- Pharo-dev - lists.pharo.org

[pharo-project/pharo-core]

by GitHub

Branch: refs/tags/30844 Home: https://github.com/pharo-project/pharo-core

April 29, 2014

[pharo-project/pharo-core] e619ad: 30844

by GitHub

Branch: refs/heads/3.0 Home: https://github.com/pharo-project/pharo-core Commit: e619adda13ca6dcd51b7c760b5d71494b09ffdd7 https://github.com/pharo-project/pharo-core/commit/e619adda13ca6dcd51b7c760… Author: Jenkins Build Server <board(a)pharo-project.org> Date: 2014-04-29 (Tue, 29 Apr 2014) Changed paths: A ScriptLoader30.package/ScriptLoader.class/instance/pharo - scripts/script106.st A ScriptLoader30.package/ScriptLoader.class/instance/pharo - updates/update30844.st M ScriptLoader30.package/ScriptLoader.class/instance/public/commentForCurrentUpdate.st A Slot.package/PharoClassInstaller.class/instance/notifications/sharedVariableOrSharedPoolChangedFrom_to_by_.st R Slot.package/SlotClassBuilder.class/instance/accessing/client_.st A Slot.package/SlotClassBuilder.class/instance/deprecated/client_.st M Slot.package/SlotClassBuilder.class/instance/private/applyAndUpdateSharedVariableOrSharedPool_.st A SlotTests.package/SlotMigrationTest.class/instance/tests/testAddSharedVariableKeepSubclasses.st A SlotTests.package/SlotMigrationTest.class/instance/tests/testChangeSuperclass.st Log Message: ----------- 30844 13028 Adding ClassVariables corrupts class hierarchy https://pharo.fogbugz.com/f/cases/13028 http://files.pharo.org/image/30/30844.zip

April 29, 2014

Re: [Pharo-dev] [Pharo3] New Sources file starting 30841/30842

by Esteban A. Maringolo

Great news!!! Thanks people. Hopefully in Pharo 4 we won't need source files anymore :) Esteban A. Maringolo 2014-04-29 7:53 GMT-03:00 Marcus Denker <marcus.denker(a)inria.fr>: > Hi, > > We have a sources file for Pharo3! > > http://files.pharo.org/sources/PharoV30.sources.zip > http://files.pharo.org/sources/PharoV30.sources > > > The image that goes with it is 30 842 or newer. > zipped image is now 5MB smaller due to the empty changes. > > http://files.pharo.org/image/30/ > > this now needs some careful testingâ¦ the build is green > > https://ci.inria.fr/pharo/ > > > > > > Marcus > > > >

April 29, 2014

Re: [Pharo-dev] [Pharo-users] [Pharo3] New Sources file starting 30841/30842

by Sven Van Caekenberghe

On 29 Apr 2014, at 15:05, Norbert Hartl <norbert(a)hartl.name> wrote: > > Am 29.04.2014 um 14:19 schrieb Sven Van Caekenberghe <sven(a)stfx.eu>: > >> Tested on Mac OS X: >> >> $ mkdir pharo3 >> $ cd pharo3/ >> $ curl get.pharo.org/vm | bash >> $ curl get.pharo.org/30 | bash >> $ ./pharo-ui Pharo.image >> > Just a side note. You know there is a > > get.pharo.org/30+vm > > ? Yes. Of course I know ;-) I was explicitly going step by step since the source were new. > Norbert > >> Loaded some configurations, ran some tests, recompiled all. All OK. >> >> Super ! >> >> On 29 Apr 2014, at 12:53, Marcus Denker <marcus.denker(a)inria.fr> wrote: >> >>> Hi, >>> >>> We have a sources file for Pharo3! >>> >>> http://files.pharo.org/sources/PharoV30.sources.zip >>> http://files.pharo.org/sources/PharoV30.sources >>> >>> >>> The image that goes with it is 30 842 or newer. >>> zipped image is now 5MB smaller due to the empty changes. >>> >>> http://files.pharo.org/image/30/ >>> >>> this now needs some careful testingâ¦ the build is green >>> >>> https://ci.inria.fr/pharo/ >>> >>> >>> >>> >>> >>> Marcus >>> >>> >>> >>> >> >> > >

April 29, 2014

Re: [Pharo-dev] Amber-Pharo based coding dojo in a coworking space

by kilon alios

Sorry I meant "Do you seriously believe people will give *up* JS for Amber alternatives ? " On Tue, Apr 29, 2014 at 4:03 PM, kilon alios <kilon.alios(a)gmail.com> wrote: > I was talking mainly about the IDE and other areas where Pharo and Amber > overlap. I think Amber is awesome for what it is and should be kept > developed independently. > > "If your ass is being kicked by AngularJS <https://angularjs.org/> or > Meteor <https://www.meteor.com/> because you reach the goal faster using > those, your next contract is not to be powered by Amber." > > is this not the cases if you cannot beat them join them ? Do you seriously > believe Amber or Pharo can compete AngularJS , correct me if I am wrong, > when behind this framework is Google itself ? > > "In regard to Pharo, becoming Amber friendly is âattracting javascript > people to the smalltalk worldâ are you joking? is the best marketing > strategy Pharo can dream about!" > > Did I said anything against that ? I agree that is great. > > "With some add ons Amber can do the things Meteor does in the front end." > > Thats all great but again I ask why not leverage the power of Meteor.js > for Amber and Pharo in the first place and sprinkle it with some smalltalk > libraries to bring it closer to Smalltalk philosophy ? Isnt this the whole > point of doing OOP, to reuse code ? > > Why should it be Amber / Pharo vs the Others ? Instead as "Amber a way to > make your favorite javascript libraries even easier and more accessible / > dynamic" > > Do you seriously believe people will give JS for Amber alternatives ? > > No thats not a best marketing strategy. > >

April 29, 2014

Re: [Pharo-dev] a Pharo talk from a ruby conference

by Esteban A. Maringolo

2014-04-28 23:48 GMT-03:00 Sean P. DeNigris <sean(a)clipperadams.com>: > Esteban A. Maringolo wrote >> Plays well with "choose your favorite text editor" (Sublime, Vim, >> etc.) and IDEs (RubyMine, etc.), with source control systems (any file >> based system), with unix in general (several cli commands), has >> binding for any major/mainstream library* (databases, network, etc.). > > But again these boil down to community size/interest > - To use "your favorite text editor", Craig Latta serves Smalltalk via > WebDav [1], but who has jumped at this opportunity? > - source control - now that there is community interest, progress on git > support has been moving ahead rapidly with minimal resources This is my point, who uses WebDAV as its file support? It is a "workaround" to enable file based development. A VERY CLEVER one, but still. > - unix in general - with FFI and OSProcess, what can't you do? Are we > talking about the lack of cool Ruby backtick syntax? While definitely cool, > that special-purpose syntax is the kind of cognitive load Smalltalk > overcomes. All those little syntactical twists and turns to remember lead > away from "syntax on a T-shirt" to manuals with hundreds of pages I'm not talking about syntax, and I wouldn't trade Smalltalk syntax for anything else. FFI and OSProcess IS NOT unix interoperability, are ways to get "out" of the image (the island in Byte's '81 cover). Think about this... Windows and Unix/Linux culture. One started from an "all windows" architecture, whilst the other was the other way around. Windows is still thriving to get a command line culture for many of its products. Only in the last years they've been adding more and more features through command line, because of the demand of scripting and who knows what else. And maybe that "command line" need comes from people coming from the unix world. And, to me, this is "playing well with others", not being able to call "external" programs via any mechanism. I think that one big part of the success of Mac notebooks in the developer community, is the fact is has an underlying unix. Oversimplified example: You don't need a version of gzip with bindings to tar in order to have both working together. > - bindings - again, obviously just a question of community size and interest Here I agree. That's why I asked if you would recommend Pharo today to somebody who has current needs. > So the "play well with others" is a self-fulfilling prophecy. There are no > bindings because there are no people to write them because there are no > bindings... I have to agree with you about the self-fulfilling prophecy. > At inception, Ruby (and every other language) didn't have those bindings either. Ruby got to be what it is because of Rails. Because of 37signals. Pharo doesn't have that. It could have happened with Seaside. But it didn't. It worries me that when an outsider pinpoints so clearly what are our weak points we can't think outside of our box. Proportions aside, it remembers me when you talk with a 100% minded PHP programmer about something, and they can only think inside of their box, and instead reply that you can do OO in PHP and you don't need an interactive debugger.:) Having all that said, I still bet on Pharo. :) Regards, Esteban A. Maringolo

April 29, 2014

Re: [Pharo-dev] [Pharo-users] [Pharo3] New Sources file starting 30841/30842

by Norbert Hartl

Am 29.04.2014 um 14:19 schrieb Sven Van Caekenberghe <sven(a)stfx.eu>: > Tested on Mac OS X: > > $ mkdir pharo3 > $ cd pharo3/ > $ curl get.pharo.org/vm | bash > $ curl get.pharo.org/30 | bash > $ ./pharo-ui Pharo.image > Just a side note. You know there is a get.pharo.org/30+vm ? Norbert > Loaded some configurations, ran some tests, recompiled all. All OK. > > Super ! > > On 29 Apr 2014, at 12:53, Marcus Denker <marcus.denker(a)inria.fr> wrote: > >> Hi, >> >> We have a sources file for Pharo3! >> >> http://files.pharo.org/sources/PharoV30.sources.zip >> http://files.pharo.org/sources/PharoV30.sources >> >> >> The image that goes with it is 30 842 or newer. >> zipped image is now 5MB smaller due to the empty changes. >> >> http://files.pharo.org/image/30/ >> >> this now needs some careful testingâ¦ the build is green >> >> https://ci.inria.fr/pharo/ >> >> >> >> >> >> Marcus >> >> >> >> > >

April 29, 2014

Re: [Pharo-dev] Amber-Pharo based coding dojo in a coworking space

by kilon alios

I was talking mainly about the IDE and other areas where Pharo and Amber overlap. I think Amber is awesome for what it is and should be kept developed independently. "If your ass is being kicked by AngularJS <https://angularjs.org/> or Meteor<https://www.meteor.com/> because you reach the goal faster using those, your next contract is not to be powered by Amber." is this not the cases if you cannot beat them join them ? Do you seriously believe Amber or Pharo can compete AngularJS , correct me if I am wrong, when behind this framework is Google itself ? "In regard to Pharo, becoming Amber friendly is âattracting javascript people to the smalltalk worldâ are you joking? is the best marketing strategy Pharo can dream about!" Did I said anything against that ? I agree that is great. "With some add ons Amber can do the things Meteor does in the front end." Thats all great but again I ask why not leverage the power of Meteor.js for Amber and Pharo in the first place and sprinkle it with some smalltalk libraries to bring it closer to Smalltalk philosophy ? Isnt this the whole point of doing OOP, to reuse code ? Why should it be Amber / Pharo vs the Others ? Instead as "Amber a way to make your favorite javascript libraries even easier and more accessible / dynamic" Do you seriously believe people will give JS for Amber alternatives ? No thats not a best marketing strategy.

April 29, 2014

Re: [Pharo-dev] Amber-Pharo based coding dojo in a coworking space

by Sebastian Sastre

On Apr 29, 2014, at 9:35 AM, kilon alios <kilon.alios(a)gmail.com> wrote: > I think that Amber and Pharo should be merged, maybe not as code bases or not as projects or websites or whatever but as communities. The way I see it both work towards the same goal, making the Smalltalk voice heard loud and clear. I think a closer integration is for the benefit of both communities. The communities are too small to afford to spread thin. > > Inventing the future is all cool and great but this world runs on money , not on progress and innovation. In most cases money is a big obstacle to progress. > > "I love progress , I just dont like change" (sorry dont remember where I got the quote from) Well Iâd keep Amber free to use other backends. Specially node. But you bring very important points there. If your ass is being kicked by AngularJS or Meteor because you reach the goal faster using those, your next contract is not to be powered by Amber. So do fucking something about that, right? Okay.â¦ In regard to Pharo, becoming Amber friendly is âattracting javascript people to the smalltalk worldâ are you joking? is the best marketing strategy Pharo can dream about! With some add ons Amber can do the things Meteor does in the front end. With some add ons, Pharo can also do the backend things that Meteor does in its backend. Meteor is the new cool kid in the block and is proven itself worth using. With a little code in the back and in the front we can catch up with Meteor (which be honest! is currently kicking our ass big time).

April 29, 2014

Re: [Pharo-dev] a Pharo talk from a ruby conference

by Esteban A. Maringolo

2014-04-29 4:42 GMT-03:00 Frank Shearar <frank.shearar(a)gmail.com>: > On 29 April 2014 03:48, Sean P. DeNigris <sean(a)clipperadams.com> wrote: >> So the "play well with others" is a self-fulfilling prophecy. There are no >> bindings because there are no people to write them because there are no >> bindings... At inception, Ruby (and every other language) didn't have those >> bindings either. > > That's largely true, in the sense that someone needs to grab a shovel > to dig that trench. > > But if you start out with an external text editor, with an external > version control system, with (only) stdout/stderr/stdin, you end up > building a different system than if you're already in an insular > environment and want/need to learn to "play well with others". This is what I meant by built from the ground (cli) up. And sticking with unix philosophy. > Ruby plays well with other - interfaces well with external systems - > precisely because it didn't have that integrated environment. It NEEDS external tools, because there isn't a "Ruby" toolkit (as with any smalltalk distro). The concept of IDE is clearly separated from the language and vm. Of course it has its drawbacks. Regards!

April 29, 2014