Squeak-dev June 1998

squeak-dev@lists.squeakfoundation.org

83 participants
130 discussions

Re: Smalltalk-76
by glenn 29 Jun '98

29 Jun '98

At 03:47 PM 6/26/98 -0700, Dan Ingalls wrote: ><snip> >>Perhaps the syntax of ST-80 has been simplified too much. >>I wonder whether ST-76 already featured block closures. Where there >>additional features which are not detailed in the paper and why where they >>dropped in ST-80? What has been added in ST-80 beyond meta classes (and >>perhaps block closures). > >The implementation of blocks was pretty much the same as ST-80, but I >fixed a couple of problems in ST-80. Similarly, the implementation of meta classes was pretty much the same. All classes were instances of classes; the difference was that in '76 all classes were instances of a single class and in '80 we fixed a couple of problems so that only half of the classes were instances of a single class. glenn

1 0

REQUEST: Polling PWS Use
by Mark Guzdial 29 Jun '98

29 Jun '98

Hi Folks! I'd like to get a sense of what people are doing with PWS and PWS-based applications such as Swiki. My motivation is two-fold, and both mercenary :-) - It's tenure packet time, and I'd like to be able to say something about where these tools are being used and for what. - I'm working on a paper on educational uses of Swiki's due Aug. 1, and I would like to be able to say something about uses of Swiki's in educational settings but outside of our campus. If you are using PWS, could you please drop me a personal note (rather than posting to the list) at guzdial(a)cc.gatech.edu? I'd like to know (0) What organization are you with? (1) What are you using PWS for? (2) If you're using it in an education context, what classes are you using it in? (3) (Whether or not you're in an educational context) What's your sense of how the use of PWS is working? Is it successful? Will you continue to use it? If unsuccessful, I'd also appreciate feedback on how it's not working and why. (4) Finally, if I may quote directly from your note, please let me know that directly. I'd appreciate being able to cite people directly. Many thanks! Mark -------------------------- Mark Guzdial : Georgia Tech : College of Computing : Atlanta, GA 30332-0280 (404) 894-5618 : Fax (404) 894-0673 : guzdial(a)cc.gatech.edu http://www.cc.gatech.edu/gvu/people/Faculty/Mark.Guzdial.html

2 1

MessageTally
by sqrmax＠cvtci.com.ar 29 Jun '98

29 Jun '98

Hi. While checking some heavy number crunching, again in the method #spyOn:every: a variable was rendered into nil without the tally knowing about it. This time, the nilled variable came out to be ObservedProcess, and because it was nil, it didn't understand #suspendContext. Maybe the comment there about long primitives has something to do... the tally was as follows: - 681 tallies. **Tree** 100.0 ProbabilityInterval>>binaryUpscale 44.3 Fraction>>timesTwoPower: |44.3 LargePositiveInteger>>bitShift: | 44.3 LargePositiveInteger(Integer)>>bitShift: | 43.9 LargePositiveInteger(Integer)>>digitRshift:bytes:lookfirst: | 43.6 primitives 29.8 Fraction(Number)>>floor |29.8 Fraction>>truncated | 29.8 LargePositiveInteger>>quo: | 29.8 LargePositiveInteger(Integer)>>quo: | 29.8 LargePositiveInteger(Integer)>>digitDiv:neg: | 29.7 primitives 16.4 Fraction>>+ |15.7 LargePositiveInteger(Integer)>>gcd: | 7.8 LargePositiveInteger(Number)>>rem: | |5.1 LargePositiveInteger>>quo: | | |5.1 LargePositiveInteger(Integer)>>quo: | | | 5.1 LargePositiveInteger(Integer)>>digitDiv:neg: | | | 4.8 primitives | |2.6 LargePositiveInteger>>* | | 2.6 LargePositiveInteger(Integer)>>* | | 2.6 LargePositiveInteger(Integer)>>digitMultiply:neg: | 6.2 LargePositiveInteger>>* | 6.2 LargePositiveInteger(Integer)>>* | 6.0 LargePositiveInteger(Integer)>>digitMultiply:neg: 9.4 Fraction>>- 9.4 Fraction>>- 9.4 Fraction>>+ 8.5 LargeNegativeInteger(LargePositiveInteger)>>* 8.5 LargeNegativeInteger(Integer)>>* 8.5 LargeNegativeInteger(Integer)>>digitMultiply:neg: **Leaves** 44.2 LargePositiveInteger(Integer)>>digitRshift:bytes:lookfirst: 35.7 LargePositiveInteger(Integer)>>digitDiv:neg: 17.3 LargePositiveInteger(Integer)>>digitMultiply:neg: As you can see, LOTS of primitives. Andres.

1 0

Can variableByteSubclasses: have fixed inst vars?!
by Andreas Raab 29 Jun '98

29 Jun '98

Hi, The subject says it all. Can variableByteSubclasses have fixed inst vars? The compiler seems to be happy with it but I get all sorts of strange crashes. So the question is: Do we need to fix the compiler or the Interpreter?! ;-) Andreas -- Linear algebra is your friend - Trigonometry is your enemy. +===== Andreas Raab ============= (raab(a)isg.cs.uni-magdeburg.de) =====+ I Department of Simulation and Graphics Phone: +49 391 671 8065 I I University of Magdeburg, Germany Fax: +49 391 671 1164 I +=============< http://isgwww.cs.uni-magdeburg.de/~raab >=============+

2 2

Re: Re: Two questions
by sqrmax＠cvtci.com.ar 27 Jun '98

27 Jun '98

Hi. >>^self class new: value + (self negated and: aProbability negated) value >I assume that the class this method is in implements an "and:" message, >which is what you are trying to use in the code above. Yes. >If this is the case, then you will have to choose a different names for >your operations than "and:" and "or:". Oh, boy... probabilities look so ugly without them... anyway, I chose andWith:, orWith: and so. Do you know of any nicer selectors? Andres.

1 0

Solaris problem w/ dynamic linking
by Andrew M. Duncan 27 Jun '98

27 Jun '98

All: I'm running Solaris 2.5.1 and have a strange linking problem. Running the executable gives this error: % ./SqueakVM* Squeak*.image ld.so.1: ./SqueakVM-2.0-sparc-solaris-2.5.1: fatal: libXext.so.0: can't open file: errno=2 But running ldd reports this: % ldd SqueakVM-2.0-sparc-solaris-2.5.1 libX11.so.4 => /usr/lib/libX11.so.4 libXext.so.0 => (not found) <-----+ libm.so.1 => /usr/lib/libm.so.1 | ... etc ... | ??? libXext.so.0 => /usr/openwin/lib/libXext/so.0 <-----+ There's a conflict between the second and fifth lines quoted here. Is libXext.so.0 found or isn't it?? Suggestions welcome. Andrew Duncan aduncan(a)cs.ucsb.edu http://www.cs.ucsb.edu/~aduncan

2 1

Re: Two questions
by Tim Olson 27 Jun '98

27 Jun '98

>Hi. Hi, Andres: >What does the star besides the cursor's arrow mean? That's the "execute" cursor; it should be displayed whenever an action will take longer than a second or so to give feedback to the user. See the Cursor class constants for a list of the cursors available. > ^self class new: value + (self negated and: aProbability negated) value > >an error occurs and a message appears saying that the arguments for #and: >have to be either a variable or a block? I assume that the class this method is in implements an "and:" message, which is what you are trying to use in the code above. Unfortunately, "and:" is one of the few messages that cannot be redefined, because it is hard-coded in the compiler. The message you see is reported from the compiler when it is performing the transformAnd: Macro Transformation. It is checking to see that the argument to and: is either an immediate block or a variable that might hold a block. If this is the case, then you will have to choose a different names for your operations than "and:" and "or:". -- tim

1 0

Re: Two questions
by GTE/staypufd 27 Jun '98

27 Jun '98

Andres, Your code should be like so: ^self class new: value + (self negated and: [aProbablity negated]) value The reason is is that the 'and:' msg is a keyword msg that takes a block of code as an argument. Its' declaration will be like so: and: aBlock < code here > Look at the implementors of 'and:' and you'll see what I mean.... Hope this helps, Sam Griffith Jr. staypufed(a)aol.com staypufd(a)gte.net

1 0

Two questions
by sqrmax＠cvtci.com.ar 27 Jun '98

27 Jun '98

Hi. What does the star besides the cursor's arrow mean? And also, why is that when I try to compile this: xor: aProbability "Answer the probability that represents the chance that either the receiver, or aProbability but not both of them occur at the same time" ^self class new: value + (self negated and: aProbability negated) value [Boy, what a GREAT cut & paste with crlf!!!] an error occurs and a message appears saying that the arguments for #and: have to be either a variable or a block? Here it is: <- argument of and: must be a block or variable -> What's going on in here??? I have already tried retyping to avoid spilled unseen characters. Andres.

1 0

Morphic is fun
by Stefan Matthias Aust 26 Jun '98

26 Jun '98

Hi! The attached file adds a small but useless feature to morphic system windows - you can add transparency to these windows. I wrote this just because it could be done. Unfortunately, it seems that you cannot rotate transparent system windows, well you can rotate them but they loose their tranparency. Have fun, bye Content-Type: application/octet-stream; name="Transparent Hack.26Jun1009pm.cs" Content-Disposition: attachment; filename="Transparent Hack.26Jun1009pm.cs" Attachment converted: Cog:Transparent Hack.26Jun1009pm.cs (????/----) (0000DA61) -- Stefan Matthias Aust // Are you ready to discover the twilight zone?

1 0