Squeak-dev February 2000

squeak-dev@lists.squeakfoundation.org

163 participants
655 discussions

who's who on swiki
by Steven Elkins 28 Feb '00

28 Feb '00

The Who's Who page on the swiki http://minnow.cc.gatech.edu/squeak/8 still exists, but no longer has a list of Squeak people. Is the list located elsewhere? Thanks, Steve

1 0

Re: Visual Basic? HyperCard? An early morning ramble.
by Bob Arning 28 Feb '00

28 Feb '00

On Mon, 28 Feb 2000 07:07:56 -0500 "Jay Carlson" <nop(a)nop.com> wrote: >My favorite is when you select the wrong level of morph to manipulate---like >the text, rather than the scrolling view that surrounds it. The first time >I did this, I said "wow! look at all the generality in the system!". The >second time, it made me wonder "is there some way of automatically locking >substructure that an app isn't dealing with?". The third time I just got >annoyed. I'm still trying to get my head around Paragraph, so these kinds >of changes in Morphic are beyond me for the moment. Jay, I don't think this change would be particularly difficult to implement. What's tricky is knowing which morphs the user might wish to have locked out of all the morphs that exist in the world at a given time. Locking everything in a bowser, e.g., is not hard, but if the user wants some parts of the browser locked and others not, the devil is in defining which is which. Do you have further thoughts/examples of what those might be in your case? Cheers, Bob

1 0

Squeak license & CVSTProject
by G�ran Hultgren 28 Feb '00

28 Feb '00

Hi all! Just found this page: http://ntua.linuxberg.com/conhtml/dev_languages_smalltalk.html It says (falsely I presume) that Squeak is GPLed. Perhaps someone should fix it! Anyway, I wonder if someone could enlighten me on the license issues (if there are any) regarding Squeak: If I were to develop something that would be useful to bundle in the Squeak image - what license can I use? GPL? LGPL? Artistic? Another small announcement: The two miniprojects "SqueakCVS" and "CVSTProject" have merged (it sounds like we are BIG, doesn't it? :-) which now makes us around five hundr... err, five developers. But hey - on the bright side: This weekend I checked out a file from CVS using pure Squeak code. Yiha. Join us in our crusade for "marrying CVS and Smalltalk" (including Squeak, VW etc.): http://cvstproj.sourceforge.net ...or at our swiki: http://www.bluefish.se:8080/sqcvs.1 regards, G�ran PS. Yes, I know that I - du to my own silly pledge :-) - now have to come up with a Class Comment (Go, go Dwight, you lead the race) - I just happen to have one at home waiting to be mailed... :-) DS ===== G�ran Hultgren, goran.hultgren(a)bluefish.se icq#:6136722, GSM: +46 709 472152, http://www.bluefish.se "First they ignore you. Then they laugh at you. Then they fight you. Then you win." -- Gandhi __________________________________________________ Do You Yahoo!? Talk to your friends online with Yahoo! Messenger. http://im.yahoo.com

2 1

Re: Visual Basic? HyperCard? An early morning ramble.
by Alan Kay 28 Feb '00

28 Feb '00

Yep! Cheers, Alan ------ At 4:07 AM -0800 2/28/00, Jay Carlson wrote: >I think all this talk about Visual Basic stirs up unnecessary loathing. >Lemme summarize and rephrase some of the messages I've made it through: > >It Would Be Nice If Squeak had the same level of UI construction, >extensibility, and friendliness to users at different sophistication levels, >as HyperCard. > >HyperCard has a collection of different control types, roughly mapped to >what's in the Mac toolbox. You can build a UI that (more or less) looks >like what users normally see on some platform. Just drag the components off >the palette. > >HyperCard explictly has multiple access levels. At the browsing level, I >can just interact with an app without worrying about messing it up. At the >scripting level, I can write code associated with everything I see. In >between, there are some construction activities that don't require me to >know as much about the internals of the system. (I think that any Squeak >level system would have at least one level beyond what HC regards as >"scripting"; there's more detail available in Squeak than in HC.) > >HyperCard has Undo. That's important. No, you can't reverse arbitrary >computations, but when you do something disasterous in Morphic it's often >not clear what to do to get back. For example, I keep moving subpanes out >of their containers, and they never seem to want to fit back properly. > >My favorite is when you select the wrong level of morph to manipulate---like >the text, rather than the scrolling view that surrounds it. The first time >I did this, I said "wow! look at all the generality in the system!". The >second time, it made me wonder "is there some way of automatically locking >substructure that an app isn't dealing with?". The third time I just got >annoyed. I'm still trying to get my head around Paragraph, so these kinds >of changes in Morphic are beyond me for the moment. > >I don't need to convince Squeak people that some aspects of a HyperCard-y >model are the right thing. From what I can tell, Squeak *is* moving in >this general direction. OK, maybe consensus on a requirements set could >better congeal a roadmap on how to get there (and thus suck in more people >to work on it, which would make me happier). But construction is at the >core of the dynabook; I know we'll be working on it, even if a >deployable-app capability doesn't fall out as soon as everyone would like. > >Maybe that's a key difference between HC and VB. Both make it easy to build >UIs, but VB is concerned, above all else, with being able to deploy apps. > >Jay

1 0

Re: How Do You Do Business Apps? (Morphic Design Philosophy)
by Doug Way 28 Feb '00

28 Feb '00

On Sun, 27 Feb 2000, Bob Arning wrote: > On Mon, 28 Feb 2000 "Mike Thomas" <miketh(a)ptmnet.com> wrote: > > > >How do I find what I have added/changed? Sometimes after an extended > >session you forget and save the image. > > The top three items in the useful expressions list are: > > Smalltalk recover: 5000. > ChangeList browseRecentLog. > ChangeList browseRecent: 2000. > > These will tell you just about anything you want to know about what > you have been doing. Also, probably the most organized way of looking at classes/methods that you've changed is to use a Change Sorter... select "open..." then "simple change sorter" from the World Menu. By default, all of your changes will appear in a "changeset" called something like Unnamed1. You can then save these changes to a separate file, etc. Also see the other items in the "changes..." submenu. - Doug Way EAI/Transom Technologies, Ann Arbor, MI http://www.transom.com dway(a)mat.net, @eai.com

1 0

[ENH] Scamper image maps/buttons, take 2
by Bolot Kerimbaev 28 Feb '00

28 Feb '00

http://fce.cc.gatech.edu/~bolot/squeak/scamper/Scamper-imagemaps.28Feb0032.… (also available Scamper-imagemaps.28Feb0032.cs.gz Scamper-imagemaps.28Feb0032.zip) (improvement over the previous implementation: doesn't introduce extraneous classes DownloadingImageMapMorph, ImageMapMorph, ImageMapLinkMorph, etc. Instead, makes use of Morphic's event handlers to create links) Known issue: uses a hack to make the links clickable (in HtmlArea>>linkMorphForMap:andBrowser:), for some reason the default (transparent links) didn't always work (that's a mystery: it worked sometimes, for some links...) Change Set: Scamper-imagemaps Date: 27 February 2000 Author: Bolot Kerimbaev Enhances Scamper to support: - client side image maps - image buttons (form inputs)

1 0

Re: Polling
by Bob Arning 28 Feb '00

28 Feb '00

On Sun, 27 Feb 2000 22:16:21 -0500 "Charles-A. Rovira" <crovira(a)wt.net> wrote: >Want to run that one by me again? > >You surely don't have a loop polling the mouse to see what happened... > >Most GUIs generate separate events for mouse down and mouse up so you can bring the >mouse down over a widget and not trigger by moving the mouse off and then releasing >it so something else get the mouse up. They also generate mouse mode and get lose >focus events... Charles-A., There are actually levels within levels here, some polling and some events: 1. How the underlying OS communicates with the VM varies platrorm by platform. 2. How the VM communicates with the image is (in the case of Morphic) a polling loop (see PasteUpMorph>>doOneCycleNow and HandMorph>>processEvents). 3. How the HandMorph communicates those events to the appropriate morph - see various implementators of #mouseDown: and #mouseUp:, e.g. Cheers, Bob

2 1

[CCC] False class comment
by Dwight Hughes 28 Feb '00

28 Feb '00

----------------------------------------------------------------- False defines the behavior of its single instance, false -- logical negation. Notice how the truth-value checks become direct message sends, without the need for explicit testing. Be aware however that most of these methods are not sent as real messages in normal use. Most are inline coded by the compiler as test and jump bytecodes - avoiding the overhead of the full message sends. So simply redefining these methods here will have no effect. If you want to actually send them as messages, you will need to send them using #perform:with: and #perform:with:with:. ----------------------------------------------------------------- -- Dwight

1 0

Re: RE: [newbie] clipped method display in system browser
by Bob Arning 28 Feb '00

28 Feb '00

On Sun, 27 Feb 2000 20:28:20 -0500 "Lex Spoon" <lex(a)cc.gatech.edu> wrote: >It's just the Mac distribution which has this particular gotcha. Is >there really no way to work around this? Has anyone tried setting the >app-type-thingy to something other than TEXT and seeing if StuffIt will >still do it's auto-conversion? Umm... I think it's more than just Mac, at least in some cases. I have seen Linux ftp clients do a conversion from CR to LF. Cheers, Bob

1 0

[CCC] Boolean class comment
by Dwight Hughes 28 Feb '00

28 Feb '00

----------------------------------------------------------------- Boolean is an abstract class defining the protocol for logic testing operations and conditional control structures for the logical values represented by its subclasses True and False. Boolean redefines #new so no instances of Boolean can be created. It also redefines several messages in the 'copying' protocol to ensure that only one instance of each of its subclasses True (the global true, logical assertion) and False (the global false, logical negation) exist in the system. ----------------------------------------------------------------- -- Dwight

1 0

← Newer
1
...
4
5
6
7
8
9
10
...
66
Older →

Jump to page: