Squeak-dev June 2001

squeak-dev@lists.squeakfoundation.org

122 participants
362 discussions

Re: On interfaces - Morphic Wrappers
by Gerardo Richarte 06 Jun '01

06 Jun '01

John Kozak wrote: > > Just downloaded and fileIn'd this, and seem to have hit a bugette > (this with squeak 3.1 4081). I'm sure you haven't done anything wrong. This is what I meant when I said that it was hard to mantain a package up to date. The problem with the text appearing outside the balloon is not new, it appeared with one of the latests Squeak updates (between 2000 and 4048), we tried to fix it, but it will take some time (does anybody want to give it a try?). The other problem (the doesNotUnderstand:) is pretty new, I haven't heard of it before, so it must have appeared with an update between 4048 and 4081. As soon as we update our images, and have some time, we'll fix it... sorry for the inconvenience... I'll announce the updated version on the list as soon as we have it. Buggy Bye! Richie --- For a personal reply use gera(a)core-sdi.com

2 1

Distributed squeak?
by Francisco Colho 06 Jun '01

06 Jun '01

Hi, Is there some option to "execute" some squeak code on a remote VM/Image? If not, wouldn't this be great? -- Francisco Coelho ==================================== Departamento de Matematica Universidade de Evora Rua Romao Ramalho, 59 7000-671 Evora tel. 351 91 900 63 79 email fc(a)dmat.uevora.pt www http://www.dmat.uevora.pt/~fc ====================================

4 4

Re: CAS and Squeak
by Henrik Gedenryd 06 Jun '01

06 Jun '01

I enjoyed your explanation, I could sort of guess to some % of certainty what you meant with MGP, by trying to blend the idea of metaclasses with that of genotype/phenotype, but I couldn't be sure I had blended it correctly. Now I think my interpretation was valid that in MGP phenotype corresponds to instance and genotype to class. > For implementing adaptive systems > with a complexity like that above(including GAs), MGP is very useful. And in that context I believe it becomes more concrete and therefore easier to understand as well. I can also picture how MGP becomes useful in that context, while almost seeming absurd in its hyper-abstractness if presented without a useful application. I would then assume that Whole Value in this context is about reifying the apparently abstract concept of phenotype, for example. Consider yourself encouraged to keep this coming! regards, Henrik

1 0

Re: On interfaces - Morphic Wrappers
by Gerardo Richarte 06 Jun '01

06 Jun '01

Hi Alan! This are great news! I whould love to help with the integration, but I'm not sure what it means... > There clearly should also be an integration of > some of Ned Konz's stuff into the "writing on air" interface. Here at work, Diego and Luciano are using Ned's connectors (a modified version, I think) together with MWs. Is that what you meant? What kind of integration do you have in mind? Integral Bye! Richie --- For a personal reply use gera(a)core-sdi.com

1 0

RE: On interfaces - Morphic Wrappers
by Jerry Bell 06 Jun '01

06 Jun '01

Extremely good news! I for one am really looking forward to seeing what this "omniuser" environment shapes up to be. Are there any clues in the image as it stands now, experimental add-ons, or notes/sketches available? Jerry Bell jdbell(a)fareselaw.com > -----Original Message----- > From: Alan Kay [mailto:Alan.Kay@squeakland.org] > Sent: Wednesday, June 06, 2001 11:24 AM > To: squeak(a)cs.uiuc.edu > Cc: recipient list not shown; @cs.uiuc.edu > Subject: RE: On interfaces - Morphic Wrappers > > > We like it too. We have an itch to make it fit in a little more with > the main stream "omniuser" environment (and with Nathanial's Genie > recognizer) that is s l o w l y being built by us, > and we just haven't had the time to give Morphic Wrappers the > attention it deserves. There clearly should also be an integration of > some of Ned Konz's stuff into the "writing on air" interface. > > Also, the exit from Disney has slowed things down a bit. > > Cheers, > > Alan and Dan >

1 0

Re: On interfaces - Morphic Wrappers
by Dan Rozenfarb 06 Jun '01

06 Jun '01

--- In squeak@y..., Jerry Bell <jdbell@f...> wrote: > I've always liked playing around with Morphic Wrappers. I'm surprised that > it hasn't migrated into the main release yet- is there any reason it hasn't > been folded in? Typing on air is fun! I strongly second this. Although I believe MorphicWrappers still need some work, IMHO they are an indispensable part of the Morphic world, mostly typing in the air. Just as Smalltalk, you've got to use them for a while to really graps their usefulness. Regards, Dan Rozenfarb > > Jerry Bell > jdbell@f... > > > -----Original Message----- > > From: Gerardo Richarte [mailto:core.lists.squeak@c...] > > Sent: Tuesday, June 05, 2001 1:43 PM > > To: squeak@c... > > Cc: recipient list not shown; @cs.uiuc.edu > > Subject: On interfaces - Morphic Wrappers > > > > > > This days we have a thread on user interfaces, this > > time is about > > how to use Squeak with non-standard input devices, or, more precisely, > > without keyboard. It is not the first time we can enjoy this kind of > > discussions, there have been a lot of them in the list. > > Personally, I love > > to read about UIs. > > > > Today, we can say that once again there is a version of the > > MorphicWrappers available to download. This latest version is ready to > > be used in a Squeak3.1a-4048. It's not a long time since that previous > > update (just 100 updates), but before that it was a little longer (800 > > updates). > > > > It's a really tough job to maintain MorphicWrappers (we > > all suffer > > for maintaining packages for squeak up to date...) We almost > > gave up, but > > the team keeps growing and the new people comes with fresh energy to > > maintain the package. > > > > So here it comes, from deep in the heart of the latest > > adopters of > > the MorphicWrappers way of interacting with the World: > > > > We started the project almost 3 years ago with the first version > > of Morphic, in Squeak 2.3 (I'm not really sure when it was). > > MW is today > > an ancient project, but we think it's still current, and even > > a little in > > the future. We´ve been using it continously all this time, and we got > > adicted to them. > > > > Here at core where I work, there is people that learned > > Smalltalk & Squeak using them, and the other day, one of them > > asked what a Workspace > > is! just imagine that! > > > > Don't wait! You have to live the MorphicWrappers > > experience, then, you'll have the experience to say "I don't > > like MWs", but you won't say > > it. > > Go get them at http://mathMorphs.swiki.net/MorphicWrappers > > > > Interfacing Bye! > > Once again, Richie > > > > --- For a personal reply use gera@c... > > > > _________________________________________________________ ¿Lo probaste? Correo gratis y para toda la vida en http://correo.yahoo.com.ar

1 0

RE: On interfaces - Morphic Wrappers
by Jerry Bell 06 Jun '01

06 Jun '01

I've always liked playing around with Morphic Wrappers. I'm surprised that it hasn't migrated into the main release yet- is there any reason it hasn't been folded in? Typing on air is fun! Jerry Bell jdbell(a)fareselaw.com > -----Original Message----- > From: Gerardo Richarte [mailto:core.lists.squeak@core-sdi.com] > Sent: Tuesday, June 05, 2001 1:43 PM > To: squeak(a)cs.uiuc.edu > Cc: recipient list not shown; @cs.uiuc.edu > Subject: On interfaces - Morphic Wrappers > > > This days we have a thread on user interfaces, this > time is about > how to use Squeak with non-standard input devices, or, more precisely, > without keyboard. It is not the first time we can enjoy this kind of > discussions, there have been a lot of them in the list. > Personally, I love > to read about UIs. > > Today, we can say that once again there is a version of the > MorphicWrappers available to download. This latest version is ready to > be used in a Squeak3.1a-4048. It's not a long time since that previous > update (just 100 updates), but before that it was a little longer (800 > updates). > > It's a really tough job to maintain MorphicWrappers (we > all suffer > for maintaining packages for squeak up to date...) We almost > gave up, but > the team keeps growing and the new people comes with fresh energy to > maintain the package. > > So here it comes, from deep in the heart of the latest > adopters of > the MorphicWrappers way of interacting with the World: > > We started the project almost 3 years ago with the first version > of Morphic, in Squeak 2.3 (I'm not really sure when it was). > MW is today > an ancient project, but we think it's still current, and even > a little in > the future. We´ve been using it continously all this time, and we got > adicted to them. > > Here at core where I work, there is people that learned > Smalltalk & Squeak using them, and the other day, one of them > asked what a Workspace > is! just imagine that! > > Don't wait! You have to live the MorphicWrappers > experience, then, you'll have the experience to say "I don't > like MWs", but you won't say > it. > Go get them at http://mathMorphs.swiki.net/MorphicWrappers > > Interfacing Bye! > Once again, Richie > > --- For a personal reply use gera(a)core-sdi.com > >

1 0

RE: Wormhole Instability (was Why do things keep changing?)
by Bruce ONeel 06 Jun '01

06 Jun '01

Hi, In one of my past lives I ran a load of email lists via different list processors. It is suprising how hard this is to get running well. You'd think it would be easy, but, you want to distribute an inbound message out to multiple outbound processes and then those multiple out bound processes try to send the mail out. On a high traffic mail list you eassily can get replies going out before the origional message goes out just by having one of the outbound process hang trying to contact someone's inbound email server. And you can easily get multiple copies if the last little bit of the SMTP protocol is dropped on the floor by TCP. Sigh.. Another choice is to force single threading, ie, one message is delivered to the whole list of users before the next message is delivered. On a list of any size this will result in very few messages per day since some node on the list is always disconnected and requires a timeout and wait for N minutes to retry. Oh yea, the number of bounces, even from well connected nodes is stunning. Kudos to teh uiuc folks :-) cheers bruce "Jarvis, Robert P. $Contingent$" <Jarvisb(a)timken.com> wrote: > Ouch - that came out harsher than I meant. > > To be fair to the people at UIUC, who have been incredibly generous in the > support of the Squeak and Smalltalk communities, my observation is that > *ALL* list servers have problems. > > Bob Jarvis > Compuware @ Timken > > > -----Original Message----- > > From: Jarvis, Robert P. (Contingent) > > Sent: Thursday, May 31, 2001 8:17 AM > > To: 'squeak(a)cs.uiuc.edu' > > Subject: RE: Wormhole Instability (was Why do things keep changing?) > > Importance: High > > > > > > > -----Original Message----- > > > From: JArchibald(a)aol.com [mailto:JArchibald@aol.com] > > > Aha! Now I understand Fleeberz's comment: "I've only had a > > > few bounces in the last 17 years or so". > > > > > > THEY ARE BOTH WRITING TO US FROM SEVENTEEN YEARS IN THE FUTURE. > > > AND THE MAILING LIST IS STILL SCREWING UP !!! > > > > Yes, well...isn't it nice to know that despite the ups and > > the downs, the > > vicissitudes of life, the general tendency towards chaos and > > anarchy, and > > the lack of a good five cent cigar, one can count on some > > things and say, > > "Well, no matter what else may happen you can always count on > > the Squeak > > list server to have problems". It's rather endearing, really... :-) > > > > Bob Jarvis > > Compuware @ Timken > >

1 0

Re: CAS and Squeak
by G�ran Hultgren 06 Jun '01

06 Jun '01

Hi Robert and guys! I am totally uneducated on GAs but did dabble with them for fun in VW 5 years ago. My silly question with some background first: I remember I built a small framework with different DNA mixing algorithms, different rules for mating and survival etc. etc. and toyed on the "travelling salesman" with it. When I was kneedeep in parameters - like 30 or so - trying them out and seeing what made it go faster/better etc. I realized that THAT in itself was an optimizationproblem that ought to be applicable to GAs... But pretty soon my old PC choked on the idea of having to run say 50 generations on 100 individuals for each fitness function evaluation... But I still liked the idea though! Q: Has that been done? Have any of you working with GAs more professionally done it? regards, G�ran ===== G�ran Hultgren, goran.hultgren(a)bluefish.se GSM: +46 70 3933950, http://www.bluefish.se "Department of Redundancy department." -- ThinkGeek __________________________________________________ Do You Yahoo!? Get personalized email addresses from Yahoo! Mail - only $35 a year! http://personal.mail.yahoo.com/

1 0

SyntaxHighlighting on 3.0
by Aron Brener 06 Jun '01

06 Jun '01

When I was using version 2.7 I used SyntaxHighlighting to highlight my code. I upgraded to 3.0 and loaded it into my image. It worked ok for a while but not it gives me a mnu #writeOnScan: Is there a version of SyntaxHighlighting compatible with 3.0? Or some other package that is similar that I could use? TIA -- ------------------------------------------------------------------------- Aron Brener, BASc | Caminante no hay camino, brener(a)home.com | se hace camino al andar. | Antonio Machado ------------------------------------------------------------------------- miliHelen: Quantity of beauty required to launch one ship.

5 5

← Newer
1
...
28
29
30
31
32
33
34
...
37
Older →

Jump to page: