is called under other horizons? Now there is a bootstrap problem. How can a community show that an open-source model is better than a closed one, if the software is closed? STef
On 03/10/2010 05:06 PM, Stéphane Ducasse wrote:
What the starting point is will depend on to what extent Cog has been open sourced (Teleplace may choose to open source single-threaded Cog initially, keeping back the threaded FFI for a while, it may not open source Cog at all; we'll see :) ). May be I the only one to notice the:) which I have problem to understand since for me it announces that COG may not be open-source.
Isn't this what you wanted to allow companies to do, when you chose the MIT license? I don't understand, why should you care? I see some irony... Paolo
What the starting point is will depend on to what extent Cog has
been open sourced (Teleplace may choose to open source single-threaded Cog initially, keeping back the threaded FFI for a while, it may not open source Cog at all; we'll see :) ). May be I the only one to notice the:) which I have problem to understand since for me it announces that COG may not be open-source.
Isn't this what you wanted to allow companies to do, when you chose the MIT license? I don't understand, why should you care?
I see some irony... Not me. Freedom of choice is a political attitude. I understand GPL goal but I do not adhere to it. I respect people pushing it but not in my way. I'm not sure that we should debate that here but we do not have the single answer.
2010/3/10 Stéphane Ducasse <stephane.ducasse@inria.fr>:
What the starting point is will depend on to what extent Cog has
been open sourced  (Teleplace may choose to open source single-threaded Cog initially, keeping back the threaded FFI for a while, it may not open source Cog at all; we'll see :) ). May be I the only one to notice the:)  which I have problem to understand since for me it announces that COG may not be open-source.
Isn't this what you wanted to allow companies to do, when you chose the MIT license? Â I don't understand, why should you care?
We shouldn't. Well, except if previous annoucements strongly suggested this would be the case...
I see some irony...
Not me. Freedom of choice is a political attitude. I understand GPL goal but I do not adhere to it. I respect people pushing it but not in my way. I'm not sure that we should debate that here but we do not have the single answer.
Not sure the goals differ much, but indeed these are two radically different strategies. The question is: would COG have ever started under a GPL derivative? Who knows? Since it did not happen, current choice is between an hypothetical something MIT or nothing... Bah, at least we already get a closure VM in Squeak. Nicolas
_______________________________________________ Pharo-project mailing list Pharo-project@lists.gforge.inria.fr http://lists.gforge.inria.fr/cgi-bin/mailman/listinfo/pharo-project
On Wed, Mar 10, 2010 at 7:37 PM, Nicolas Cellier < nicolas.cellier.aka.nice@gmail.com> wrote:
2010/3/10 Stéphane Ducasse <stephane.ducasse@inria.fr>:
What the starting point is will depend on to what extent Cog has
been open sourced (Teleplace may choose to open source single-threaded Cog initially, keeping back the threaded FFI for a while, it may not open source Cog at all; we'll see :) ). May be I the only one to notice the:) which I have problem to understand since for me it announces that COG may not be open-source.
Isn't this what you wanted to allow companies to do, when you chose the MIT license? I don't understand, why should you care?
We shouldn't. Well, except if previous annoucements strongly suggested this would be the case...
I see some irony...
Not me. Freedom of choice is a political attitude. I understand GPL goal but I do not adhere to it. I respect people pushing it but not in my way. I'm not sure that we should debate that here but we do not have the single answer.
Not sure the goals differ much, but indeed these are two radically different strategies. The question is: would COG have ever started under a GPL derivative? Who knows? Since it did not happen, current choice is between an hypothetical something MIT or nothing... Bah, at least we already get a closure VM in Squeak.
I would guess that Cog would not have started because if squeak were under GPL, squeak wouldn't have been used by Teleplace to begin with...of course, that's just speculation on my part (and Cog may very well have gotten started under different circumstances). I actually appreciate the role that the GPL played in the evolution of the GNU Unix tools. Without GPL, the Unix vendors would very likely have simply co-opted that code and sucked the life out of the GNU project very early on. I don't believe GPL should be used for squeak however (and I think there are general problems with that license as it relates to Smalltalk code (i.e. it was written with C like linking in mind)). What I believe is needed is a license that has time limited, GPL like requirements for sharing enhancements and after that time period reverts to a pure and simple MIT license (where the conversion date can be chosen by the author). It is essential that such a date be specified in the license upon initial release. Each new version would also come under a new license that could have a timeout further into the future. That would help ensure that a project isn't co-opted early on by commercial interests while simultaneously ensuring that at a defined point in time, it becomes available for anyone to use without any restrictions of any kind (except the limitations on liability in the MIT license). It also would not preclude commercial interests from paying for a different license in that early period. - Stephen
Well, I certainly care.. Stef wasn't remarking about license rights, he said didn't understand the smiley face. Frankly, I don't either. After years of teasing from Bryce and now this? What a crushing disappointment. On Wed, Mar 10, 2010 at 6:37 PM, Nicolas Cellier <nicolas.cellier.aka.nice@gmail.com> wrote:
2010/3/10 Stéphane Ducasse <stephane.ducasse@inria.fr>:
What the starting point is will depend on to what extent Cog has
been open sourced  (Teleplace may choose to open source single-threaded Cog initially, keeping back the threaded FFI for a while, it may not open source Cog at all; we'll see :) ). May be I the only one to notice the:)  which I have problem to understand since for me it announces that COG may not be open-source.
Isn't this what you wanted to allow companies to do, when you chose the MIT license? Â I don't understand, why should you care?
We  shouldn't. Well, except if previous annoucements strongly suggested this would be the case...
I see some irony...
Not me. Freedom of choice is a political attitude. I understand GPL goal but I do not adhere to it. I respect people pushing it but not in my way. I'm not sure that we should debate that here but we do not have the single answer.
Not sure the goals differ much, but indeed these are two radically different strategies. The question is: would COG have ever started under a GPL derivative? Who knows? Since it did not happen, current choice is between an hypothetical something MIT or nothing... Bah, at least we already get a closure VM in Squeak.
Nicolas
_______________________________________________ Pharo-project mailing list Pharo-project@lists.gforge.inria.fr http://lists.gforge.inria.fr/cgi-bin/mailman/listinfo/pharo-project
_______________________________________________ Pharo-project mailing list Pharo-project@lists.gforge.inria.fr http://lists.gforge.inria.fr/cgi-bin/mailman/listinfo/pharo-project
On 12 March 2010 00:31, Chris Muller <asqueaker@gmail.com> wrote:
Well, I certainly care.. Â Stef wasn't remarking about license rights, he said didn't understand the smiley face. Â Frankly, I don't either. After years of teasing from Bryce and now this? Â What a crushing disappointment.
I think that Teleplace realizing that there is no reason to keep this technology private. Simply because there is nothing too much compelling comparing to other open-source implementations which already having open-sourced JITs. Instead, by giving it away, it would increase their own chances to success, because it will attract more developers, more professionals to Squeak & friends and rise it visibility. Oh.. unless Teleplace wants to use marketing strategy to hide the fact that their product is based on Squeak and implemented in smalltalk. I'm only hope that people won't ashame themselves with postings like this: http://www.opengl.org/news/comments/croquet_sdk_v10_opengl_api_based_open_so... a âlate-binding object-orientedâ programming language WTF?!?!?!
On Wed, Mar 10, 2010 at 6:37 PM, Nicolas Cellier <nicolas.cellier.aka.nice@gmail.com> wrote:
2010/3/10 Stéphane Ducasse <stephane.ducasse@inria.fr>:
What the starting point is will depend on to what extent Cog has
been open sourced  (Teleplace may choose to open source single-threaded Cog initially, keeping back the threaded FFI for a while, it may not open source Cog at all; we'll see :) ). May be I the only one to notice the:)  which I have problem to understand since for me it announces that COG may not be open-source.
Isn't this what you wanted to allow companies to do, when you chose the MIT license? Â I don't understand, why should you care?
We  shouldn't. Well, except if previous annoucements strongly suggested this would be the case...
I see some irony...
Not me. Freedom of choice is a political attitude. I understand GPL goal but I do not adhere to it. I respect people pushing it but not in my way. I'm not sure that we should debate that here but we do not have the single answer.
Not sure the goals differ much, but indeed these are two radically different strategies. The question is: would COG have ever started under a GPL derivative? Who knows? Since it did not happen, current choice is between an hypothetical something MIT or nothing... Bah, at least we already get a closure VM in Squeak.
Nicolas
_______________________________________________ Pharo-project mailing list Pharo-project@lists.gforge.inria.fr http://lists.gforge.inria.fr/cgi-bin/mailman/listinfo/pharo-project
_______________________________________________ Pharo-project mailing list Pharo-project@lists.gforge.inria.fr http://lists.gforge.inria.fr/cgi-bin/mailman/listinfo/pharo-project
_______________________________________________ Pharo-project mailing list Pharo-project@lists.gforge.inria.fr http://lists.gforge.inria.fr/cgi-bin/mailman/listinfo/pharo-project
-- Best regards, Igor Stasenko AKA sig.
On Thu, 2010-03-11 at 16:31 -0600, Chris Muller wrote:
Well, I certainly care.. Stef wasn't remarking about license rights, he said didn't understand the smiley face. Frankly, I don't either. After years of teasing from Bryce and now this? What a crushing disappointment.
I'm still here.... And still making progress though slowly. Bryce
Bryce Kampjes a écrit :
On Thu, 2010-03-11 at 16:31 -0600, Chris Muller wrote:
Well, I certainly care.. Stef wasn't remarking about license rights, he said didn't understand the smiley face. Frankly, I don't either. After years of teasing from Bryce and now this? What a crushing disappointment.
I'm still here.... And still making progress though slowly.
Can't you get support to progress more easily? Should not be difficult regarding the importance of your work? Hilaire
On Tue, 2010-03-23 at 19:30 +0100, Hilaire Fernandes wrote:
Bryce Kampjes a écrit :
On Thu, 2010-03-11 at 16:31 -0600, Chris Muller wrote:
Well, I certainly care.. Stef wasn't remarking about license rights, he said didn't understand the smiley face. Frankly, I don't either. After years of teasing from Bryce and now this? What a crushing disappointment.
I'm still here.... And still making progress though slowly.
Can't you get support to progress more easily? Should not be difficult regarding the importance of your work?
Help's always welcome but it's likely easier to help after Exupery supports closures. Bryce
Bryce Kampjes a écrit :
On Tue, 2010-03-23 at 19:30 +0100, Hilaire Fernandes wrote:
Bryce Kampjes a écrit :
On Thu, 2010-03-11 at 16:31 -0600, Chris Muller wrote:
Well, I certainly care.. Stef wasn't remarking about license rights, he said didn't understand the smiley face. Frankly, I don't either. After years of teasing from Bryce and now this? What a crushing disappointment. I'm still here.... And still making progress though slowly. Can't you get support to progress more easily? Should not be difficult regarding the importance of your work?
Help's always welcome but it's likely easier to help after Exupery supports closures.
I was thinking more about financial support to encourage you in your effort. Hilaire
On Sun, 2010-03-28 at 16:41 +0200, Hilaire Fernandes wrote:
Bryce Kampjes a écrit :
On Tue, 2010-03-23 at 19:30 +0100, Hilaire Fernandes wrote:
Bryce Kampjes a écrit :
On Thu, 2010-03-11 at 16:31 -0600, Chris Muller wrote:
Well, I certainly care.. Stef wasn't remarking about license rights, he said didn't understand the smiley face. Frankly, I don't either. After years of teasing from Bryce and now this? What a crushing disappointment. I'm still here.... And still making progress though slowly. Can't you get support to progress more easily? Should not be difficult regarding the importance of your work?
Help's always welcome but it's likely easier to help after Exupery supports closures.
I was thinking more about financial support to encourage you in your effort.
That's not practical as it would create a conflict of interest. And money can't easily buy extra hours in the day. Though the children will get older and every month makes a big difference with our youngest. Bryce
participants (9)
-
Bryce Kampjes -
Chris Muller -
Eliot Miranda -
Hilaire Fernandes -
Igor Stasenko -
Nicolas Cellier -
Paolo Bonzini -
Stephen Pair -
Stéphane Ducasse