Squeak-dev October 2005

squeak-dev@lists.squeakfoundation.org

180 participants
296 discussions

Election Team
by Cees De Groot 09 Nov '05

09 Nov '05

Hi, As Goran announced - http://lists.squeakfoundation.org/pipermail/squeak-dev/2005-October/095378.… - the SqF board thinks it would be a good idea to have an election team outside the board to take care of everything around succession. It doesn't need to be a large team, and there's not a lot of work involved - but it is a very responsible task, and the Election Team must jump into action whenever necessary, so we don't leave empty seats on the board for too long. And, of course, you must be prepared to defend whatever you come up with, procedure-wise, to the Squeak Community - that could be a bit of a daunting task ;-). Anyway, this is not a Formal Board Communique, but rather a reminder by yours truly that we think this Election Team thing is a good idea, and that if anyone volunteers to be part of it, we'd like to hear it (here on the list, preferrably, but otherwise mail to board(a)discuss.squeakfoundation.org - please don't contact any of us through private mail, we want to keep an archive of everything in this place). Regards, Cees --------------------------------------------------------------------- Repeating Goran's write-up for reference: Representational model --------------------------- Now when we merge we also think it is important to take a first, small careful step towards a more representative/democratic model for SqF. Note that the model below is NOT anywhere near a perfect model - but instead of getting caught in an endless debate we chose something simple and careful that we could quickly agree on. So please keep that in mind. :) The model is this: - We want to create an Election Team with a Team leader that is NOT in the SqF board. - Whenever someone of us current 7 board members decides to step down *and* the board wants to get a replacement, it is the job of the election Team, in *whichever* way it finds suitable, to come up with a replacement. This means that the first job of the Election Team is probably to establish some kind of voting procedure (or whatever) ready for use. - Whenever the board wants to recruit someone specific (or non specific) thus growing the board, we do it "through" the Election Team. - The Election Team also has the task to come up with a better future model than the above. Any such new election model it presents needs to be approved by a united board though. A few notes on the above: - We leave it to squeak-dev to discuss and find a suitable Election Team leader. We will only try to help if noone steps forward willing to pick up the ball. So consider the ball rolling. :) - We deliberately did not include an option for the board to refuse the replacement presented by the Election Team. We will trust the Election Team to do a good job and to communicate with the board during its process. - The full procedure to find a replacement is entirely delegated to the Election Team. If the team chooses to use public voting or throwing a dice :) - it is up to the Team. - When we want to recruit a specific person we will go through the election Team and simply give them our name and motivation. Then they use their process, whatever that might be. In theory they could end up rejecting that person or whatever. - Whichever new future model the Election Team produces, we still currently reserve the power to reject it. Yes, a bit on the cautious side, but as we said, the above model is not perfect. ;) - The fact that the model still leaves no options for the community to *initiate* any changes in the board (any initiative still comes from inside) should motivate the Election Team and the community to come up with a better model to replace this one :).

5 5

[ANN] Squeak Videos
by stéphane ducasse 08 Nov '05

08 Nov '05

Hi all after playing a bit with a nice tool on mac (SnapZ Pro), I produced the following list of videos showing all kinds of cool aspects of Squeak - class creation - method creation - workspace - method finder - monticello connection - squeaksource registration - squeakmap - unit test writing.... http://www.iam.unibe.ch/~ducasse/Web/Videos/Squeak/ So if you want to help - you can check if you can see these videos - convert them in any format (and send them to me, I will add them to the repository) If I have some time, I plan to do the same for - refactoring, and SmallLint, - seaside (avi :) you can do it too - debugger if one of you want to do it. Please do. I will do the same with VisualWorks now.... (tomorrow) Stef PS: it took me one afternoon so now we do not have any excuse not to have more of these. PSPS: my plan is to create a repository with all this material + the books + all my lectures and the ones I will find....

6 10

Instrumenting Send and what is StrikeFontSet>displayString:on:from:to:at:kern:baselineY: doing?
by John M McIntosh 08 Nov '05

08 Nov '05

Michael was sitting in my living room this evening and we were discussion performance issues and I was musing about adding an instrumentation printf to the Squeak VM to grab the method name on each execution of a message send. That way we *know* what messages are being sent in which order. So I decided to code up a test macintosh carbon VM that is found at hftp://ftp.smalltalkconsulting.com/ experimental/SqueakMethodTrace3.8.9b7.app.zip which dumps data to 'messageTraceFile.txt' One of the things I saw a lot of when opening an image, which generates 40MB of data, is this sequence. Process context Increment GC count method 107493580 111893480 16 StrikeFontSet>glyphInfoOf:into: 107493580 111893572 16 StrikeFont>ascentOf: 107493580 111893940 16 StrikeFont>hasGlyphOf: 107493580 111894032 16 Character>charCode 107493580 111893940 16 StrikeFont>hasGlyphOf: 107493580 111893940 16 StrikeFont>hasGlyphOf: 107493580 111894032 16 Magnitude>between:and: 107493580 111893940 16 StrikeFont>hasGlyphOf: 107493580 111893572 16 StrikeFont>ascentOf: 107493580 111893480 16 SequenceableCollection>first 107493580 111893480 16 SequenceableCollection>first 107493580 111893480 16 SequenceableCollection>second 107493580 111893572 16 SequenceableCollection>checkedAt: 107493580 111893572 16 SequenceableCollection>checkedAt: 107493580 111893480 16 SequenceableCollection>third 107493580 111893572 16 SequenceableCollection>checkedAt: 107493580 111893572 16 SequenceableCollection>checkedAt: 107493580 111893480 16 SequenceableCollection>fifth 107493580 111893572 16 SequenceableCollection>checkedAt: 107493580 111893572 16 SequenceableCollection>checkedAt: what is interesting is that StrikeFontSet>displayString:on:from:to:at:kern:baselineY: does self glyphInfoOf: (aString at: charIndex) into: glyphInfo. g := glyphInfo first. leftX := glyphInfo second. rightX := glyphInfo third. (glyphInfo fifth ~= aBitBlt lastFont) ifTrue: [ glyphInfo fifth installOn: aBitBlt. ]. Now for example SquenceableCollection>>fifth "Answer the fifth element of the receiver. Raise an error if there are not enough elements." ^ self checkedAt: 5 which does SquenceableCollection>>checkedAt: index index > self size ifTrue: [self error: 'not enough elements']. ^ self at: index which does the at: index which would in fact toss an self errorSubscriptBounds: when the primitive discovers a range check issue. So do we need the checkedAt: logic? Could we just not say fifth ^self at: 5 Of course saying (self at: 5) in displayString:on:from:to:at:kern:baselineY: would be way more efficient. Other fun things like Rectangle>insetBy: invokes isKindOf: versus saying isThisARectangle or something... String>convertFromWithConverter: is quite message sending happy. -- ======================================================================== === John M. McIntosh <johnmci(a)smalltalkconsulting.com> 1-800-477-2659 Corporate Smalltalk Consulting Ltd. http://www.smalltalkconsulting.com ======================================================================== ===

9 28

Three more squeak videos on video.google.com
by Marcus Denker 05 Nov '05

05 Nov '05

Hi, Three more videos are now on video.google.com: Dan Ingalls: Seven (give or take) Smalltalk implementations http://video.google.com/videoplay?docid=2409496407757723940 Alan Kay: Understanding powerful ideas, and how computers can help Kyoto Laureate Symposium http://video.google.com/videoplay?docid=-124161889929978086 Video showing Smalltalk 80. Keywords: Adele Goldberg, Smalltalk, GUI, Graphical User Interface, Object Oriented, Programming http://video.google.com/videoplay?docid=-4365247885921962429 Marcus

6 8

OpenAL 1.1
by Brad Fuller 04 Nov '05

04 Nov '05

The Game Developers Conference ended this week and Creative discussed OpenAL 1.1. This new version includes some extensions that may be beneficial to me and potentially to others. In a previous msg Mark mentioned that OpenAL plugins have been made and are used in Croquet. I have a few questions: * How can Squeak users obtain the plugins and use it in our current images? * How can OpenAL plugins be integrated into 3.9 to be available for all? * Has there been any discussion regarding extending the plugins to OpenAL v1.1? * To facilitate the migration from 1.0 to 1.1, the OpenAL community is providing a few extensions to v1.0. One of the main issues of OpenAL, for me anyway, is that it does not have facilities for recording. Fortunately, this has changed in v1.1. The 1.0 recording extension is: EXT_capture. Has anyone started work on this or on any other extensions? (more info on extensions: http://openal.org/extensions.html) brad -- ________________________________ Hear us online: www.Sonaural.com Brad Fuller (408) 799-6124 ** Sonaural Audio Studios ** (408) 799-6123 West San Jose (408) 799-6124 Cambrian

4 5

Pending Mac Carbon VM 3.8.9b7 posted
by John M McIntosh 04 Nov '05

04 Nov '05

I've posted a macintosh VM 3.8.9b7 See http://www.smalltalkconsulting.com/squeak.html It's fixes to issues raised by the 3.8.9b5 distribution. If there are no complaints raised over the next day or two (please check) then I'll ask if this can go into the current mac distribution dmg on the main download site. 3.8.9b7 sqMacMain.c #ifdef change to enable compile under os-9 sqMacUIAppleEvents.c Handle case where the file name ends with *.image but we have no hfs meta data. sqMacUIEvents.c #ifdef change to enable compile under os-9. sqMacWindow.c Recheck window handle on window open, or window title set. 3.8.9b2,b3,b4,b6 had limited distribution for testing This VM also contains the OS-9 executable in the bundle. -- ======================================================================== === John M. McIntosh <johnmci(a)smalltalkconsulting.com> 1-800-477-2659 Corporate Smalltalk Consulting Ltd. http://www.smalltalkconsulting.com ======================================================================== ===

6 12

About package responsible
by stéphane ducasse 03 Nov '05

03 Nov '05

Hi I was going over mantis to see what were the bugs and next actions I should integrate. So now I imagine that I should let the team/team leader work and they will produce new version of packages. Is it right? Here is the list I collected Andreas: Compression, FFI, Graphics, Balloon Goran: Network Juan: Morphic Cees: Files Stef

4 4

For the morphic team
by stéphane ducasse 03 Nov '05

03 Nov '05

Hi guys It would be good that after the PLM refactoring is integrated, morphicers get a look at http://bugs.impara.de/view.php?id=1083 because having multiple selection in list is miiiisssssssssiiiiiinnnnng. Stef

2 1

Project "Kolibri" help wanted
by Cees De Groot 03 Nov '05

03 Nov '05

Hi, As some of you may remember, I've been busy with a Squeak project the past 9 months or so under the banner of the Digital Society of the Past (DGV), a non-profit that aims to bring (Dutch) private persons and institutions together by means of peer-to-peer software. We built the software in wxSqueak on top of a homebrew p2p layer (dubbed 'Gossip'). During the project, we decided to name the base software 'Kolibri', so that's what the project is going to be called from now :) What the package does is organize the world in communities, which have files, members, etcetera. Also institutional collections can be made available, currently through the OAI protocol. Files serve as starting points for discussion threads, have Dublin Core metadata, and files, institutional collection items, etcetera can be linked together in relations (which themselves serve as starting points for discussions). Data travels over the network as required, and there's a (primitive) protocol to download blobs (file contents) from multiple sources at once. The software has largely been completed, and I'm uploading new snapshots almost daily on SqueakSource (http://www.squeaksource.com/dgv). You'll need wxSqueak 0.4 for this (although 0.4.1 will work as well). We still have to formally license it, but it'll probably end up as SqueakL+MIT code or something similarly liberal. I am biased, but I think this software is interesting from multiple perspectives: - it contains probably the most 'real' and most complete p2p layer for Squeak to date (using UDP exclusively allowing it to work through most residential broadband routers); - it is a real-world application with a native L&F using wxSqueak (currently the only one?); - even though the application was paid for by the DGV, nothing in the application is specific to the use of it for history. We're already having interest of a group of schools wanting it to disclose teaching materials, for example. Therefore, I think it would be useful if some people outside the scope of the original project, i.e. from the larger Squeak community, would help out to make Kolibri more useful for a larger audience. Specifically, some items that I can think of which are clearly out of scope of the project: - Translation to English. The software is translatable (we send #translated to every string literal, and have some infrastructure in place to use that) but the actual translation work still needs to be done. Probably we want to reverse the current situation where the default strings are Dutch and everything needs to be translated into English; - A Seaside UI as an alternative for wxSqueak - I have some ideas on how this could look, and this would let people play with the software without requiring them to install it or to ask system admins to pierce holes in firewalls; - Testing on Mac and Linux, maybe build installers for these platforms; - A good review of the p2p layer, there are some weaknesses there that ideally need cleaning up (most notably the whole 'presence' stuff which was conceived as a quick hack to support instant messaging but has grown out of proportion, but the file download code could also use some reviewing because performance is under par); - Nifty stuff people come up with - both code hacks and new uses of the software. - And of course, I have a long list of ideas but this post is too long already :-) Anyway, if you're interested, please shout. Oh, and thanks to all the people who have helped out with advice and debugging during the project. Especially Rob Gayvert, who put a lot of hours in wxSqueak support.

4 9

Exupery's new mailing list
by Bryce Kampjes 02 Nov '05

02 Nov '05

Exupery, a bytecode to machine code compiler (a JIT) for Squeak, now has it's own mailing list. Join if you're interested in either helping develop or using Exupery. Exupery is currently 15% faster than VisualWorks for the bytecode benchmark and twice as fast as Squeak's interpreter for the send benchmark. Exupery's next goal is to become seriously useful. So join even if you're interested in using a beta compiler rather than writing a compiler. Some user domain expertise would be helpful. I know very little about graphics, multimedia, Morphic, or crypto which are all things that might benefit from a little more speed. The list is here: http://lists.squeakfoundation.org/cgi-bin/mailman/listinfo/exupery Bryce

5 6

Jump to page: