September 2012 - Pharo-dev - lists.pharo.org

Re: [Pharo-project] Bad behaviour at startup

by StÃ©phane Ducasse

Thanks Sven. I created an issue. BTW esteban was making progress on a simpler integration process and it will easier for non core people to commit :). Stef On Sep 29, 2012, at 9:02 PM, Sven Van Caekenberghe wrote: > Name: Zinc-HTTP-SvenVanCaekenberghe.307 > Author: SvenVanCaekenberghe > Time: 29 September 2012, 9:00:49.72 pm > UUID: 9387492b-daa2-4ebf-a49d-44950586bda3 > Ancestors: Zinc-HTTP-SvenVanCaekenberghe.306 > > Modified ZnServer class>>#startUp: to use a deferred startup action to start all registered servers; this should allow for normal error handling when something goes wrong initializing server sockets (thanks Igor Stasenko for the fix; thanks Denis Kudriashov for the error report)

Sept. 29, 2012

Re: [Pharo-project] Metacello bleedingEdge question

by Sven Van Caekenberghe

On 29 Sep 2012, at 21:38, StÃ©phane Ducasse <stephane.ducasse(a)inria.fr> wrote: > On Sep 29, 2012, at 5:39 PM, Dale Henrichs wrote: > >> Stef, >> >> #latestVersion is there for backward compatibility. > > Ok > I would simply remove it. > > Stef I was about to say: what's that 'backward compatibility' ? ;-) No seriously, unless there is a very good, important reason to keep it, confusing stuff should be removed going forward. Sven -- Sven Van Caekenberghe http://stfx.eu Smalltalk is the Red Pill

Sept. 29, 2012

Re: [Pharo-project] Fuel and Tanker in older Pharo images [WAS] Re: How to try Shapes and Gaucho?

by StÃ©phane Ducasse

> On 29 September 2012 14:57, Fernando Olivero <fernando.olivero(a)usi.ch> wrote: >> Cool! Thanks Mariano for the summary, and all the others for this >> much needed features! Looking forward to adopt such a process, to have >> a minimal Pharo image, customized to my needs. >> > > indeed, sometimes even i forgetting why we doing things we do.. don't worry I'm the compass :) > and then people who are poorly informed about our end goals, making > wrong assumptions > about where Pharo moves, bashing us for killing compatibility, > breaking what was not broken, etc etc :) Because they did not read the vision document :) Stef

Sept. 29, 2012

Re: [Pharo-project] Fuel and Tanker in older Pharo images [WAS] Re: How to try Shapes and Gaucho?

by StÃ©phane Ducasse

Thanks mariano for the history :) This is the vision we have and that is presented in the vision file available in the consortium web page on the pharo-project.org :) > Hi Denis. I want to say a couple of things about this. First, you are not the only one wanting this. In fact, that scenario is one of the reasons we are investing a lot of time in the infrastructure: > > 1) Pavel Krivanek has been working a lot in making the image more modular, being able to cleanly unload and reload packages. This was the first step. > 2) Jannik Laval continue that effort trying to remove lots of cyclic dependencies between packages. > 3) Guillermo Polito started to work in the bootstrap. He is now able to bootstrap images from scratch (contrary to Pavel's PharoKernel where he was unloading packages). > 4) Esteban Lorenzano is working on a new ConfigurationOfPharo that will let you take a kernel/bootstraped/minimal image and load just the packages you want. > 5) Camilo Bruni, Martin Dias, Guillermo Polito, and Toon Verwaest have been working in a new class builder based on layouts. The "old ClassBuilder" is really difficult to maintain, understand, fix, etc. When you load packages, the class builder takes an important role. Therefore, they were working in a new class builder, which, from my point of view, is way better, object-oriented, tested, etc. > 6) Martin Dias and Me were working in Tanker, a tool to export/import packages of code using Fuel without Compiler. > > So.....as you can see there is/was a lot of work going on related to what you need :) It is not 100% ready, but we are getting there. > > Cheers,

Sept. 29, 2012

Re: [Pharo-project] Metacello bleedingEdge question

by StÃ©phane Ducasse

On Sep 29, 2012, at 5:39 PM, Dale Henrichs wrote: > Stef, > > #latestVersion is there for backward compatibility. Ok I would simply remove it. Stef > > The #stable symbolic version is the preferred mechanism for declaring the "latest version". > > Dale > > ----- Original Message ----- > | From: "StÃ©phane Ducasse" <stephane.ducasse(a)inria.fr> > | To: Pharo-project(a)lists.gforge.inria.fr > | Sent: Friday, September 28, 2012 11:37:30 AM > | Subject: Re: [Pharo-project] Metacello bleedingEdge question > | > | dale > | > | at the end do we really need latestVersion because I'm still confused > | by it. > | > | On Sep 28, 2012, at 3:42 PM, Dale Henrichs wrote: > | > | > #latestVersion is defined as the latest version whose blessing is > | > neither #baseline nor #development. > | > | > | >

Sept. 29, 2012

Re: [Pharo-project] [Seaside] Re: change n same words in browser with one action

by Igor Stasenko

On 29 September 2012 10:17, Camillo Bruni <camillobruni(a)gmail.com> wrote: > On 2012-09-29, at 10:04, Sabine KnÃ¶fel <sabine.knoefel(a)gmail.com> wrote: >> I agree with Igor and want to add, that a newbie will not be disturbed by shortcuts and can use the dialogs. It is the normal way, if you learn a new system, you get to know one thing after the other and someday, you get to know the shortcuts and they help to work faster. >> >> As you can see in the screenshot (left side), the feature is documented. >> At the right side you can see, that it is not implemented ;.) >> >> Perhaps it is possible to implement it in one of the next releases? >> > > I don't think replacements on a textual basis make much sense: > > 1. methods are only a few lines of code, thus limiting the usefulness > of such replacements > 2. besides renaming temporaries, most renames should be applied to > other methods as well. For instance inst-var renames, or method renames.. > > Though I would like to see on-the-fly variable renaming. For instance as > soon as you start typing on the name of the variable definition it will > automatically rewrite all the other accesses to that variable in the current > method. But for that we need a decent AST-based text editor. > > So unless you are a pro-pharo-shortcut user, think about the time you > loose otherwise by point-and-clicking with your mouse ;), a little search > and replace dialog interaction won't add too much :) > no, Cami, with cmd-j you don't need to click and point. you just keep pressing cmd-j till all old occurrences is replaced with one which you over-typed. but i agree, that this feature has limited use in smalltalk code, knowing that most of methods is 1-liner and, i would say, if you need such thing, most of the times it is a sign that you violating DRY principle somehow and should really think how to refactor your code, instead of changing single variable name/expression 10 times. > best > cami -- Best regards, Igor Stasenko.

Sept. 29, 2012

Re: [Pharo-project] [Seaside] Re: change n same words in browser with one action

by Igor Stasenko

On 29 September 2012 07:28, Chris Muller <ma.chris.m(a)gmail.com> wrote: >>> You should not expect that -- Pharo used to have it but it was removed >>> in favor of a more mainstream editor look-and-feel -- so that it be >>> would more attractive to new users. Command+J / Control+J are just >>> two of several powerful code-editing capabilities Pharo originally >>> inherited from Squeak, along with my other favorites Command+E = >>> Exchange the last two selections and Command+[, (, {, or | = Surround >>> the highlighted text with that bracket-type pair. (not sure if those >>> are still there in Pharo). >> >> Now what would be really cool is to have a documentation about all >> those, and make them easily >> discoverable. I worked with Squeak long enough, and didn't saw >> anything close to that. >> I said that it is 'command-g' instead 'command-j' exactly because of >> that, because you never know... >> And for me it was always felt like: >> - let user discover shortcuts by pressing random key combinations and >> see what happens. >> And then tell me, what are the chances that people will understand >> what exactly cmd-j does without >> reading documentation? In 99% of cases it will be 'oh.. what was just happen' .. >> This is about "more attractive". >> And if you hiding knowledge behind sealed doors, don't be surprised >> that someone will come and build a new library with open doors, >> replacing yours. > > You didn't look hard enough -- it's under the "help" menu. Squeak 3.0 > is the earliest I have on my HD but.. yes, it was there then and its > still there now. Maybe i didn't look hard enough (well actually i did, cause i remember looking for the code which implements/introduces them).. Still since you cannot remember all shortcuts at once , unless they are catching your eye (in context menus/elsewhere), you tend to forget/not using them. What i can say about Pharo 2.0 today, that yes, a new shortcut is a bit alienating (because many you need to learn from scratch), but in other turn they are much easier to discover. And if you try Nautilus, you can figure, that you can do everything without even touching a mouse. I can hardly characterize this as "newbie-friendly"/anti-poweruser approach, but IMO it is much more friendly to power-users.. once you learn shortcuts, of course. Still, this is a work-in-progress, but i like the way guys doing it.. we'll see the end result once dust settles :) > Ok, I know you just forgot we miss you too Igor. ;) > but i cannot spend my time on everything, everywhere. I need to focus on stuff i do :) > I admire Pharo blazing new trails but notice a pattern in some of the > "gold" feedback from Pharo customers which worth taking to heart. > Like: > > http://lists.gforge.inria.fr/pipermail/pharo-project/2012-May/064702.html > What can i say, it is of course not nice when you breaking old habits, unless you have a good reasons for doing so, and when you don't provide alternative(s). >> This is, what i think, happens with shortcuts changes in Pharo, except >> form those which was remapped. >> >>> Pharo favors _familiarity_ for new users somewhat at the cost of >>> productivity for its experienced users. It somewhat resembles eclipse >>> -- with pop-up dialogs and a seemingly "modal" philosophy. Squeak, by >>> constrast, has kept pushing the "simple-but-powerful" philosphy of its >>> IDE to near extremes. While it's less familiar for new uesrs, once >>> learned, general navigation and editing require _far fewer_ gestures. >>> >> Another thing, is that i first time hear that pharo is in favor of >> newcomers at expense of experienced users. >> I am not involved in changes to key bindings / shortcuts, but i doubt >> that people, who doing that, making changes having such strange aim. >> I think they are focused on making it good for newcomers and >> experienced users both. Otherwise i don't see much sense in doing this >> at all, and would be first who will be against it. > > Ok, glad to hear that. I would rather have the best system possible > than one with the most mainstream acceptance. i can say you, it has nothing to do with mainstream: it is just new people , who learned things differently, and of course their preferences influencing the choices they do. i , for instance, i a long-time windows user. And many things in squeak were alien to me, like using alt-key instead of control key for most of shortcuts. Still, it doesn't matters as long as they meeting two key criteria: - consistent - easy to learn. -- Best regards, Igor Stasenko.

Sept. 29, 2012

Re: [Pharo-project] Bad behaviour at startup

by Sven Van Caekenberghe

On 29 Sep 2012, at 20:09, Igor Stasenko <siguctua(a)gmail.com> wrote: > In case of Zn server, apparently, it is not absolutely necessary to > bind socket(s) during startup phase, because image will function > properly with or without them. So deferring this action after startup > is finished, is the right way to do it. --- Name: Zinc-HTTP-SvenVanCaekenberghe.307 Author: SvenVanCaekenberghe Time: 29 September 2012, 9:00:49.72 pm UUID: 9387492b-daa2-4ebf-a49d-44950586bda3 Ancestors: Zinc-HTTP-SvenVanCaekenberghe.306 Modified ZnServer class>>#startUp: to use a deferred startup action to start all registered servers; this should allow for normal error handling when something goes wrong initializing server sockets (thanks Igor Stasenko for the fix; thanks Denis Kudriashov for the error report) --- -- Sven Van Caekenberghe http://stfx.eu Smalltalk is the Red Pill

Sept. 29, 2012

Re: [Pharo-project] Fuel and Tanker in older Pharo images [WAS] Re: How to try Shapes and Gaucho?

by Mariano Martinez Peck

On Sat, Sep 29, 2012 at 8:32 PM, Igor Stasenko <siguctua(a)gmail.com> wrote: > On 29 September 2012 14:57, Fernando Olivero <fernando.olivero(a)usi.ch> > wrote: > > Cool! Thanks Mariano for the summary, and all the others for this > > much needed features! Looking forward to adopt such a process, to have > > a minimal Pharo image, customized to my needs. > > > > indeed, sometimes even i forgetting why we doing things we do.. > But we also have the Pharo vision document for that :) https://gforge.inria.fr/frs/download.php/30434/PharoVision.pdf > and then people who are poorly informed about our end goals, making > wrong assumptions > about where Pharo moves, bashing us for killing compatibility, > breaking what was not broken, etc etc > > So, Mariano's summary is good rehearsal of our aims :) > > > Fernando > > -- > Best regards, > Igor Stasenko. > > -- Mariano http://marianopeck.wordpress.com

Sept. 29, 2012

Re: [Pharo-project] Fuel and Tanker in older Pharo images [WAS] Re: How to try Shapes and Gaucho?

by Igor Stasenko

On 29 September 2012 14:57, Fernando Olivero <fernando.olivero(a)usi.ch> wrote: > Cool! Thanks Mariano for the summary, and all the others for this > much needed features! Looking forward to adopt such a process, to have > a minimal Pharo image, customized to my needs. > indeed, sometimes even i forgetting why we doing things we do.. and then people who are poorly informed about our end goals, making wrong assumptions about where Pharo moves, bashing us for killing compatibility, breaking what was not broken, etc etc So, Mariano's summary is good rehearsal of our aims :) > Fernando -- Best regards, Igor Stasenko.

Sept. 29, 2012