Squeak-dev June 2001

squeak-dev@lists.squeakfoundation.org

122 participants
362 discussions

RE: [BUG][FIX]Unstable Squeak Won't Load Modules on Macintosh
by Wiebe Baron 03 Jun '01

03 Jun '01

Joseph, StSq was released at this point "In the interest of getting this proto release into as many hands and minds as possible". Hands and minds means comments, complaints and fixes. Your comments on the StSq list: "We don't need more traffic. This list consists of two groups - the StSq developers (a paltry few), and the lurkers. Let's keep it this way" seem to suggest that input from anyone, including other developers, is not welcome. I don't believe that this was the intention of the Squeak World Tour. Many of the concerns expressed on this list are not unfounded. If StSq becomes the pink-plane basis of *our* Squeak, its functioning will impact us all. We're not all just looking for a free Visual Works to base our work on. Many of us share SqC's vision of the "personal computing environment". If the *real* computer revolution is going to happen with Squeak, the pink-plane is going to have to follow the blue-plane into the clouds. Now is a good time to comment. Yesterday perhaps, would have been better. Cheers, Wiebe > -----Oorspronkelijk bericht----- > Van: Andreas Kuckartz [mailto:a.kuckartz@dokom.net] > Verzonden: 03 June 2001 12:30 > Aan: squeak(a)cs.uiuc.edu > Onderwerp: Re: [BUG][FIX]Unstable Squeak Won't Load Modules on Macintosh > > > > Lots of complaints here. People want a preview of a work-in-progess, and > > then complain that not everything works. > > If the StSq-people are reading bug reports (and fixes) and > suggestions from > people who spent some time with StSq as complaints then it really might be > better for others not to spend time with it at all. > > Is that what you are suggesting? > > I think open source projects which are not perceptive for users reactions > are about as bad as hidden source projects. > > Andreas > > >

1 0

Re: [ENH] BrowseLocal-nk (3.1 CS4055 or later ONLY)
by Ned Konz 03 Jun '01

03 Jun '01

On Friday 01 June 2001 09:57, I wrote: > Recent changes broke the browse local functionality because the menu got changed. I will have a better (smaller) version of this change set soon. The one I posted steps on some methods it probably doesn't need to. Sorry. -- Ned Konz currently: Stanwood, WA email: ned(a)bike-nomad.com homepage: http://bike-nomad.com

1 0

Re: [FIX] ProtocolBrowserSuperclassFix-nk
by Ned Konz 03 Jun '01

03 Jun '01

On Friday 01 June 2001 09:52, I wrote: > This fixes a problem with the ProtocolBrowser where attempting to view methods in superclasses causes a walkback."! Please ignore the change set. The ProtocolBrowser changes were fighting some of my local changes. I made the mistake of not checking a stock image... -- Ned Konz currently: Stanwood, WA email: ned(a)bike-nomad.com homepage: http://bike-nomad.com

1 0

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: More advanced Graphing
by Diego Gomez Deck 03 Jun '01

03 Jun '01

If you want, I can send my PlotMorph... it looks like the atachs... TIA, Diego Gomez Deck >Hi, > >Is there a way to graph multiple sets of data on one GraphMorph? > >Like right now it's just a bar graph, only handling one array of data. >I'd like to be able to graph multiple sets of data as lines and have >them appear in different colors, and later other more advanced things. > >Does GraphMorph handle this? Has anyone extended it to? Is there a >more advanced graphing facility? > >Sincerely, >jsj > >P.S. Thanks Bert & Mark for your wonderful answers to my last >questions.

4 4

Re: [EXPERIMENTAL!][ENH] HintingBrowser 0.2
by danielv＠netvision.net.il 02 Jun '01

02 Jun '01

Roel Wuyts <roel.wuyts(a)iam.unibe.ch> wrote: > On 02/06/01 17:27, "danielv(a)netvision.net.il" <danielv(a)netvision.net.il> > wrote: > > Stephane Ducasse <ducasse(a)iam.unibe.ch> wrote: > >> What could be interesting to add is: > >> - abstract methods > > Done. Wasn't sure whether methods that call #shouldNotImplement also > > count as abstract. What do you think? > Well, no, as the intention is not that subclasses should override this to > implement the behavior, quite the contrary. Hmm, right. It's not delaying implementation, but refusing it. >ShouldNotImplement is used for > 'cancellation': 'removing' unwanted behavior inherited from superclasses. > Maybe this is another category. I think it should be, but it's not so important. > >> - extends (overide + call super) [super calls on different names] > Yes. Methods with a selector foo: that in their implementation call do a > super call to another method then foo: are considered very bad and could > lead to hard-to-track bugs. I know - the question is whether this is worth showing. As in the previous case, I'm not sure what I want to do with the relatively rare cases. > I normally call this is a 'bad super send', and > show it differently from regular super sends. and > Yes, it quickly does, does it ? I never found a good intuitive way of doing > it... Where do show these sort of things? Are you implying that SOUL got a body? how is that going? > > BTW, you have better ideas about how to display these things? I think > > the single-letter-flag idea is past it's usefulness ;-) > Icons are also not very practical. Maybe letters or icons in > combinations with tooltips ? That way you can hover over an icon/flag and > get more info ? I was thinking of icons too, but I like the tooltip idea. The problem with icons is you need pretty icons (a problem by itself ;-) that designate some pretty abstract ideas... Daniel

1 0

[EXPERIMENTAL!][ENH] HintingBrowser 0.2
by danielv＠netvision.net.il 02 Jun '01

02 Jun '01

Stephane Ducasse <ducasse(a)iam.unibe.ch> wrote: > What could be interesting to add is: > - abstract methods Done. Wasn't sure whether methods that call #shouldNotImplement also count as abstract. What do you think? > - extends (overide + call super) It now says SU if the method calls upon super. This raises two issues - 1. For the other indications, if you get them you don't need to know the methods complexity. Should one get things like SU3 for a complex, super calling method? 2. SU isn't exactly like extends, because some methods call super differentMethod. Is this an important distinction? > - override without super call > - delegate define delegate... (same parameters? same name? or just only a single send? does it have to return the result of the delegation?) BTW, you have better ideas about how to display these things? I think the single-letter-flag idea is past it's usefulness ;-) > Stef Daniel from preamble: "Change Set: HintingBrowser 0.2 Date: 2 June 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. Now also: ABS - the method is a subclassResponsibility SU - the method calls upon the super class This is very alpha code, much hackery, not meant for inclusion in update stream... Having gotten this clear, this code can say 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"!

2 1

Stable squeak isn't (on mac, maybe)
by jack 02 Jun '01

02 Jun '01

hi, i have had consistant problems getting squeak to run from after decompacting. i have a suspicion that it might have to do with stuffit expander, can't be sure about that. however, if you have a high speed connection and/or some patience, this workaround might get your squeak working right. i had os x running, i had fixed my squeak and was running it from a carbon vm. since then my os x crashed my powerbook firewire *hard*. now, when i open os x it doesn't show any of its own app icons or folders properly, they are all blank (if anyone knows how to fix this get back to me at jeaton @ trentu.ca or on this list)... it happens despite any number of reformats and reinstalls. anyway, i reformatted and am back to os 9.0.4 but this might work for you: go to the platform independent section of your favorite squeak ftp site, and get the package that has the changes and image, and get the one that has the sources... go to www.versiontracker.com or some similar site and get macgzip and zipit. uncompact the files you download with those programs, and then change the file types and creators with res edit or its mac os x equivalent. creator for all of these must be: FAST file type for the image is: STim file type for the sources and changes is: STch another thing is to give the vm more memory than it asks for... mine was set too low initially. at worst you might waste a half an hour or so, at best you'll get a stable squeak. it worked for me. jack

2 2

Solutions Conference Confirmation Monday, April
by Les Tyrrell 02 Jun '01

02 Jun '01

11, 2001 Stephens Convention Center 5555 N River Road Rosemont, IL 60018 USA You are registered for the following: Leslie A Tyrrell UIUC 234 Apple Ree Dr Urbana IL 61802-7371 Fee Description: Fee: Payments: Balance: Full Conf Reg w/2 Tutorial by 3/1/2001 $1,247.

2 1

building a simple app with "active" gui in squeak?
by jack 02 Jun '01

02 Jun '01

hi, i am a 3rd year computer studies student at university in canada. i am in a summer course right now doing a 3 person group project to create a prototype for degree planning and academic advice software. i have attached a small .jpg that shows what we are trying to achieve (it shows the project in a web browser, but that is not essential). i have been reading about squeak and playing with the demo projects a little bit for maybe 2 weeks, and am now eager to try and use it more powerfully as an idea processor but don't know where to begin. someone suggested smalltalk for wrapping my head around what oo really is, and i chose squeak. i have just spent the day with self 4.1 (mac) because i read that would help with morphic but i didn't make much progress there either. the rest of my group wants to do this project in visual basic, and i wanted to try and get it done in squeak before the end of the weekend, so that we would have to resort to vb at all... and i could get some good experience with squeak. here is a textual description of our functional requirements [see the diagram for more clarity]: the student shall be presented with a grid. there shall be four columns, one to represent each year of an undergraduate degree. there shall be 5- rows, and each box in the grid will represent a course. for example: 100, 122, 205, etc. there are two features that comprise the desired core functionality: i) active gui: the initial state of the grid will be as follows: courses for which no prerequisites are required shall be yellow. courses for which the student lacks the appropriate prerequisite shall be red. once the student clicks on any given course, signifying that he/she has taken that course, or would like to see the effects of taking that course it shall turn green. upon any course turning green, indicating successful completion of that course and therefor its eligibility as a prerequisite, any courses for which that newly green course alone or another combination of previous green courses with that new green course shall turn from red to yellow to indicate their availability. this is mostly an exploratory way for the student to visualize the relationships between courses in a single major. after indicating all of the courses he/she has completed and pressing the "process" button the student shall receive ii) suggestion screen the software must be able to store the courses that the student has taken and sort them against a list of the required courses for their major. the suggestion screen shall provide the student with a written indication of the courses he/she should take to complete the degree. ---------------------------- ideas so far: i think that the active gui will have something to do with embedding morphs inside of each other, the buttons will go on top of rows or columns? i was in self all day, so i don't know how similar the squeak morphic is yet. i don't know how to implement that [to me] complex behaviour where the other buttons would all change the appropriate color in any language.... not a clue. we were thinking that the gui state could be translated somehow into an array of courses taken and then sorted against an array of courses required or something like that, or maybe using an orderedcontainer or bag? then some simple text processing could transform that container into english sentences? i would greatly appreciate any and all suggestions... i have been reading some of the broad kind of context level papers on smalltalk and squeak and i like the vision and goals much more than visual basic... besides i don't even have a pc so it would be hard for me to do any meaningful [all night] work on that platform. i think that these days it is irresponsible to develop in any language that is not ported to at least 3 hardware platforms anyway. lastly, please cc your responses to me directly at jeaton(a)trentu.ca, as for some reason the list server won't send me the list traffic. *big* thanks in advance, jack eaton trent university canada

2 1

Jump to page: