Squeak-dev September 1998

squeak-dev@lists.squeakfoundation.org

108 participants
197 discussions

Re: Squeak/2 - still some trouble
by Boris G. Chr. Shingarov 05 Sep '98

05 Sep '98

Hi Michael, > Thanks for the note re the updated image addressing a > segementation fault. > > Cheese begins to start, but fails (and this time gives > an error log entry: Oh yes. Sorry. That's my fault. You need the VM v.59, http://www.plugcom.ru/~squeak/gCheese.exe.zip . Again, sorry for the inconvenience. Boris

1 0

Windoze-2-1-1.cs coming soon
by Sam Adams 04 Sep '98

04 Sep '98

Squeakers, I have ported Windoze-1-3-4 to Version 2.1 and uploaded it to UIUC. All of an afternoon's work. I really like the new pluggable views framework. Should take a couple od days to show up. Perhaps Reinier van Loon will soon port all his enhancements as well! Regards, Sam Sam S. Adams, Distinguished Engineer, IBM Network Computing Software Division 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: Base image, Windows look
by vuletich＠ar.ibm.com 04 Sep '98

04 Sep '98

To: squeak(a)cs.uiuc.edu cc: From: Hi. Alejandro Reimondo made sqAPIs, a complete framework for accesing any host OS facility (including native widgets). It works on the Mac and Windows, and will be ported to other OSes. It is at http://www.sugarweb.com/Miniatures/downloads.html. The Cheese project from Boris Shingarov is at http://www.plugcom.ru/~squeak .. It is a new environment, based on Squeak, but evolving to using only host native windows. It runs on OS/2, but Boris needs help for making it work on other OSes. I thank theese folks for their great work, but I don't think this stuff should be in the base image. I worked with Smalltalk/V Windows and VisualAge Smalltalk for several years and got a lot of headaches because of native widgets. The frameworks for native widgets are more convoluted than MVC and Morphic, and what's worse, they are seriously limited by the functionality supported by the OS, because you can't control all the code that runs. Also, when something breaks, there are things you can't inspect. The reason for leaving something out of the base image is not only to keep it small. It is also to keep it consistent and easy to understand. And to follow Squeak's philosophy. Juan Manuel Vuletich jvuletic(a)dc.uba.ar

8 9

Re: Scientific Computing
by m3rabb＠stono.com 04 Sep '98

04 Sep '98

Before I get flamed, I want to back away from this suggestion a bit. I went further back in the archives, and read Dave N. Smith's posts about doubles, and the many posts about immediate small floats. In the particular application for which I need FloatArrays, I only need a few digits of precision, and the exponential range is not at all extreme. I wonder how often this is the case. I started thinking about this when I checked all of the instance of Float in the system. My guess is that most of the values in the virgin system are a result of windowing ratios, etc. None of these floats seem to require more than single precision. Would it make sense to have a SmallFloat as an immediate that would convert itself to a normal double precision (or more) when necessary? I know my application would also work with fixed points as well. Perhaps Andreas 3D graphics application could use them as well. At 7:01 PM 9/2/98, Christopher Oliver wrote: >To be blunt, why should one CPU receive preferential attention? >Maybe this should be a compile time issue for the VM. I should have said both single and double precision arrays. --Maurice BTW, I see that it was only Hans-Martin Mosner's experimental system that uses more than a single tag bit. Hans, is your website temporarily down? I haven't been able to get on for the past few days? --------------------------------------------------------------------------- Maurice Rabb 773.281.6003 Stono Technologies, LLC Chicago, USA

2 1

Re: Controller loops
by Gerardo Richarte (CORE-SDI) 04 Sep '98

04 Sep '98

vuletich(a)ar.ibm.com said: > This makes the MVC controllers poll for activity no more than 20 times a > second, what is perfectly enough. > > I found that this frees almost all the processor time to other programs in > a multitasking OS. > All that is needed is to have the ioRelinquishProcessorForMicroseconds() > function in the > VM properly implemented. I tried it and works OK in Windows and OS/2 > versions of Squeak. > > Does this modification deserve to be in the base image? I'm sure it does, and as soon as possible!!! Hurried Bye! Richie++

1 0

Re: Could somebody help a newbie please?
by Dwight Hughes 04 Sep '98

04 Sep '98

Along with Stefan's overview you might find "Applications Programming in Smalltalk-80(TM): How to use Model-View-Controller (MVC)" by Steve Burbeck at http://st-www.cs.uiuc.edu/users/smarch/st-docs/mvc.html to be a help. It covers MVC as it was implemented in ParcPlace Smalltalk-80 v2.0 -- an early version not much divergent from Squeak's MVC implementation (starting with v2.5 the MVC architecture changed quite a bit). -- Dwight Joshua Johnson Marker wrote: > > I haven't read all of your reply, but it looks great. You should post it > to c.l.smalltalk, and to the FAQ -- it's the kind of infor that I had a > hard time finding as well. > > --- > Joshua (lux) * HMRL: 734.647.5778 > Sorting out my life. . . .in O(n lg n). > There's only one way to use 'Windows NT' and 'security' in the same > sentence, and this is it. > --- > Windows '98 -- guaranteed Year 1990 compliant.

2 1

Re: Standards ... Who needs them?
by JArchibald＠aol.com 04 Sep '98

04 Sep '98

In a message dated 9/3/98 2:05:21 PM Eastern Daylight Time, Allen_Wirfs- Brock(a)Instantiations.com writes: << While the Squeak community is certainly free to do whatever it can agree upon (i.e. Squeak ~~ Smalltalk) it is probably worth noting that there is an ANSI standard definition of the core Smalltalk classes. Unless or until the Squeak community seriously reworks the entire Smalltalk selector vocabulary, confusion will probably be minimized if issues like this were resolved by following the consensus that was achieve by the commercial Smalltalk community. >> Oh! Alan ... You wouldn't want us to get bogged down with standards, now, would you. Isn't more fun to have a surprise everyday. Where would this mailing list be if things were standardized--there wouldn't be anything to talk about. BTW, When you say Squeak~~Smalltalk, What Smalltalk do you mean? This community has often been defined by who had the hottest (most closely watched, if you will) Smalltalk of the day. At the moment, That _is_ Squeak. So maybe, Squeak==Smalltalk. I have close to a dozen Smalltalks running on my machine right now. I'll bet that none of them is ANSI standard (sticking my neck out here a bit ...) All kidding aside, how do you convince _experimenters_ that it is worth their while to pay attention to such issues. Re-engineering the wheel seems to be a greater _learning_ experience, to them. Jerry.

1 0

R: Convex hulls
by Alessandro Manunza 04 Sep '98

04 Sep '98

Hi, excuse me, what is a "convex hull algorithm "? Thanks Alessandro ---------- > Da: ocit.inc <ocit.inc(a)MCI2000.com> > A: squeak(a)cs.uiuc.edu > Oggetto: Convex hulls > Data: venerdì 4 settembre 1998 6.56 > > I am on the verge of attempting to implement an ST version of a convex hull > algorithm however I'm kind of busy at the moment and would prefer to borrow > one. > > Can anybody help? > > Thanks. > > Charles > >

1 0

Re: Base image, Windows look
by Stefan Matthias Aust 04 Sep '98

04 Sep '98

Doug Way wrote: >Still, after playing around with Squeak for a few months, there are some >things I'd like to see moved to the base image. (Additional to the things already mentioned, I'd like to see the monolithic base image getting extended and eventually replaced by some kind of packages/parcels, similar to Dolphin ST or VisualWorks ST.) >A related item is supporting widgets with a Windows (or Mac, Motif, etc.) >look & feel. Although I could understand keeping this kind of thing out >of the base image, as it could start to take up a decent amount of >space... it could be part of a larger Windows-look development goodie or >something. Probably the way to go with this would be emulated (Morphic) >widgets rather than native widgets, sticking with the philosophy of >keeping Squeak very portable. Has anyone tried creating Windows-look >widgets with Morphic? Well, I didn't tried this so far, but I already wrote a Windows look for VisualWorks and I'm currently experimenting with a NextStep look (but no feel) for Squeak's MVC classes. First of all, emulating an existing GUI is a damn big challenge you cannot code in a weekend (unfortunately, I once tried ;-). There's always the problem that you get the look only 90% right and the last remaining 10% make the people notice that everything is only fake. Then, at least as important as the look is the right feel. It's not a big problem if button shadows don't look right, but if you cannot use the usual short cut keys, or if menus behave differently or if you cannot use the GUI also without a mouse, then people will refuse the emulation. Both Squeak's MVC and Morphic GUI behave very different compared to Apple's GUI or Microsoft's Windows (not even thinking about X Windows here). This means, if you want to get the right feel, you've basically to rewrite the whole GUI. Currently, Squeak doesn't support keyboard navigation (keyboard support is overall different, you've to move the mouse over the window you want to type in, for example), doesn't support accelerator keys for menus, no keyboard menu control, no concept of an input focus. Squeak doesn't support a large number of common widgets like tree views, table views, file selectors, sliders, spin buttons, (c/t)oolbars, floating toolboxes, etc. It has no real concept of scrolling panes (horizontal scrollbars are missing altogether and the vertical ones are hacked into the ScrollController). The text widget (although quite powerful) if very different to something you are used to use in Windows. Pulldown menus and submenus are missing, popup menus behave different. I prefer to perform two clicks instead of continuesly pressing the mouse button. The request paradigm of Squeak is different. It uses special popup menus where the Windows user would expect messages boxes or dialogs. Last but no least, Squeak's cursors are only monochrome and doesn't resemble the Windows ones. For my experiments with a NextStep look, I changed the MVC classes. So far, this wasn't very difficult after I eventually figured out the meaning of the existing stuff. There's definitely not enough documentation, (but that seems to be a typical Smalltalk problem). Using the browsers to change the underlying GUI framework if however a little bit like brain surgery :-) The feel is however still Squeakish and probably not Next-like at all (I don't know as I never had the chance to see or use a Next, but I always liked its look.) but that's wasn't my goal. I also looked at Morphic and found this much more difficult to use. Morphic is very promising, but still very experimental. As it mixes up model, view and controller in just one class, it's difficult for me to tell what method is used (if it's used at all) for which task. It's lacking documentation! (also Morphic is too slow on my P100) I've attached a Win4-style button written in Morphic. Feel free to extend this :-) I'd be interested in a NextStep- (or Win4-)like look (and feel) based on the MVC framework. You cannot scale and rotate this, but it's much faster and (at least for me) easier to understand. Something that looks better and can be used to create some good looking applications without too much difficulties. BTW, there's already a patch at the UIUC Smalltalk archive which gives you a sort-of Win4 look, but again just the look, not the feel, I think. bye Content-Type: application/octet-stream; name="Morphic-Win4.1Sept944pm.cs" Content-Disposition: attachment; filename="Morphic-Win4.1Sept944pm.cs" Attachment converted: Cog:Morphic-Win4.1Sept944pm.cs (????/----) (0000E42A) -- Stefan Matthias Aust // Are you ready to discover the twilight zone?

2 1

Convex hulls
by ocit.inc 04 Sep '98

04 Sep '98

I am on the verge of attempting to implement an ST version of a convex hull algorithm however I'm kind of busy at the moment and would prefer to borrow one. Can anybody help? Thanks. Charles

1 0

← Newer
1
...
12
13
14
15
16
17
18
19
20
Older →

Jump to page: