Squeak-dev May 2001

squeak-dev@lists.squeakfoundation.org

143 participants
597 discussions

RE: Squeak textbox and events.
by Wiebe Baron 03 Jun '01

03 Jun '01

I'm also not very keen on inlined error messages. I have the habit of pressing Alt-s twice in rapid succession, the first time the red borders disappears but I only notice the second time when the whole edit box flashes (or perhaps I just have a need to double check anyway). This extra keystroke seems to pay off against the frustration of debugging code I've forgotten to save but you'll understand that things go badly wrong when the first Alt-s generates an inlined error message. Besides, having text inserted in your code can feel intrusive and having to press delete before you can edit your code doesn't seem make much cognitive sense. It's not that it's hard work, it's just the wrong kind of work! There's also a bit of inconsistency in that using an undefined variable in your code brings up a message box rather then an inlined message. Completely logical, but inconsistent never-the-less. Using inlined error messages seems to be the simplest solution to the problem and as a proponent of Occam's razor, I can live with it. Somewhere down the line we should look at better ways though, because as Einstein warns: "Everything should be made as simple as possible, but not simpler." Cheers, Wiebe > -----Oorspronkelijk bericht----- > Van: Jim Benson [mailto:jb@speed.net] > Verzonden: 13 May 2001 02:35 > Aan: squeak(a)cs.uiuc.edu > Onderwerp: Re: Squeak textbox and events. > I'll assume that you mean you have gotten an error message while > compiling a > method. Two things here. First, notice that the error is highlighted > (selected). In order to delete the error message, a one kestroke > sequence is > required. Second, in order to recompile and save the code, you > must fix your > error, which will require you to edit the text. If you are going > to edit the > text anyway, an extra keystroke does not seem to be a major penalty.

3 3

RE: Community and Artifact Define One Another
by Stephen Pair 01 Jun '01

01 Jun '01

I was not attempting to rewrite history (indeed, attempts at rewriting history really boil my blood...so please be careful when making such accusations). I guess it comes down to your interpretation of what exactly constitutes a fork. My definition of a fork is any line of development that occurs in parallel to any other line of development. When you download Squeak and begin writing code, you have just forked. (this very technical view of forking is perhaps influenced by my experience in developing source code management systems) The real question is: "how often are the forks synchronized?" If Je77 and Bolot's answer to this question for ComSwiki was never, I would consider that unfortunate. However, ComSwiki does get syncronized periodically...and the more often it is synchronized, the less work involved. A nice service that someone could provide is to maintain a web page that listed the major forks (lines of development) and for each fork, maintain a "synchronization index" that was calculated based on the frequency that the fork was updated to load and run correctly in the current Squeak release (forks could also have SIs relative to other forks as well). This information would give people an idea of how current a particular line of development stays and perhaps influence their decision whether or not to follow it (of course, past results are not necessarily an indication of future performance). So, regardless of what the StSq people may say about forks, according to my definition, they forked long ago. And, I hope they sync up soon and often. - Stephen > -----Original Message----- > From: Andreas Kuckartz [mailto:a.kuckartz@dokom.net] > Sent: Thursday, May 31, 2001 3:12 PM > To: squeak(a)cs.uiuc.edu > Subject: Re: Community and Artifact Define One Another > > > Stephen Pair wrote: > > > I think people seem to forget that for a long time now, we've had a fork > > that has quite successfully co-existed with the rest of the Squeak > > community. I think it has also been mutually beneficial. It's called > > ComSwiki and can be downloaded at: > > > > http://minnow.cc.gatech.edu/swiki > > Sorry, but this is wrong - please do not try to rewrite history to justify > forks. For a long time now the up-to-date ComSwiki-sources have been made > available by the authors and they could be filed in to an unforked image. > Maybe the source is not (yet) available for the most recent version for > whatever reason but that was not the case before. > > Andreas > > BTW: The StSq-people have made a decision that they do *not* want to be > forkers and therefore all debates about forks have become much > less relevant > as far as I am concerned. > >

2 2

Re: newbie question about morphs
by Rosemary Michelle Simpson 31 May '01

31 May '01

Thanks to everyone who responded to my question about the use Morphic with the kind of data model I'm evolving. I must confess that its a great relief since Morphic is so utterly enchanting; I had been very sad at the thought of being shut out from its growing capabilities. Thanks again, R.

2 1

Re: Community and Artifact Define One Another
by Mark Guzdial 31 May '01

31 May '01

I think that Stephen is referring to forking the community, not the artifact. The pws(a)cc.gatech.edu mailing list is live and well (though not NEARLY as busy as the Squeak list! :-). But you're right -- Comanche and Swiki run just fine on normal Squeak. Mark At 9:11 PM +0200 5/31/01, Andreas Kuckartz wrote: >Stephen Pair wrote: > > > I think people seem to forget that for a long time now, we've had a fork > > that has quite successfully co-existed with the rest of the Squeak > > community. I think it has also been mutually beneficial. It's called > > ComSwiki and can be downloaded at: > > > > http://minnow.cc.gatech.edu/swiki > >Sorry, but this is wrong - please do not try to rewrite history to justify >forks. For a long time now the up-to-date ComSwiki-sources have been made >available by the authors and they could be filed in to an unforked image. >Maybe the source is not (yet) available for the most recent version for >whatever reason but that was not the case before. > >Andreas > >BTW: The StSq-people have made a decision that they do *not* want to be >forkers and therefore all debates about forks have become much less relevant >as far as I am concerned. -------------------------- Mark Guzdial : Georgia Tech : College of Computing : Atlanta, GA 30332-0280 Associate Professor - Learning Sciences & Technologies. Collaborative Software Lab - http://coweb.cc.gatech.edu/csl/ (404) 894-5618 : Fax (404) 894-0673 : guzdial(a)cc.gatech.edu http://www.cc.gatech.edu/gvu/people/Faculty/Mark.Guzdial.html

1 0

Community and Artifact Define One Another (Was: SqF purpose: supporting the Squeak community?)
by Andrew C. Greenberg 31 May '01

31 May '01

>> That is, we are both looking inwards to discussing purposes and >> projects >> to support "Squeak the artifact" as opposed to looking outwards to >> purposes and projects to support "Squeak the community". The two are >> naturally related, but the choice of perspective might have a profound >> influence on how we all think about the Squeak Foundation, as well as >> how the Squeak Foundation carries out its operations and sets its >> priorities. > > I would most certianly agree. I think it is always better, perhaps > infinitely better, to support the community rather than "just" the > artifact that the community formed around. The latter will come from > the former. Excuse me, gentlemen, this is mere sophistry. A seasoned advocate can trivially make the contrary argument with equally compelling rhetorical force. But I'll spare you the demonstration, because to do so would be mere demagoguery -- indeed, mere pabulum. The question is not "whether the artifact drives the community or the community drives the artifact?" Even to attempt an answer (using an exclusive or) is losing, for this is a fundamental principle of open source communities and, to use your terms, their "artifacts:" COMMUNITY AND ARTIFACT DEFINE ONE ANOTHER. Put another way, the community and artifact are not only "naturally related," but are inherently intertwined. At least for a viable community. In an open source world, the artifact joins and defines the community. Just as surely, the community defines the artifact. This is because of what we are -- of what is an open source community: a community comprising those who are concerned with a program they program. It is because the word "program" is both NOUN and VERB. Projects and communities that enjoyed these synergies prospered. Those projects and communities who have not became irrelevant. It is as grievous an error to rationalize a dangerously conservative posture regarding the artifact on the theory that the artifact --as is-- defines the community as it is to rationalize an unconcionably unnecessary fork on the theory that the artifact is irrelevant if it is not changed to serve some special interest identified for particular purposes as "the community." If we try to define the community with words approximating a functional specification at any level, we have already lost. As such, it leads to hopeless and divisive advocacy -- and eats at the core of a community. I am not saying you are right or wrong. I am saying that THIS argument is pointless, possibly harmful, and disserves community and artifact equally. Let us define our community by our dynamic and limitless good works. It is time to write much less political prose and write much more code and documentation. All of us. Let our works (meaning both the efforts and the products), not our words, define us.

4 6

Re: The Joys of the Script Manager
by Joseph Pelrine 31 May '01

31 May '01

aw, shucks. I'm turning red here ;-) If you have the latest version, you'll note that you can also maintain multiple managers on multiple data files. Just export the scripts into another file name than 'scripts'. That's what I do to keep big groups of scripts separate, and to transfer them from image to image. Joseph At 18:02 31.05.2001 , you wrote: >I just wanted to say thanks for the Script Manager to Joseph Pelrine >whoever else might've worked on it. > >I'm doing work in an area that is usually would be done in little awk >or perl scripts. Parse file, compute some minor stats, and generate >an XML file with this info. The format is full of exceptions, and it's >always a battle finding the new one screwing up your output. >Consequently, >I've been writing a lot of little scripts. Nothing I want to formalize >in a class, just transient, often changed code. > >The Script Manager allows me to dump it somewhere, without giving me >that >not-so-fresh-and-clean feeling of putting it in some "MiscScripts" class >that just has methods hanging off of the class side. > >Aaron -- - Joseph Pelrine [ | ] Daedalos Consulting Email: jpelrine(a)acm.org Web: www.daedalos.com/~j_pelrine Smalltalk - scene and not herd!

1 0

Re: [EXPERIMENTAL!][GOODIE] HintingBrowser
by Stephane Ducasse 31 May '01

31 May '01

on 5/27/01 11:25 PM, danielv(a)netvision.net.il at danielv(a)netvision.net.il wrote: > If you're planning on studying some code soon, I'd love to know if this > helps you any, or any other comments. > > from preamble: > > "Change Set: HintingBrowser > Date: 28 May 2001 > Author: Daniel Vainsencher > > Experimental!! > Adds hints about the contents of methods in the Browser. > A - accessor > S - just return self > C - just return some constant > P - a primitive > a number - regular code, higher number mean more complicated code. > > This very alpha code, much hackery, not meant for inclusion in update > stream... > Having gotten this clear, this code says various things - > 1. We can get some information more easily than entering the method, > especially when browsing code. The info found by this version is only > the trivially available, we could do more. > 2. Cute icons might be better than letters > 3. A smarter PluggableList would be able to get lists of things other > than texts, and display each appropriately - (not like in this code by a > fixed mapping) maybe by asking the list items, maybe by being configured > with the right morph Factory by the application"! > What could be interesting to add is: - abstract methods - extends (overide + call super) - override without super call - delegate What is also a problem with common browsers is that they only show the delta that the class is defining. In fact we would like to see the class as the local class and its superclasses. Inheritance promotes incremental definition but it would be good to have a synthetic view. One problem with this approach is clearly if you take initialize you will have to find a way to present all the initialize methods defined in the superclasses. Stef

3 3

RE: Squeakland Evolution project thingy
by Withers, Robert 31 May '01

31 May '01

Sam, Sorry about that. You're absolutely right; we shouldn't go into the faith issues. I meant to question technical merits of evolution, as opposed to other mechanisms describing organic system formation, with the goal of doing simulations in Squeak. I am intrigued by a system capable of negotiating distributed meta-services with feedback mechanisms to drive configuration and activity. John Holland developed the original "Selectrons" system that used a blackboard and a bucket brigade feedback mechanism. He was simulating thought processes. Rob PS. is there an essay on comparative religion from an Islamic perspective about? That would be interesting! -----Original Message----- From: Sam Adams [mailto:ssadams@us.ibm.com] Sent: Thursday, May 31, 2001 9:06 AM To: squeak(a)cs.uiuc.edu Subject: Re: Squeakland Evolution project thingy After carefully reading the many recent posts on this topic, I would like to strongly suggest that those on the list who want to discuss matters of faith, and that includes the faith in evolution, chaos, complexity, or God, please take them to another forum. As I understand it, the Squeak community is one based on a common interest in using and advancing certain software technologies, not personal ideologies (scientific, religious, or otherwise). It is one thing to discuss astrophysical simulation (at any scale) or generative variation/selection (evolutionary) algorithms, but this "we, the enlightened of science" vs "they, the endarkened of faith" kind of discussion presumes a uniformity of belief on this list and presses a defacto statement of faith on its membership. I do not think any of us either believe this uniformity exists or want this result, so please have some consideration for those of your Squeak "brethren"(and "sistren" !) who may not share your ideology or wish to discuss it on this list. Do we really want this already very full list to be cluttered with such things? If you quote Dawkins or Kaufman (both of which I read), not for technical reference but in defense of your ideology, how would you react if others quote Jesus, Paul, or selections from Ecclesiastes (which I also read, and believe) with equal force in defence of theirs? Innapropriate for the list? "Go and do likewise" - Jesus As for the *content* of applications that some member of the list *uses* Squeak to create, I do not believe the list should stand in judgement as an open forum. What anyone posts on BobsSuperSwiki, like any swiki, is their own responsibility, unless of course the swiki master sets and enforces limits. This should be true for the 8 year old in an open school, the members of SqC or anyone else. How would you react to a long discussion on the list of the literary and spiritual merits of an interactive essay on the book of Philippians or, say, an interactive essay on comparative religion from an Islamic perspective? But would that essay be allowed on the SuperSwiki as an example use of Squeak technology? My point is this. If we succeed in creating a truly wonderful environment for dynamic and interactive media, and succeed in getting many in the world to use it, they will create there own content without our permission or ideological agreement. As a technologist, that is what I expect and that's good enough for me. As a christian, I choose to take my stand on matters of faith in other forums. Go and do likewise. Sam S. Adams, IBM Distinguished Engineer, IBM Research tie line 444-0736, outside 919-254-0736, email: ssadams(a)us.ibm.com <<Hebrews 11:6, Proverbs 3:5-6, Romans 1:16-17, I Corinthians 1:10>>

1 0

Re: Squeakland Evolution project thingy
by Brent Pinkney 31 May '01

31 May '01

Keith mate, We are here to cut code. You would be a lot happier if you first perused the Skeptical Enquirer's Field Guide to Critical Thinking. http://www.csicop.org/si/9012/critical-thinking.html Then you would be happy that any rational discipline cannot even discuss the validity or compare religious beliefs, art, etc because is doesn't meet any, let along all the conditions. Valid, not sure, but arguable no. Further, and more to the point, there is no one interpretation of the word "know" or any of the other words that bother you. All becomes clear in the context of a rational argument. Brent programmer, philosopher, labourer, martyr ----------------------------------------------------------------- Visit our Internet site at http://www.reuters.com Any views expressed in this message are those of the individual sender, except where the sender specifically states them to be the views of Reuters Ltd.

1 0

Graphing, Concatenate strings, and misc questions
by J Scott Jaderholm 31 May '01

31 May '01

Hi, I need to write some software that listens to the serial port, inputs a comma delimited list of values and then graphs them. I've looked at the SerialPort code a bit so say once i read it in I have '63,65,3,34,35,88,54cr32,33,34,43,22,23,22' any ideas on how I'd change that into #(#(63 65 3 34 35 88 54) #(32 33 34 43 22 23 22)) or where to look for hints or any code that does somethign similiar? I wonder if there is something more advanced for graphing data than GraphMorph? Also, how can I create a new instance of GraphMorph but give it a name or some way to reference it? I can do New Morph > GraphMorph and get one but then I can't change it's data or whatnot. From browsing the class source I don't see any way to create a new one and have it display on the screen via code. I was also unable to find a method that would concatenate two strings, any ideas? Thanks for any help. Sincerely, jsj -- the sky is tired of being blue

3 2

Jump to page: