Squeak-dev April 2001

squeak-dev@lists.squeakfoundation.org

165 participants
614 discussions

Re: Smalltalk & Squeak featured on Slashdot
by Douglas Brebner 28 Apr '01

28 Apr '01

On Thursday 26 April 2001 04:21, Richard A. O'Keefe wrote: > Christopher Sawtell wrote: > >I am not exactly in the flush of youth any more, and for me the pin-head > > size of the default font is difficult to read. > > Bob Arning <arning(a)charm.net> replied: > I'm curious, do you know about how many pixels per inch your > monitor generates? Mine is in the low 80's and, although I have > upgraded my reading glasses several times in the last 4 years, I find > the Squeak font to be much more legible than, e.g. the TV listings in > the newspaper. > > For what it's worth, on the Solaris box where I usually use Squeak, > xdpyinfo tells me > dimensions: 1152x900 pixels (325x254 millimeters) > resolution: 90x90 dots per inch > > If "12 point" type should have a baseline-to-baseline distance of 12 > points, that means 15 pixels (14.944, using "true" points). > > One understands that StrikeFonts are bitmap fonts, so they're not going to > scale easily, but surely 90 dots per inch has been common enough for > several years to warrant a set of fonts that are the right size for it? > Can any of the X fonts be adapted? FWIW I find the default fonts uncomfortable to read on my 100dpi monitor and my vision is perfect.

1 0

How to enable profiling in Mac VM?
by Fleeberz＠aol.com 28 Apr '01

28 Apr '01

If this has been posted, I sure can't find it ;-) I goofed around, as new Squeakers often do, and managed to get Squeak to create enough C sources so that I could compile a working VM. (Whatever I can get from Squeak 3.1 Alpha, update 3910.) Looking through the code, I saw that I could remove a comment and allow profiling... I did that, recompiled, verified that the profiling code was there (have necessary library, too, this is for PPC non-Carbon). Back into Squeak with the new VM, looking around in the Method Browser, I found profile, profile: and similar. Pulling up a browser on profile: showed some code to profile a block; I tried manually executing, in order, the "steps" shown in the code. (did this in a text window with "doIts," since I'm still too wet behind the ears to know how to specify a "block" - don't tell anyone, eh? ;-) Problem? Can't seem to ever create the "Profile.out" file at any point when the profile is dumped and/or the VM quits (Quit from Squeak world menu or VM "Mac menubar" Quit). Anyone with the patience to walk me through the basics of what I should be doing? BTW, the VM sure is a lot slower with the profiling code compiled in! Thank you all, Jean

2 1

Debugger-step into.
by olczyk＠interaccess.com 28 Apr '01

28 Apr '01

Ok. To stop the debugger in a method at a line, I put "self halt." just before that line. But what if I'm tracing through some code and I want to step into a method being passed a message. ( Then step through code in that message.) How can I do this?

3 2

Re: Digital Video in Squeak
by Karl Ramberg 28 Apr '01

28 Apr '01

Jan Bottorff wrote: > > >I would love to have the codec written in Squeak, though. It's more fun > >to > >debug and program that way. (I don't even know how to program C) Then > >run the code > >trough one of those machine-code-converter-classes out there :-) > > Having written commercial high performance video codec's, I can tell you it > generally takes hand optimized assembler to get the kind of performance you > see in commercial products. Languages like C are mostly clueless about > properly utilizing the short vector instructions on modern processors. For > the M-JPEG codec we did, the decompression performance between optimized C > and more optimized Intel MMX assembler (and using special OS interfaces for > video) was about 5:1. The MMX code alone was worth about 3:1. The better OS > interface included using an overlay surface in video memory and writing YUV > color space pixels. This code is java. http://www.sureplayer.org/ It's not fullscreen not 25 or 30 fps but it works with streaming video. Seems like I have some fun hacking ahead of me:-) Karl

1 0

Re: massively parallel smalltalk
by Jecel Assumpcao Jr 28 Apr '01

28 Apr '01

While I really like Linda (and have seen at least two nice implementations in Smalltalk), it does require using a very different design for your programs. Several other alternatives have been developed for massively parallel Smalltalk and I would recomend at least a quick look at: http://cva.stanford.edu/j-machine/cva_j_machine.html Though this Concurrent Smalltalk (there was another project with the same name) had a Lisp-like syntax instead of the keywords we love so much, it really was Smalltalk and was running on 1024 node machines. Hmmm... I see only a brief mention of "distributed objects" in there, though I was pretty sure that was where the notion of "concurrent aggregates" came from. The idea was that if collections were allocated with its members spread out through several nodes, messages such as #do:, #select: and so on would automatically express parallel execution. So many "normal" Smalltalk programs could become highly parallel simply by using concurrent versions of collection objects instead of having to rewrite everything to use tuples. -- Jecel

2 2

[ENH]VMMaker file copying plugin
by Tim Rowledge 28 Apr '01

28 Apr '01

Here is a trivial fix/workaround/what have you for the problem of copying file/dirs without losing the type and other attributes. A copy has also been added to my website, along with an Acorn implementation of the platform support (one measly little function) for example reference. tim -- Tim Rowledge, tim(a)sumeru.stanford.edu, http://sumeru.stanford.edu/tim Let the machine do the dirty work. - Elements of Programming Style

1 0

runing squeak at sun Solaris 8.0
by TECHIN.A.KANG＠mail.sprint.com 28 Apr '01

28 Apr '01

Hi, I can not run squeak from Sun Solaris 8.0. It could be my X-window settings problem, but I have no knowledge of X window. I download unix version for Sun from uc website. gunzip and tar and move .image and .change to squeak/3.0 directory. here is its ls: /toolkits/src/squeak/squeak/3.0> ls -lg total 52876 -rwxr-xr-x 1 other 1432 Mar 23 14:52 inisqueak -rwxr-xr-x 1 other 648 Mar 23 14:52 Profiler.la -rwxr-xr-x 1 other 26364 Mar 23 14:52 Profiler.so -rwxr-xr-x 1 other 1996756 Mar 23 14:52 squeak -rw-r--r-- 1 other 39621 Mar 23 14:52 squeak.map -rw-r--r-- 1 other 885527 Mar 31 12:12 Squeak3.0.changes -rw-r--r-- 1 other 9518760 Mar 31 12:12 Squeak3.0.image -rw-r--r-- 1 other 14542313 Apr 9 16:41 SqueakV3.sources -rwxr-xr-x 1 other 636 Mar 23 14:52 System.la -rwxr-xr-x 1 other 8648 Mar 23 14:52 System.so 1) running squeak from Exceed terminal emulator at WinNT4.0, got the following error message: X Error of failed request: BadMatch (invalid parameter attributes) Major opcode of failed request: 1 (X_CreateWindow) Serial number of failed request: 7 Current serial number in output stream: 14 The display is: /toolkits/src/squeak/squeak/3.0> echo $DISPLAY 172.25.54.29:0.0 2) running squeak from webpage connect to server using VNC, got the following error: Visual class TrueColor is not supported at depth 8 The display is: /toolkits/src/squeak/squeak/3.0> echo $DISPLAY unix:1.0 THANKS, Techin Alex Kang

1 0

massively parallel smalltalk
by Steve Wart 27 Apr '01

27 Apr '01

I knew some people years ago at a company called Myrias (http://www.myrias.com) -- they are now a consulting company -- but they had some really cool stuff. Their computer was a board of 1024 68000s with a special version of Unix to support parallel processing. Had C and Fortran compilers (and an assembler called Myriasm) with constructs like pardo {} instead of do {}. It's not hard to think about the equivalent constructs in Smalltalk (eg. pardo: , parcollect: pardetect:, etc., although the concurrency semantics could be interesting). There's also this: http://www.beowulf.org/ But why stop with Linux boxes? What about routers? Aren't they just really fast computers optimized for sending messages? Isn't that what Smalltalk is all about? Aren't Cisco routers going for about $40 each on eBay right now? :) Not sure why I think squeakers might be interested. I just think it is a cool concept to turn the Internet into a really big Smalltalk image. Physicists and Engineers hate Fortran anyhow :) Cheers, Steve

1 0

Re: [BUG] Smalltlk addToStartUpList: NewClass
by Michael Rueger 27 Apr '01

27 Apr '01

"Daniel V. Oppenheim" wrote: > > Turns out that if you want a new class to get notified on StartUp (such as > for initializing a new MidiPlugin), sending <Smalltalk addToStartUpList: > NewClass> does NOT work. By default the NewClass is put at the end of the > list -- which is what one would expect. BUT, the StarrtUp list iteration is > abandoned either after SecurityManager or AutoStart, so NewClass is never > iterated (this is in: <SystemDictionary send: startUpOrShutDown Hmm, I just tried it with a class and startUp "Smalltalk addToStartUpList: Test" Transcript show: 'Test started'; cr The method has to be implemented on the class side and is spelled "startUp". Sounds silly, but, implementing a method "startup" caused me to scratch my head for a while when I implemented the AutoStart/Launcher stuff 8-\ Do you run Squeak inside a browser or standalone? Michael -- "To improve is to change, to be perfect is to change often." Winston Churchill +------------------------------------------------------------+ | Michael Rueger m.rueger(a)acm.org ++1 (310) 937 7196 | +------------------------------------------------------------+

2 4

Re: Ameliorating/refactoring Squeak
by Florin X Mateoc 27 Apr '01

27 Apr '01

Sorry, I am in conversation with myself again (that's because I like so much the company of intelligent people). Actually, the Squeak Foundation could do something very important here: hire somebody to review and reject/approve the patches/goodies, test (and regression) test them on multiple platforms, then incorporate them into the core. It would still not be as good as many users, but it would force everyone into a user once they are incorporated. This could all happen on multiple branches as well. I am sure this would increase developer satisfaction and productivity, as well as the quality of the Squeak platform in general. It would still not bring us new users though and eventually we need them. Florin mateoc_florin(a)jpmorgan.com on 04/27/2001 01:44:23 PM Please respond to squeak(a)cs.uiuc.edu To: squeak(a)cs.uiuc.edu cc: (bcc: Florin X Mateoc) Subject: Re: Ameliorating/refactoring Squeak Hi, Like many others on this list, I know exactly what you are saying, but I have no idea on how to improve things, because what we need is users (one can only develop in the void for so long) and I don't know how to get them. The problem, as I see it, is that we are not (nor do we have) a community of users (or application developers), who might be interested in a specific functionality or improvement. It would be those users (or application developers working on top) of the main system who would be interested to test and take improvements even when not radiating from the center. Then they would give us feedback and multiple virtuos circles would emerge. Instead, we all are ring 1 developers, only willing to try new things coming from ring 0, so the only chance to get our code tested (and maintainable) is if ring 0 accepts it in the core and pushes it outside. I don't think there is much else ring 0 can do for us. As an example, some time ago I sent some code directly to Dan Ingalls. He was very nice (he always is) and he posted it to the list saying it was interesting and asking people to test it. Although this was coming from Ring -1 Himself, nobody replied. I guess nobody went to the trouble of manually filing it in, trying to understand it, trying to break it and so on. And I am not better than enybody else here. I noticed similar things happening with other people's code and I did not find the time to do testing, debugging. And this is because I am not a user of the system, but a developer and in the little spare time that I have I prefer to do my own hacking. I hope the Squeak Foundation can come up with some answers but I doubt it. I think the only real hope is that Squeak becomes a Disney success story as a kid's learnig/playing tool. Then we'll have all the users we want (need). Florin wuyts(a)iam.unibe.ch on 04/27/2001 10:22:12 AM Please respond to squeak(a)cs.uiuc.edu To: squeak(a)cs.uiuc.edu cc: (bcc: Florin X Mateoc) Subject: Ameliorating/refactoring Squeak Hello, The last days I am thinking about lots of things in Smalltalk, Smalltalk communities, Smalltalk development environments, ... You could call this an existential problem, only that it has to do with Smalltalk, and not so much with my 'inner self'. It's not that I am posing questions whether or not to keep on using Smalltalk. Quite the contrary: I have been using it for 8 or 9 years now, and its conceptual beauty still strikes me regularly. So, the question is not whether or not to keep on using Smalltalk, but what to do with it. Personally, what I typically end up doing is to clean up parts of it. Every time I have to implement something, I sooner or later end up making bigger or smaller refactorings in the System itself (as well as in my code of course). While this pleases me enormously (it makes it far more conceptually esthetic), I never know what to do with these changes afterwards. Sometimes I send them to the list here, but they are not included with the base release. Note that this is no remark whatsoever against Squeak central (they already do a terrific job, for which I am very thankful); I completely understand that they have other goals and do not have the manpower to try and integrate such changes. However, the net result is that these 'ameliorations/refactorings/...' are forgotten over time. After discussing with Stephane Ducasse (my colleague here in Bern), he created the Swiki pages to note these kinds of things. And we are in contact with John Sarkela to add to the Stable Squeak effort. But still... Is there anything that can be done ? Am I just to overly aesthetic ? Did I overlook a special prefix that I can use to send this kinds of things in the mailing list ? Do they have a chance to be included if I make them for the latest Squeak versions ? Etc etc etc -- Roel Wuyts Software Composition Group Roel.Wuyts(a)iam.unibe.ch University of Bern, Switzerland Board member of the European Smalltalk User Group: www.esug.org This communication is for informational purposes only. It is not intended as an offer or solicitation for the purchase or sale of any financial instrument or as an official confirmation of any transaction. All market prices, data and other information are not warranted as to completeness or accuracy and are subject to change without notice. Any comments or statements made herein do not necessarily reflect those of J.P. Morgan Chase & Co., its subsidiaries and affiliates. This communication is for informational purposes only. It is not intended as an offer or solicitation for the purchase or sale of any financial instrument or as an official confirmation of any transaction. All market prices, data and other information are not warranted as to completeness or accuracy and are subject to change without notice. Any comments or statements made herein do not necessarily reflect those of J.P. Morgan Chase & Co., its subsidiaries and affiliates.

1 0

← Newer
1
2
3
4
5
6
7
8
9
...
62
Older →

Jump to page: