Squeak-dev February 2005

squeak-dev@lists.squeakfoundation.org

191 participants
430 discussions

The quest for an Usability Team
by Giovanni Corriga 26 Feb '05

26 Feb '05

Hi everyone, One of the areas where Squeak is currently lacking is Usability. Both the language and the environment can offer amazing possibilities in term of productivity, ease of use, tinkering, discovering and learning. But my feeling is that a large part of this potential is hindered and blocked by inconsistant tools, more-complex-than-necessary interfaces, conflicting metaphors. So, as a mean to solve some of Squeak's usability problems, I propose the creation of a Usability Team. The Usability Team should: - analyze the usability of the current Squeak releases, identifying the usability problems and proposing solutions for that problems either as fixes or as suggestions. - as a byproduct of the analysis and solving process, create some Squeak Interface Guidelines which would help the Squeak developers create better tools. - work in close contact with the other teams and the Squeak developers' community. Please note that this is not (yet) a formal team proposal. It is more than a call for everyone who is interested in this topic to come out and join this project. I can act as an interim team leader while the team is forming, but I have no problem in stepping down if someone else wants the job. So, if you are interested in this topic, and want to give your two cents, please reply to this message. As soon as I have feedback from at least a couple of person, I'll ask the coordinators to create a usability mailing list, so that we can start draft a formal team proposal. Giovanni

4 4

Re: The quest for an Usability Team
by Dominique Dutoit 26 Feb '05

26 Feb '05

I agree that a Usability Team is much needed. A few weeks ago, I was wandering in the "Add a new morph" menu and discovered the Phoneme Recognizer which is very interesting for generating mouth positions for animation programs (Blender and Moho in my case). I have fixed and tweaked some parts of it but yet it need much more work to make it usable. As far I have seen, hundred of morphs are just sitting there waiting to be brought to light (or life) -- time to start the "Adopt a Morph" program. On the other end, the risk of such team is to walk on the feet of the UI team and yet, there are some interesting things to explore and some opportunities to move beyond the windows paradigm. First things that come in mind are agents, bricks (no other way to explain the cool stuff found on ZOOSqueakDT [1]) and proximal user interfaces [3] (found today in Scala, Blender, Lightwave ... and in some extends Squeak). So in a sense the job of the Usability Team is not only to clean the Squeak UI or provide a Human Interface Guide but also to inspire. [1] http://www.languagegame.org:8080/zoo/194 [3] http://www.basden.u-net.com/R/proximal.html

2 1

Re: a doubt on squeak license
by Bruce O'Neel 26 Feb '05

26 Feb '05

Hi, (I've added squeak-dev to the To: line so someone else can respond) My unexpert opinion. Open Source has a very specific meaning and by their meaning squeak is not Open Source. (notice the Upper Case) The FSF folks also have a very specific thing in mind and by their definition Squeak is not Free Software. I believe that there are three sections of the license which cause problems with the above groups. - The fonts. You can't produce derivative works from the fonts and you can't charge a fee for the works that contain the fonts. - Clause 5 - "You agree to indemnify and hold Apple harmless from any and all damages..." - Clause 6 - Export Law Assurances. This is basically what https://www.fsf.org/licensing/licenses/license-list.html says: The Squeak license As applied to software, this is not a free software license because it requires all users in whatever country to obey US export control laws. As applied to fonts, it also does not permit modification. In addition, it has a requirement for users to indemnify the developer, which is enough to make many users think twice about using it at all. Now my current take on the above is: - Fonts. We've gotten rid of the origional ones and therefore this doesn't apply anymore. - Clause 5 - Well, you can't sue Apple. - Clause 6 - Unless you plan to have no dealings with the US you probably have to keep this in mind anyway. It's suprising how far the attempts to apply US export laws go, and, one probably doesn't want to try to be a test case. Your call though. There has been some chatter about just ignoring the problem and releasing squeak under the Croquet license (see http://www.opencroquet.org/Croquet_Technologies/license.html) since they have just gone and done that (basically) for Squeak 3.6. Anyway, there has been a lot of debate over the years on the squeak-dev mailing list and there continues to be debate :-) cheers bruce On Fri, Feb 25, 2005 at 06:57:07PM +0530, Vimal Joseph wrote: > Hi Bruce O'Neel, > > i found your address on the download page of the project Squeak. I am > not sure you are the right person to contact for my query. I couldn't > found whom to contact for issues regarding license of squeak. > > In the Squeak's home page and related pages, it is written that squeak > is a open source program. So i downloaded it. But on a search on the net > I found the following page > http://sourceforge.net/support/noss_exception.php?group_id=5506 which > clearly says that squeak is not an open source project. In the fsf.org > license page also i found that squeak is not a Free Software. > > I am now totally confused. I really want to use squeak and spread it > among the teacher community in my place. But if it is a Non Free > software, i cant do that. Please help me clear this confusion. > > Thanks and Regards, > > Vimal Joseph > > > -- > Free as in Freedom <www.gnu.org> > -- edoneel(a)sdf.lonestar.org SDF Public Access UNIX System - http://sdf.lonestar.org

4 3

[OB][FIX] OBDragAndDropFix
by bergel＠iam.unibe.ch 26 Feb '05

26 Feb '05

from preamble: "Change Set: OBDragAndDropFix Date: 26 February 2005 Author: Alexandre Bergel Dragging a method from a method pane over the class or class categories pane trigger a DNU. The reason is that a node is not always returned from OBColumn>>nodeForDroppedMorph: transferMorph event: evt inMorph: listMorph This fix solves this bug, but I feel that the problem is deeper than that..."!

1 0

Harvesting Party Wednesday March 2nd 12:00 GMT
by Ken Causey 26 Feb '05

26 Feb '05

Just a reminder that the next Harvesting Party will take place between 12:00 (Noon) and 16:00 (4 PM) GMT Wednesday March 2nd, 2005. As usual we will meet in #squeak on irc.freenode.net. More information about the #squeak IRC channel can be found at http://people.squeakfoundation.org/article/7.html Information about the Harvesting Party idea can be found at http://people.squeakfoundation.org/article/29.html If you have the time please attend and work toward the future of Squeak. The next seven parties are scheduled for 16:00 (04:00 PM) GMT Sunday March 06, 2005 20:00 (08:00 PM) GMT Thursday March 10, 2005 00:00 (12:00 AM) GMT Tuesday March 15, 2005 04:00 (04:00 AM) GMT Saturday March 19, 2005 08:00 (08:00 AM) GMT Wednesday March 23, 2005 13:00 (01:00 PM) GMT Sunday March 27, 2005 17:00 (05:00 PM) GMT Thursday March 31, 2005

1 0

Re: [Q][balloon2D] is transparency supported?
by Baveco, Hans 26 Feb '05

26 Feb '05

To illustrate my problem, compare the result of the two code fragments, the upper one provided by Boris, the second one illustrating what I have been doing. I would really like to have the result of the first fragment, but not on the Display but on a morph... Hans | bc fs ramp | ramp := (0 to: 360 by: 60) collect: [:h | h / 360.0 -> (Color h: h s: 0.7 v: 0.8 alpha: 0.2)]. fs := GradientFillStyle ramp: ramp. fs origin: (10@10). fs direction: 600@300. fs radial: false. bc := BalloonCanvas on: Display. bc aaLevel: 2. bc drawPolygon: (Array with: 10@10 with: 500@300 with: 300 @500 with: 50 @ 300) fillStyle: fs borderWidth: 2 borderColor: ((Color red) alpha: 0.3). | bc fs ramp form | form := Form extent: 600@600 depth: Display depth. ramp := (0 to: 360 by: 60) collect: [:h | h / 360.0 -> (Color h: h s: 0.7 v: 0.8 alpha: 0.2)]. fs := GradientFillStyle ramp: ramp. fs origin: (10@10). fs direction: 600@300. fs radial: false. bc := BalloonCanvas on: form "Display" . bc aaLevel: 2. bc drawPolygon: (Array with: 10@10 with: 500@300 with: 300 @500 with: 50 @ 300) fillStyle: fs borderWidth: 2 borderColor: ((Color red) alpha: 0.3). ^form asMorph openInWorld On Friday 25 February 2005 7:26 am, Baveco, Hans wrote: > When displaying polygons on a BalloonCanvas, setting the alpha of the > fill-color apparently does not work. > It changes the color though: when alpha approaches 1.0 the original > color shows, when alpha approaches 0.0 the resulting color approaches > black/opaque. Is this a known limitation, or a bug.... (the code looks > like transparency should be handled). The final drawing is done in > primitives, so a bit difficult to check.. I think the problem is more in the way that you displayed the contents of the BalloonCanvas. Was there an intermediate canvas? If you look at (for instance) the fill in the inside of Connectors (which are rendered with Balloon2D) you will see that they can have translucency. -- Ned Konz

2 1

[ANN][ENH] more Refactoring Browser improvements
by Ned Konz 26 Feb '05

26 Feb '05

I have put another Monticello diff file (against the current SqueakMap version of Refactory, which is Refactory-md.3.8.12) on my site at: http://bike-nomad.com/squeak/Refactory-nk.3827.mcd You will have to load Refactory-md.3.8.12 first. Since 3823: added support for opening finder/rewrite tool on environment browsers added hierarchy find/rewrite. made environment browsers use RB menus Allowed changing class selection from Finder and Rewrite tools. Added to flaps registry. Refactored LintDialog to re-use category/class selector as component made ChooserMorph work right for multiple selections improved help text made lint results always display proper selection made environment browsers display class definition if no method selections added support for defining and displaying rationale for Lint rules. The 3823 announcement: The big news here is the addition of the RewriteTool, which is like a pressure washer for code janitors. It is like the FinderTool in that it lets you search for parse tree (rather than text) patterns in your code. But it also lets you specify a replacement pattern. When you say "Replace" it puts up a FileContentsBrowser from which you can view the proposed changes and load selected methods, classes, or the entire set of changes. I have also added the beginnings of a color formatter for the RB, but haven't finished all the colors yet. You can now use the RB formatter for pretty-printing tasks throughout the system. I have also improved its printing of various things, including array literals. Changes from Refactory-md.3.8.12 to Refactory-nk.3823 - added RewriteTool (like FinderTool but you can do search/replace with it) - added lint rule for override of superclass method without calling it. - added lint rule for sending of 'new initialize' and friends - added lint rule for sending a deprecated method to a known global - improved pretty printing (especially for literal arrays, etc.) - added preference for using RB to pretty-print globally - added RB-based color pretty-printing for use in browsers - deprecation fixes: fixed 'super new initialize', 'self beep', etc. - made sure that metaclasses were searched too by lint tool! - misc. bug fixes - changed lint rule for missing super sends to look for #subclassResponsibility in superclass - fixed opening of editor in some cases - sorted class names in RB environment browsers I am presenting it this way so people can get a chance to use it for a bit. -- Ned Konz http://bike-nomad.com/squeak/

2 1

Re: Improving Squeak's Mutlimedia
by Martin Kuball 26 Feb '05

26 Feb '05

Am Sunday 20 February 2005 02:10 schrieb Russell Penney: > > Now one of my biggest concerns is that we should try to develop > EVERYTHING in Squeak with as few plugins as possible. I wrote the > OGG decoder in "pure" Smalltalk so it is possible. Of course by > converting a small part of the decoder to a plugin I sped > everything up 10 fold but that came later ;) > > I think this is important to remove platform dependencies or > dependencies on external libraries. We may end up using them purely > for performance but I think we should have a goal of not using them > in the first part. I don't share your point of view here. Writing codecs in pure smalltalk is interesting and may teach you a lot. But I don't think we have the manpower to do this. I want to have support for mpeg2, mpeg4 (xvid, divx), theora, etc, and windows people may want to use the microsoft codecs. Do you want to implement all this again - if it were possible at all. And I think the speed of a codec is very important. Every cpu cycle counts. I would like to have small Class for each codec that uses the FFIPlugin as interface to the external library. This way everybody with a basic knowledge of his favourite codecs API could integrate it integrate it in no time. We should concentrate on the container stuff and nice guis to manipulate the basic multimeda entities. This is work enough for the next month. Martin

5 7

Problems w/Squeak on Linux
by chris＠workinglinux.com 26 Feb '05

26 Feb '05

I'm running Debian Woody (stable), Squeak VM 3.6.3 and Squeak3.7-5989-full.image, Squeak3.7-5989-basic.image, and Squeak3.6-5429. I downloaded 3.6, the sources and the VM from the "officialUnix" site, and the 3.7 images from the link at squeak.org. Each site provides different information and versions, and squeak.org says nothing about how files need to be organized on disk or how to install, though the "Unix site" does, if one reads the comments at the beginning of the INSTALL script. Perhaps my primary interest in Squeak is collaboration and web services, the educational aspect is also important, but without the others the value of Squeak to me is greatly reduced. I have been trying for two days now to get a web server running, either Kom*/Comanche or Swazoo, without success. I have in the process learned a considerable amount, a good deal of which might be useful should I continue with Squeak. The KomHTTP simply won't install into any of the images listed - in the 3.6 image, I have tried 7.0.2 (after installing NamedProcess, DynamicBinding and KomServices) and version 6.2. The latter won't install, the former will, but it won't run - ModuleAssembly does a 'super initalize', and no superclasses implement it, so I get 'Object does not understand initialize' message. If I comment out the offending statement, ModCore does the same thing. That is as far as I am willing to pursue it - I am new to Squeak and KomHTTP, so either I don't understand what the issue is here, or I am doing something very wrong. I can manage to get Swazoo running in the 3.6 image, however when I point the Galeon (Mozilla-based) browser at it, I get 'page contains no data', if I try a text browser (I've tried 3) they all report an error trying to read the socket. I have Seaside loaded, and I if try to get the /confg page, same thing. I have looked through the both this mailing list and the Seaside list, and can find nothing, Googled, etc. Any help would be greatly appreciated. In both 3.7 images, the Package Loader refuses to install anything - as soon as I try, I get "UndefinedObject(Object) doesNotUnderstand: #<". Furthermore, both the man page for the 'squeak' executable and the output of the -help switch seem to be incorrect/outdated - the options '-lazy' and '-xshm' cause the outpput of '-help' to be displayed. Version 3.7 is certainly more attractive in appearance than version 3.6, and the 'Smalltalk' kernel of Squeak seems to be fairly solid, but after 25 years of evaluating, installing and developing software in many different languages on many different platforms, my instincts and the experiences described above are urging caution about all the other code in Squeak. The Linux kernel would be pretty useless without all the rock-solid utilities, servers, compilers that the GNU project produced, and Squeak-as-a-platform community might wish to consider that. Otherwise, it may be quite a long, long time before Croquet becomes anywhere near viable. Based on what I've read in these newgroups, I wonder if the ability to publish changes so quickly and easily via SqueakMap might not lead to unwelcome interactions between components, or the tempatation to publish without sufficient testing. In the Debian community, package maintainers, usually different people than the developers, do extensive testing before publishing anything - I think this separation of roles might contribute significantly to overall quality and fewer bugs released. This may slow releases down a bit, but the much improved quality has greatly enhanced Debian's reputation and popularity. Besides, whats the rush, anyway? Don't get enough rushing at work? I have noticed that some Squeak packages do run SUnits test as the last step of install, a practice I like a lot and wish were more widespread. I sure hope that my difficulties are just my idiocy, or my bad timing in trying to test the web servers just after a new bug was introduced or something, so please feel free to let me know. Three cheers for quality and care in software development! -- I have yet to see any problem, however complicated, which, when you looked at it in the right way, did not become still more complicated. -- Poul Anderson (1926 - )

8 14

Re: Improving Squeak's Multimedia - additional opinion
by Bob.Cowdery＠CGI-Europe.com 26 Feb '05

26 Feb '05

I would love to see some real audio streaming and DSP capability in Squeak. My interest is Software Radio which requires full duplex audio streamed through DSP routines. I have written such code in 'C', 'C#' and Python using PortAudio and a mixture of libraries and home grown code. The reference to overlap-add FFT framework is something that figures in most radios as well as frequency domain filtering. I don't really know where to start with Squeak however as I have only played with Smalltalk. Would it be easy to do a PortAudio binding or is that the wrong place to start. Would DSP be too slow in ST? A software radio framework in Squeak would be really neat. Anyone willing to give me a starter for 10? Bob *** Confidentiality Notice *** Proprietary/Confidential Information belonging to CGI Group Inc. and its affiliates may be contained in this message. If you are not a recipient indicated or intended in this message (or responsible for delivery of this message to such person), or you think for any reason that this message may have been addressed to you in error, you may not use or copy or deliver this message to anyone else. In such case, you should destroy this message and are asked to notify the sender by reply email.

4 6

← Newer
1
...
4
5
6
7
8
9
10
...
43
Older →

Jump to page: