May 2014 - Pharo-dev - lists.pharo.org

Re: [Pharo-dev] a Pharo talk from a ruby conference

by GÃ¶ran Krampe

On 05/12/2014 10:54 PM, Nicolas Cellier wrote: > But Smalltalk V was cheap, small, fairly well documented and worked on > windows (DOS even). Yes, I even used it IIRC, but it was not gratis. And IMHO the only way to compete back then with the big boys (MS, Borland etc) was to either be gratis or open source. Because you couldn't match up when it came to advertising etc. Now, then came Sun with *both* a gratis download of the JDK *and* advertising as hell - bam. I mean... at the University we all used C++ in 1990s because it was free. I did have the luck to actually get to use VW then, but that took some effort from the University to get proper licenses. regards, GÃ¶ran

May 13, 2014

Immutable collections and encapsulation

by Yuriy Tymchuk

Hi, sorry if there was already this question, but I couldnât find it anywhere. Iâm looking in the OO-design concerns and it seams that Java guys are crazy about returning the collection that is used for state of an objects. The only acceptable option is returning it in the immutable wrapper. As far as I know, pharo does not have immutable collections (except from intervals and symbols). Are we missing something important, or there is a philosophy behind the building blocks we have now, and the design we come up while using them. Cheers. Uko

May 13, 2014

Re: [Pharo-dev] WTF and Programming Languages

by Nicolas Cellier

2014-05-13 2:51 GMT+02:00 Nicolas Cellier < nicolas.cellier.aka.nice(a)gmail.com>: > > 2014-05-13 2:41 GMT+02:00 Nicolas Cellier < > nicolas.cellier.aka.nice(a)gmail.com>: > >> >> The sooner failure, the better failure ;) >> >> > Ah, finally a better illustration of what you obtain when you mix the bugs: > > 'hello' asWideString copyReplaceFrom: 11/2 to: 1/2 with: 'world' > > Eventually, I could expect the behavior when invoking asWildString

May 13, 2014

Re: [Pharo-dev] WTF and Programming Languages

by Nicolas Cellier

2014-05-13 2:41 GMT+02:00 Nicolas Cellier < nicolas.cellier.aka.nice(a)gmail.com>: > > 2014-05-08 18:29 GMT+02:00 Sean P. DeNigris <sean(a)clipperadams.com>: > > Sven Van Caekenberghe-2 wrote >> > The why is the result not 'Helloworld' ? or 'Hellworldo' ? Why does the >> $o >> > appear twice ? >> >> Good point. >> >> Please take a moment to: >> 1. go to Ulysses >> 2. follow the account creation link >> 3. create an account on fogbugz >> 4. log in >> 5. create an issue report with: >> a. the vm version >> b. the image version >> c. your platform >> d. your social security number >> e. mother's maiden name >> f. name of first pet >> g. and, a reproducible way to cause the error >> 6. (optional, but preferable) followed up by a slice with tests >> a. re-log-in to fogbugz >> b. go back to the issue page because the fogbugz login page probably >> didn't redirect you to the issue >> c. add whoever is assigned as another person to notify because they will >> paradoxically not be included by default >> 7. (if 6) add class comments to classes that you didn't modify - and >> possibly have never heard of - to make the monkey happy >> 8. Weigh in on whether Pharo is or is not "Smalltalk" AND/OR whether a >> tomato is a fruit or a vegetable >> 9. Cross fingers and wait (but not as long as in Squeak) >> 10. Try to remember what you were doing before you fell down the rabbit >> hole >> 11. Check email for VPRI's final STEPS report while waiting >> >> IN CASE OF EMERGENCY: insert beer between or during any other step OR >> remember that we (meaning the Smalltalk, Smalltalk-inspired, >> Smalltalk-channelled, Smalltalk-related, and/or Dynabook-related >> identified >> persons and possibly spouses and probably children under 17.5) still have >> it >> far better than the rest of the programming world >> >> ;-P (hopefully obviously) >> >> >> > Oh, but wait, I opened some bug reports once upon a time... > I just rediscovered http://bugs.squeak.org/view.php?id=5707 which links > to http://bugs.squeak.org/view.php?id=7194 > Not my fault if a forked happened ;) > > To continue around this very same example, every one knows that this > should fail: > 'hello' copyReplaceFrom: 3/2 to: 5/2 with: 'world'. > Indeed, only integers should be used as indices! > > It should fail equally whether the argument is wide or not as you can see: > 'hello' copyReplaceFrom: 3/2 to: 5/2 with: 'world' asWideString. > Well yes, maybe non integer indices could be used, but Character>>negative > is not understood > > It should fail equally whether the receiver is wide or not as you can see: > 'hello' asWideString copyReplaceFrom: 3/2 to: 5/2 with: 'world' > asWideString. > > A good lesson: tricky hacks like this always come with a price: > index isNumber > ifTrue: [^self at: index asInteger] > ifFalse: [self errorNonIntegerIndex] > > The sooner failure, the better failure ;) > > Ah, finally a better illustration of what you obtain when you mix the bugs: 'hello' asWideString copyReplaceFrom: 11/2 to: 1/2 with: 'world' >> ----- >> Cheers, >> Sean >> -- >> View this message in context: >> http://forum.world.st/WTF-and-Programming-Languages-tp4758263p4758348.html >> Sent from the Pharo Smalltalk Developers mailing list archive at >> Nabble.com. >> >> >

May 13, 2014

Re: [Pharo-dev] WTF and Programming Languages

by Nicolas Cellier

2014-05-08 18:29 GMT+02:00 Sean P. DeNigris <sean(a)clipperadams.com>: > Sven Van Caekenberghe-2 wrote > > The why is the result not 'Helloworld' ? or 'Hellworldo' ? Why does the > $o > > appear twice ? > > Good point. > > Please take a moment to: > 1. go to Ulysses > 2. follow the account creation link > 3. create an account on fogbugz > 4. log in > 5. create an issue report with: > a. the vm version > b. the image version > c. your platform > d. your social security number > e. mother's maiden name > f. name of first pet > g. and, a reproducible way to cause the error > 6. (optional, but preferable) followed up by a slice with tests > a. re-log-in to fogbugz > b. go back to the issue page because the fogbugz login page probably > didn't redirect you to the issue > c. add whoever is assigned as another person to notify because they will > paradoxically not be included by default > 7. (if 6) add class comments to classes that you didn't modify - and > possibly have never heard of - to make the monkey happy > 8. Weigh in on whether Pharo is or is not "Smalltalk" AND/OR whether a > tomato is a fruit or a vegetable > 9. Cross fingers and wait (but not as long as in Squeak) > 10. Try to remember what you were doing before you fell down the rabbit > hole > 11. Check email for VPRI's final STEPS report while waiting > > IN CASE OF EMERGENCY: insert beer between or during any other step OR > remember that we (meaning the Smalltalk, Smalltalk-inspired, > Smalltalk-channelled, Smalltalk-related, and/or Dynabook-related identified > persons and possibly spouses and probably children under 17.5) still have > it > far better than the rest of the programming world > > ;-P (hopefully obviously) > > > Oh, but wait, I opened some bug reports once upon a time... I just rediscovered http://bugs.squeak.org/view.php?id=5707 which links to http://bugs.squeak.org/view.php?id=7194 Not my fault if a forked happened ;) To continue around this very same example, every one knows that this should fail: 'hello' copyReplaceFrom: 3/2 to: 5/2 with: 'world'. Indeed, only integers should be used as indices! It should fail equally whether the argument is wide or not as you can see: 'hello' copyReplaceFrom: 3/2 to: 5/2 with: 'world' asWideString. Well yes, maybe non integer indices could be used, but Character>>negative is not understood It should fail equally whether the receiver is wide or not as you can see: 'hello' asWideString copyReplaceFrom: 3/2 to: 5/2 with: 'world' asWideString. A good lesson: tricky hacks like this always come with a price: index isNumber ifTrue: [^self at: index asInteger] ifFalse: [self errorNonIntegerIndex] The sooner failure, the better failure ;) > ----- > Cheers, > Sean > -- > View this message in context: > http://forum.world.st/WTF-and-Programming-Languages-tp4758263p4758348.html > Sent from the Pharo Smalltalk Developers mailing list archive at > Nabble.com. > >

May 13, 2014

Re: [Pharo-dev] Github & Pharo.org

by Nicolai Hess

link for windows vm is wrong(Custom Downloads): http://files.pharo.org/vm/pharo/win/Pharo-VM-linux-stable.zip correct one: http://files.pharo.org/vm/pharo/win/Pharo-VM-win-stable.zip 2014-05-12 14:18 GMT+02:00 Esteban Lorenzano <estebanlm(a)gmail.com>: > I added it at download page: http://pharo.org/download > now we have to improve all that infrastructure :) > > Esteban > > On 03 May 2014, at 19:06, phil(a)highoctane.be wrote: > > Am I blind or am I having a hard time spotting our stuff on GitHub from > Pharo.org as well as some other properties that help us be visible? > > Meaning: > > https://github.com/orgs/pharo-project/ > > https://github.com/SquareBracketAssociates > > are very important places and aren't mentioned much. > > As are: https://github.com/tide-framework if someone wants to use Marina > so use the Pharo power... > > Maybe can we add these and > > http://en.wikipedia.org/wiki/Pharo > > as well (as we bothered updating the page...) > > Also, in http://pharo.org/get-involved we sent people directly to the > Fogbugz thing. > > We should mention this Penelope monkey as well: http://bugs.pharo.org/ > > Also, in documentation, people are doing webapps, so we need to add > "Dynamic Web Development with Seaside" to http://pharo.org/documentation > > http://book.seaside.st/book > > And, we also must also add this gorgeous > > http://spec.st/ > > and Sebastian's Mapless supernice site as well. > > > I do not have access to the CMS, so please some kind soul, do it :-) [or > tell me how] > > Phil > > >

May 12, 2014

Re: [Pharo-dev] a Pharo talk from a ruby conference

by David Astels

I worked at a company writing commercial (i.e. shrink wrapped boxes on store shelves) application using Smalltalk V. On May 12, 2014, at 3:54 PM, Nicolas Cellier <nicolas.cellier.aka.nice(a)gmail.com> wrote: > > > > 2014-05-12 22:33 GMT+02:00 GÃ¶ran Krampe <goran(a)krampe.se>: > Hi! > > On 04/30/2014 10:02 PM, kilon alios wrote: > Another mistake is that people tend to over idealising Smalltalk and it > appears as if Smalltalk used to be popular, but I have found no evidence > that Smalltalk was ever popular. Again I may be wrong but this is also > maybe a motivation to regard Smalltalk dead. > > It was quite popular in... 1985-ish to 1995-ish. I would guess that during those years VisualWorks and VisualAge (primarily) covered 33% of the OOP market and C++ about 60% - and the rest by other even smaller things like Eiffel. Those numbers I recall from some magazine, so I am not making them up. If you were into OO at the time it was quite a lot of buzz around both Smalltalk and C++ IMHO. > > But OOP was almost exclusively used in large corporations or institutions that could muster the licenses. But Smalltalk *was* fairly big and some truly huge systems were built. > > But it was not in any serious awareness outside the corporate world - since there was hardly any cheap or free Smalltalk available. C++ was though and ate up that space, and of course... > > ...you know what came in 1995. :) > > If say... Dolphin had been born as an open source (or at least gratis download) project - so that people could easily build Win32 apps for consumer use, like VB or Deplhi... then perhaps the world had been different. > > regards, GÃ¶ran > > > But Smalltalk V was cheap, small, fairly well documented and worked on windows (DOS even).

May 12, 2014

Re: [Pharo-dev] a Pharo talk from a ruby conference

by Nicolas Cellier

2014-05-12 22:33 GMT+02:00 GÃ¶ran Krampe <goran(a)krampe.se>: > Hi! > > On 04/30/2014 10:02 PM, kilon alios wrote: > >> Another mistake is that people tend to over idealising Smalltalk and it >> appears as if Smalltalk used to be popular, but I have found no evidence >> that Smalltalk was ever popular. Again I may be wrong but this is also >> maybe a motivation to regard Smalltalk dead. >> > > It was quite popular in... 1985-ish to 1995-ish. I would guess that during > those years VisualWorks and VisualAge (primarily) covered 33% of the OOP > market and C++ about 60% - and the rest by other even smaller things like > Eiffel. Those numbers I recall from some magazine, so I am not making them > up. If you were into OO at the time it was quite a lot of buzz around both > Smalltalk and C++ IMHO. > > But OOP was almost exclusively used in large corporations or institutions > that could muster the licenses. But Smalltalk *was* fairly big and some > truly huge systems were built. > > But it was not in any serious awareness outside the corporate world - > since there was hardly any cheap or free Smalltalk available. C++ was > though and ate up that space, and of course... > > ...you know what came in 1995. :) > > If say... Dolphin had been born as an open source (or at least gratis > download) project - so that people could easily build Win32 apps for > consumer use, like VB or Deplhi... then perhaps the world had been > different. > > regards, GÃ¶ran > > But Smalltalk V was cheap, small, fairly well documented and worked on windows (DOS even).

May 12, 2014

Re: [Pharo-dev] a Pharo talk from a ruby conference

by GÃ¶ran Krampe

Hi! On 04/30/2014 10:02 PM, kilon alios wrote: > Another mistake is that people tend to over idealising Smalltalk and it > appears as if Smalltalk used to be popular, but I have found no evidence > that Smalltalk was ever popular. Again I may be wrong but this is also > maybe a motivation to regard Smalltalk dead. It was quite popular in... 1985-ish to 1995-ish. I would guess that during those years VisualWorks and VisualAge (primarily) covered 33% of the OOP market and C++ about 60% - and the rest by other even smaller things like Eiffel. Those numbers I recall from some magazine, so I am not making them up. If you were into OO at the time it was quite a lot of buzz around both Smalltalk and C++ IMHO. But OOP was almost exclusively used in large corporations or institutions that could muster the licenses. But Smalltalk *was* fairly big and some truly huge systems were built. But it was not in any serious awareness outside the corporate world - since there was hardly any cheap or free Smalltalk available. C++ was though and ate up that space, and of course... ...you know what came in 1995. :) If say... Dolphin had been born as an open source (or at least gratis download) project - so that people could easily build Win32 apps for consumer use, like VB or Deplhi... then perhaps the world had been different. regards, GÃ¶ran

May 12, 2014

Re: [Pharo-dev] Roassal 3D empty windows

by Alexandre Bergel

> Thank you Alexandre, > > just one question, what system do you use (OS/Grahics). I find it strange that > I am the only one with this issue on Roassal resp. NBOpenGL > (I use Windows 7 with /AMD Graphics I tried a Intel graphics card as well but that > one does not support multisample framebuffer) Iâve just discovered that some examples do not work on my laptop but work on my desktop machine. Same OS, but apparently, different graphic card. On my desktop machine (iMac) I have a NVIDIA GeForce GTX 675MX 1024 MB and I have the last OSX. Were you able to get at least a basic rendering? Cheers, Alexandre

May 12, 2014