- Pharo-users - lists.pharo.org

Re: [Pharo-users] Where do we go now ?

by Benoit St-Jean

BTW, why put an .exe installer for Windows available when it crashes right from the start? It just doesn't work at all.Â Period.Â For everyone. And I thought looking for senders of a method was something we mastered a long time ago, like starting with Smalltalk-76.Â Am I supposed to assume that everything, even basic functionalities, are all broken because it's labeled "alpha" ? ----------------- BenoÃ®t St-Jean Yahoo! Messenger: bstjean Twitter: @BenLeChialeux Pinterest: benoitstjean Instagram: Chef_Benito IRC: lamneth Blogue: endormitoire.wordpress.com "A standpoint is an intellectual horizon of radius zero".Â (A. Einstein) On Friday, April 13, 2018, 5:20:28 a.m. EDT, Esteban Lorenzano <estebanlm(a)gmail.com> wrote: On 13 Apr 2018, at 11:07, Benoit St-Jean <bstjean(a)yahoo.com> wrote: I'm on Windows 10, using Pharo 7.0 alpha 32 bit. and btwâ¦ which part of ALPHA you do not get? Esteban

April 13, 2018

Re: [Pharo-users] Where do we go now ?

by Cyril Ferlicot D.

On 13/04/2018 11:07, Benoit St-Jean via Pharo-users wrote: "Can I at least expect to be able to save code in my local Monticello repository ? It bombs. I just reported a bug!" "Can I expect that when looking for implementors of a method things still work? I get duplicates (see attached file)." => The meaning of alpha is that things like this can happen often. Else it would be a beta or a release version. But if you stick to use it, the bug reports are really welcomed! For the windows launcher indeed there is weird behaviour and Christophe is working on it when he gets bug reports. (https://github.com/pharo-project/pharo-launcher/issues) "Can I expect reading input from the console to work on Windows? It doesn't." What do you mean by that? Having Pharo running and the user sending input from it's CMD? I never saw an application doing that on windows. "Can I expect to be able to contribute ? That is kinda sad but all the cool scripts out there are .sh files." Which scripts? I contribute to Pharo from windows 7 and windows 10 without any script. Sometimes I have some troubles but it's only natural during the version that change all the contribution process the time it stabilize. If you have troubles on windows can you open issues please? Because most of the time I see that people complain about some bugs but some of those bugs never got a bug report. -- Cyril Ferlicot https://ferlicot.fr

April 13, 2018

Re: [Pharo-users] Where do we go now ?

by Esteban Lorenzano

> On 13 Apr 2018, at 11:07, Benoit St-Jean <bstjean(a)yahoo.com> wrote: > > I'm on Windows 10, using Pharo 7.0 alpha 32 bit. > and btwâ¦ which part of ALPHA you do not get? Esteban

April 13, 2018

Re: [Pharo-users] Where do we go now ?

by Esteban Lorenzano

Can I expect a PR to help? I tell you again: That tone does not helps your case, please calm down. Esteban > On 13 Apr 2018, at 11:07, Benoit St-Jean <bstjean(a)yahoo.com> wrote: > > "You cannot take an alpha version and expect production quality." > > I'm on Windows 10, using Pharo 7.0 alpha 32 bit. > > > Can I at least expect to be able to save code in my local Monticello repository ? It bombs. I just reported a bug! > > Can I expect the Windows installer for Pharo Launcher to work on Windows? It doesn't : it crashes. > > Can I expect Pharo Launcher (not installed from the .exe installer) to work on Windows? It just doesn't. I get all kinds of error related to WindowsStore, bad UTF-8 encoding and other things... (see https://twitter.com/smalltalkdev/status/978973863332798464 <https://twitter.com/smalltalkdev/status/978973863332798464>) > > Can I expect that when looking for implementors of a method things still work? I get duplicates (see attached file). > > Can I expect reading input from the console to work on Windows? It doesn't. > > Can I expect to be able to test my stuff on Windows? That's kinda hard when Windows is always 6 months to 1 year behind VM-wise. > > Can I expect to be able to contribute ? That is kinda sad but all the cool scripts out there are .sh files. > > I'm not whining about the effort people put into this project : I'm mostly whining about duplication of effort and code and *very poor* Windows support... > > Just spend 2 weeks working with Pharo on Windows. You can start with version 5.1 if you want and let me know how it goes... Brace yourself, I'm telling you, you're off for a wild & unpleasant ride. > > > > > > > ----------------- > BenoÃ®t St-Jean > Yahoo! Messenger: bstjean > Twitter: @BenLeChialeux > Pinterest: benoitstjean > Instagram: Chef_Benito > IRC: lamneth > Blogue: endormitoire.wordpress.com > "A standpoint is an intellectual horizon of radius zero". (A. Einstein) > <implementors.JPG>

April 13, 2018

Re: [Pharo-users] Where do we go now ?

by Benoit St-Jean

"You cannot take an alpha version and expect production quality." I'm on Windows 10, using Pharo 7.0 alpha 32 bit. Can I at least expect to be able to save code in my local Monticello repository ? It bombs. I just reported a bug! Can I expect the Windows installer for Pharo Launcher to work on Windows?Â It doesn't : it crashes.Â Can I expect Pharo Launcher (not installed from the .exe installer) to work on Windows?Â It just doesn't. I get all kinds of error related to WindowsStore, bad UTF-8 encoding and other things... (see https://twitter.com/smalltalkdev/status/978973863332798464) Can I expect that when looking for implementors of a method things still work?Â I get duplicates (see attached file). Can I expect reading input from the console to work on Windows?Â It doesn't. Can I expect to be able to test my stuff on Windows?Â That's kinda hard when Windows is always 6 months to 1 year behind VM-wise. Can I expect to be able to contribute ?Â That is kinda sad but all the cool scripts out there are .sh files. I'm not whining about the effort people put into this project : I'm mostly whining about duplication of effort and code and *very poor* Windows support... Just spend 2 weeks working with Pharo on Windows.Â You can start with version 5.1 if you want and let me know how it goes...Â Brace yourself, I'm telling you, you're off for a wild & unpleasant ride. ----------------- BenoÃ®t St-Jean Yahoo! Messenger: bstjean Twitter: @BenLeChialeux Pinterest: benoitstjean Instagram: Chef_Benito IRC: lamneth Blogue: endormitoire.wordpress.com "A standpoint is an intellectual horizon of radius zero".Â (A. Einstein)

April 13, 2018

Re: [Pharo-users] Where do we go now ?

by Sven Van Caekenberghe

> On 13 Apr 2018, at 08:43, phil(a)highoctane.be wrote: > > I consider Pharo 7 as a great piece of kit but unusable for my current work. There are many new things to learn in there. When is too much too much? Also, simplifications are breaking things in unexpected ways (like the #atEnd thing). Phil, Nothing fundamental will break with #atEnd. What you are reading in pharo-dev is a constructive discussion that (for me at least) started with the desire to support one very special kind of stream (stdin in C terms), something 99.99% of Pharo users have never seen, used or heard of. Zn streams have worked well and as expected for Pharo versions going back to 3, that won't change. Sven

April 13, 2018

Re: [Pharo-users] Where do we go now ?

by Pavel Krivanek

Moreover, we are increasing the system modularity so more classes in the IDE does not mean that you need to have them all loaded. -- Pavel 2018-04-13 10:39 GMT+02:00 Esteban Lorenzano <estebanlm(a)gmail.com>: > As a general note, I have to say that I find amusing how people are often > annoyed about the growing size of the image and the amount of classes. > We have an image that contains a full IDE which incorpores more and more > modern capabilities. And you want that to stay small and fewer classes? Iâm > sorry but you are missing the point. More tools (or same tools enhanced) > means more size and classes. > > Thenâ¦ why the fear about classes? This is OOP. Doing classes is what we > do. > I prefer a new class to, for example passing same information into a > dictionary. > I prefer a new class over a string. > I prefer a new class over a tuple. > > We now have slots instead symbols to express variables. I dream with even > more classes: Selector, Protocol (we already have it, but we need to use > it), etc., etc., etc. > I will always prefer a class that defines a concept than a non-reified > usage of generic classes. > > We need better ways to organise that information? YES! (For example I was > thinking browser by default should not show âtool packagesâ, to reduce the > amount of information). > > But I welcome more classes and less esoteric usages of existing ones, > always. > > cheers, > Esteban > > On 13 Apr 2018, at 10:30, Esteban Lorenzano <estebanlm(a)gmail.com> wrote: > > > > On 13 Apr 2018, at 10:25, Ben Coman <btc(a)openinworld.com> wrote: > > On 13 April 2018 at 13:53, Benoit St-Jean via Pharo-users < > pharo-users(a)lists.pharo.org> wrote: > >> Just a quick word to get some things straight because, quite frankly, I >> really don't know where we're heading. >> > > >> Do we really need 8 delay schedulers (DelayMicrosecondScheduler, >> DelayMillisecondScheduler, DelayNullScheduler, >> DelayExperimentalSpinScheduler, DelaySpinScheduler, >> DelayTicklessScheduler, DelayExperimentalCourageousScheduler, >> DelayExperimentalSemaphoreScheduler) ? >> > > This should have been cleaned up a while ago. Thanks for the bump. I'll > get to it. > However note, that half of those have not more than two methods, so are > not contributing much to the size of the Image. > > > I didnât know. Thanks :) > > > >> Perhaps a little bit of clarity and details about what's coming and >> what's the plan would be beneficial to a lot of us. >> > > I guess you are looking for... > https://github.com/pharo-project/pharo-workingRoadmaps/ > blob/master/Pharo7/ROADMAP.md > > > cheers -ben > > > > > >

April 13, 2018

Re: [Pharo-users] Where do we go now ?

by Esteban Lorenzano

> On 13 Apr 2018, at 10:31, Cyril Ferlicot D. <cyril.ferlicot(a)gmail.com> wrote: > > On 13/04/2018 07:53, Benoit St-Jean via Pharo-users wrote: > > Hi, > > I don't have time to answer to everything but I can give some details: > > - Nautilus will be removed when Calypso will be really stable (During > the Calypso development if calypso breaks, you need an other browser to > replace it) > - The old compiler will be removed as an external project > (https://github.com/pharo-project/pharo-workingRoadmaps/blob/master/Pharo7/R…) > - I think it's planned to remove the EyeInspector > - Kommiter and Versionner will also be removed in next versions of Pharo > > "Pharo 5.1 had 5885 classes. Pharo 6.1 had 6481 classes. Pharo 7.0 alpha > has 7612 classes." > > In general I do not like metrics in term of number of classes because I > prefer a system with 7500 classes of 50 lines of code than a system of > 6000 classes of 300 lines of codes. amen! > > "Pharo 5.1 had 416 preference settings. Pharo 6.1 had 494 preference > settings. Pharo 7.0 alpha has 662 preference settings." > > The number of settings is not a bad thing if it's well organized. The > current problem is not the number of settings, it's more their > organization. > > Have a nice day. :) > > -- > Cyril Ferlicot > https://ferlicot.fr >

April 13, 2018

Re: [Pharo-users] Where do we go now ?

by Esteban Lorenzano

As a general note, I have to say that I find amusing how people are often annoyed about the growing size of the image and the amount of classes. We have an image that contains a full IDE which incorpores more and more modern capabilities. And you want that to stay small and fewer classes? Iâm sorry but you are missing the point. More tools (or same tools enhanced) means more size and classes. Thenâ¦ why the fear about classes? This is OOP. Doing classes is what we do. I prefer a new class to, for example passing same information into a dictionary. I prefer a new class over a string. I prefer a new class over a tuple. We now have slots instead symbols to express variables. I dream with even more classes: Selector, Protocol (we already have it, but we need to use it), etc., etc., etc. I will always prefer a class that defines a concept than a non-reified usage of generic classes. We need better ways to organise that information? YES! (For example I was thinking browser by default should not show âtool packagesâ, to reduce the amount of information). But I welcome more classes and less esoteric usages of existing ones, always. cheers, Esteban > On 13 Apr 2018, at 10:30, Esteban Lorenzano <estebanlm(a)gmail.com> wrote: > > > >> On 13 Apr 2018, at 10:25, Ben Coman <btc(a)openinworld.com <mailto:btc@openinworld.com>> wrote: >> >> On 13 April 2018 at 13:53, Benoit St-Jean via Pharo-users <pharo-users(a)lists.pharo.org <mailto:pharo-users@lists.pharo.org>> wrote: >> Just a quick word to get some things straight because, quite frankly, I really don't know where we're heading. >> >> Do we really need 8 delay schedulers (DelayMicrosecondScheduler, DelayMillisecondScheduler, DelayNullScheduler, DelayExperimentalSpinScheduler, DelaySpinScheduler, DelayTicklessScheduler, DelayExperimentalCourageousScheduler, DelayExperimentalSemaphoreScheduler) ? >> >> This should have been cleaned up a while ago. Thanks for the bump. I'll get to it. >> However note, that half of those have not more than two methods, so are not contributing much to the size of the Image. > > I didnât know. Thanks :) > >> >> Perhaps a little bit of clarity and details about what's coming and what's the plan would be beneficial to a lot of us. >> >> I guess you are looking for... >> https://github.com/pharo-project/pharo-workingRoadmaps/blob/master/Pharo7/R… <https://github.com/pharo-project/pharo-workingRoadmaps/blob/master/Pharo7/R…> >> >> >> cheers -ben >> >> >> >

April 13, 2018

Re: [Pharo-users] Where do we go now ?

by Cyril Ferlicot D.

On 13/04/2018 07:53, Benoit St-Jean via Pharo-users wrote: Hi, I don't have time to answer to everything but I can give some details: - Nautilus will be removed when Calypso will be really stable (During the Calypso development if calypso breaks, you need an other browser to replace it) - The old compiler will be removed as an external project (https://github.com/pharo-project/pharo-workingRoadmaps/blob/master/Pharo7/R…) - I think it's planned to remove the EyeInspector - Kommiter and Versionner will also be removed in next versions of Pharo "Pharo 5.1 had 5885 classes. Pharo 6.1 had 6481 classes. Pharo 7.0 alpha has 7612 classes." In general I do not like metrics in term of number of classes because I prefer a system with 7500 classes of 50 lines of code than a system of 6000 classes of 300 lines of codes. "Pharo 5.1 had 416 preference settings. Pharo 6.1 had 494 preference settings. Pharo 7.0 alpha has 662 preference settings." The number of settings is not a bad thing if it's well organized. The current problem is not the number of settings, it's more their organization. Have a nice day. :) -- Cyril Ferlicot https://ferlicot.fr

April 13, 2018