May 2011 - Pharo-dev - lists.pharo.org

Re: [Pharo-project] Popularity of Smalltalk in Software Industry

by Igor Stasenko

On 5 May 2011 18:57, Jimmie Houchin <jlhouchin(a)gmail.com> wrote: > On 5/5/2011 10:32 AM, Toon Verwaest wrote: >> >> On 05/05/2011 05:26 PM, CÃ©drick BÃ©ler wrote: >>>> >>>> Lastly, most Smalltalk systems are image based... >>> >>> ...which makes you feel the system is "alive", hence one **huge benefit** >>> of Smalltalk: its debugger which enables on the fly debbuging... and also >>> test driven development (real one [1]) where you can run incomplete code and >>> code what's missing iteratively when you need it (Smalltalk is a live >>> system, not only a language as somebody said lately). >>> >>> CÃ©drick >>> >>> [1] see in particular this webcast: >>> http://www.pharocasts.com/2010/01/starting-with-sunit-and-debugger.html >> >> Seriously ... these points in favor of the image are so m00t. Lets see how >> it would work without an image: >> >> I write a C application which I link to GCC. Now I run GDB on my >> application, and while running I have the whole GCC compiler collection at >> my disposal while running. While debugging (at some breakpoint) I just let >> the GCC library compile some C code for me; I turn on the executable flag >> and whooptidoo, I have a Smalltalk like debugger for C. >> >> This is totally unrelated to having an image; it's just a great debugger >> implementation. 2 completely different things. No reason why this wouldn't >> work for C; except for the fact that they didn't do it yet (those lazy >> bastards). > > Disclaimer, I am not a professional programmer and I have not programmed in > C or any other static compiled language at all. So I may not have any idea > what I am talking about. :) > > What I don't see in your example, but what I experience in my programming is > often in the *use* of my application during development. Part of my > application is written in Python (so I can interface I library I can't from > Pharo) and part in Pharo. Â (*use*, ie: Â the application is compiled and > ready for a *user* and I am the user and encounter the bug. > > My Python app connects to a server on the net and downloads data and inserts > it into a PostgreSQL database. It also provides live data to my Pharo app > via an http interface. > > While experimenting with NumPy and running some statistical analysis on a GB > or two of data. The analysis took 10-12 hours to run. I start it up and walk > away. I get back to it only to find that at the end of all the analysis it > exited with a stacktrace. Ugh! Â I concatenated a string using a ',' instead > of a '+'. Â Switching between Pharo and Python bit me. :) Â This was at the > very end of all the analysis and the generation of a report. The last line > of the code had the bug. I lost the 12hours of analysis. This bug passed the > syntax checker and the Python compiler when I imported the module. It wasn't > until it was live and running and encountered in execution that it was > discovered. > > In Pharo, the same situation, I fix it in the debugger and my report is > generated and I continue. All my data is still resident in the image and I > can explore and continue. In Python (and I presume most any other dead > language) all is lost and I have to rerun. > > Now as I said I am not a professional. Certain processes would have helped > in this situation, but not necessarily all. And I have limited experience > with the best other systems or languages have to offer. > > But how many programming languages/environments handle that well? > > I am still so very tired of every time an application on my computer says, > "You just updated. Do you want to restart?" > > Why do most every single web application outside of the Smalltalk world > require shutdowns and restarts for bug fixes and upgrades? > > Why does NetBeans consume 200mb, take 3-5minutes to startup, and do a much, > much poorer job with syntax, autocomplete, etc. on a single 500 line Python > file? > > I don't think the world out there is anywhere near what Smalltalk has. Nor > do they have the vision for it. They are happy with files, and restarts and > all their processes to reduce their pain due to not having a live > environment. > > I am a single guy with a vision, an idea, that I want to implement for my > business. I have seen nothing that gives me the productivity that Smalltalk > does. Not Java, Python, Clojure, Scala, Lua, etc. > > I have never seen any other language accomplish so much with so few. I look > at other languages and apps built with them and look at the armies they have > to do so. Then I look at Seaside, etc. with just a small group of guys with > an idea of something better. > > While I desperately want Squeak/Pharo to be able to interface outside > systems better. I have seen nothing that is better or comes close. And as I > said, I have a business requirement to interface a proprietary library > written in C, which why I am connected with Python (and Windows) against my > will. So I do understand. > > I understand, as does everybody else here, that Pharo/Smalltalk is not > perfect and has huge areas where we have not done things well or right. But > I don't think this is one of them. I think this is one where the rest of the > world is behind us. > > Squeak/Pharo/Smalltalk is empowering. > Well said. Except that i'm not sharing your view that its hard to interface with foreign libraries. Its not hard at all. Of course to connect two different worlds, you need to have knowledge in both of them. But this requirement not a bit different when you take any other pair of languages and try to connect them. > JMHO, take with a grain of salt. :) > > Jimmie > > -- Best regards, Igor Stasenko AKA sig.

May 5, 2011

Re: [Pharo-project] Popularity of Smalltalk in Software Industry

by Igor Stasenko

On 5 May 2011 17:32, Toon Verwaest <toon.verwaest(a)gmail.com> wrote: > On 05/05/2011 05:26 PM, CÃ©drick BÃ©ler wrote: >> >>> Lastly, most Smalltalk systems are image based... >> >> ...which makes you feel the system is "alive", hence one **huge benefit** >> of Smalltalk: its debugger which enables on the fly debbuging... and also >> test driven development (real one [1]) where you can run incomplete code and >> code what's missing iteratively when you need it (Smalltalk is a live >> system, not only a language as somebody said lately). >> >> CÃ©drick >> >> [1] see in particular this webcast: >> http://www.pharocasts.com/2010/01/starting-with-sunit-and-debugger.html > > Seriously ... these points in favor of the image are so m00t. Lets see how > it would work without an image: > > I write a C application which I link to GCC. Now I run GDB on my > application, and while running I have the whole GCC compiler collection at > my disposal while running. While debugging (at some breakpoint) I just let > the GCC library compile some C code for me; I turn on the executable flag > and whooptidoo, I have a Smalltalk like debugger for C. > > This is totally unrelated to having an image; it's just a great debugger > implementation. 2 completely different things. No reason why this wouldn't > work for C; except for the fact that they didn't do it yet (those lazy > bastards). > > Yes, but it affects the way how you design your applications and frameworks. In image-based setup you tend to do things in a way, that things should be always available at any moment (wanna create an instance of me? - feel free to do it). In C and other languages which having compile & run cycle it is totally different. Things are available for you not instantly, but only after you restart everything from scratch, reinitialize everything and then repeat exactly same steps to meet the same point as before. This is often tedious, time consuming and frustrating. And this difference in approaches leads to fatal flaws in design, like on Windows OS, which requires you to reboot machine every time you install new driver(s). -- Best regards, Igor Stasenko AKA sig.

May 5, 2011

Re: [Pharo-project] Popularity of Smalltalk in Software Industry

by Jimmie Houchin

On 5/5/2011 10:32 AM, Toon Verwaest wrote: > On 05/05/2011 05:26 PM, CÃ©drick BÃ©ler wrote: >>> Lastly, most Smalltalk systems are image based... >> >> ...which makes you feel the system is "alive", hence one **huge >> benefit** of Smalltalk: its debugger which enables on the fly >> debbuging... and also test driven development (real one [1]) where >> you can run incomplete code and code what's missing iteratively when >> you need it (Smalltalk is a live system, not only a language as >> somebody said lately). >> >> CÃ©drick >> >> [1] see in particular this webcast: >> http://www.pharocasts.com/2010/01/starting-with-sunit-and-debugger.html > Seriously ... these points in favor of the image are so m00t. Lets see > how it would work without an image: > > I write a C application which I link to GCC. Now I run GDB on my > application, and while running I have the whole GCC compiler > collection at my disposal while running. While debugging (at some > breakpoint) I just let the GCC library compile some C code for me; I > turn on the executable flag and whooptidoo, I have a Smalltalk like > debugger for C. > > This is totally unrelated to having an image; it's just a great > debugger implementation. 2 completely different things. No reason why > this wouldn't work for C; except for the fact that they didn't do it > yet (those lazy bastards). Disclaimer, I am not a professional programmer and I have not programmed in C or any other static compiled language at all. So I may not have any idea what I am talking about. :) What I don't see in your example, but what I experience in my programming is often in the *use* of my application during development. Part of my application is written in Python (so I can interface I library I can't from Pharo) and part in Pharo. (*use*, ie: the application is compiled and ready for a *user* and I am the user and encounter the bug. My Python app connects to a server on the net and downloads data and inserts it into a PostgreSQL database. It also provides live data to my Pharo app via an http interface. While experimenting with NumPy and running some statistical analysis on a GB or two of data. The analysis took 10-12 hours to run. I start it up and walk away. I get back to it only to find that at the end of all the analysis it exited with a stacktrace. Ugh! I concatenated a string using a ',' instead of a '+'. Switching between Pharo and Python bit me. :) This was at the very end of all the analysis and the generation of a report. The last line of the code had the bug. I lost the 12hours of analysis. This bug passed the syntax checker and the Python compiler when I imported the module. It wasn't until it was live and running and encountered in execution that it was discovered. In Pharo, the same situation, I fix it in the debugger and my report is generated and I continue. All my data is still resident in the image and I can explore and continue. In Python (and I presume most any other dead language) all is lost and I have to rerun. Now as I said I am not a professional. Certain processes would have helped in this situation, but not necessarily all. And I have limited experience with the best other systems or languages have to offer. But how many programming languages/environments handle that well? I am still so very tired of every time an application on my computer says, "You just updated. Do you want to restart?" Why do most every single web application outside of the Smalltalk world require shutdowns and restarts for bug fixes and upgrades? Why does NetBeans consume 200mb, take 3-5minutes to startup, and do a much, much poorer job with syntax, autocomplete, etc. on a single 500 line Python file? I don't think the world out there is anywhere near what Smalltalk has. Nor do they have the vision for it. They are happy with files, and restarts and all their processes to reduce their pain due to not having a live environment. I am a single guy with a vision, an idea, that I want to implement for my business. I have seen nothing that gives me the productivity that Smalltalk does. Not Java, Python, Clojure, Scala, Lua, etc. I have never seen any other language accomplish so much with so few. I look at other languages and apps built with them and look at the armies they have to do so. Then I look at Seaside, etc. with just a small group of guys with an idea of something better. While I desperately want Squeak/Pharo to be able to interface outside systems better. I have seen nothing that is better or comes close. And as I said, I have a business requirement to interface a proprietary library written in C, which why I am connected with Python (and Windows) against my will. So I do understand. I understand, as does everybody else here, that Pharo/Smalltalk is not perfect and has huge areas where we have not done things well or right. But I don't think this is one of them. I think this is one where the rest of the world is behind us. Squeak/Pharo/Smalltalk is empowering. JMHO, take with a grain of salt. :) Jimmie

May 5, 2011

Re: [Pharo-project] Making cr to autocomplete as well as tab (in Pharo 1.3)

by Igor Stasenko

On 5 May 2011 18:00, Tudor Girba <tudor.girba(a)gmail.com> wrote: > Hi, > > This is not great at all. Please read this before going forward. > > We had this in OCompletion before and then it got changed to not use CR for a reason. The problem is when using fluent APIs that are best read when written on multiple lines. For example, suppose that I want to type something like this in Glamour: > > ... > Â Â Â Â a tree > Â Â Â Â Â Â Â Â display: [ ... ] > > After typing "a tree", the first proposed item is "treeLayout" (see the attached picture). If I press enter to go to the next line "tree" is replaced with "treeLayout" and this is not the behavior I want. > > The problem is that CR has a useful meaning when typing a piece of text and this makes it not good as a completion character. Tab on the other hand, is only used in Smalltalk at the beginning of an empty line so there is no danger of overloading its functionality in the middle of a text. > This is not a problem for me, because first i typing the expression and then formatting it properly if necessary. Also, in case as you shown, one could simply type 'space' then 'enter' so discard the suggestion and begin from a new line. > > Now, if you insist, I still believe there is a place for CR. Completion has two modes: > 1. one in which I write and the completion offers me something, and > 2. one in which I am after I press the down arrow to select some completion item. > i agree, a completion should be non-intrusive (by default it should handle enter as carriage return, but if you selected an item (using up/down keys), then it should paste the selection instead). But as i said, i having a habit to use enter, when i see a popup menu and hilighted menu item selected instead of tab. > For 1. you do not want to have CR as a completion character. For 2. it is probably ok because you enter explicitly in a temporary mode and thus you are not typing anymore and this will not induce the conflict. > > Cheers, > Doru > > -- Best regards, Igor Stasenko AKA sig.

May 5, 2011

Re: [Pharo-project] Making cr to autocomplete as well as tab (in Pharo 1.3)

by Luc Fabresse

2011/5/5 Tudor Girba <tudor.girba(a)gmail.com> > Now, if you insist, I still believe there is a place for CR. Completion has > two modes: > 1. one in which I write and the completion offers me something, and > 2. one in which I am after I press the down arrow to select some completion > item. > > For 1. you do not want to have CR as a completion character. For 2. it is > probably ok because you enter explicitly in a temporary mode and thus you > are not typing anymore and this will not induce the conflict. > Yes Doru. My problem is always in the second case. Cheers, #Luc > Cheers, > Doru > > > > > > > > On 5 May 2011, at 17:37, Thierry Lebourque wrote: > > > That is really great! > > Many thanks > > > > 2011/5/5 Igor Stasenko <siguctua(a)gmail.com>: > >> Hi, i just found inconvenient that for autocompletion i should press > >> tab all the time > >> while i tend to always hit enter, which instead of pasting suggestion, > >> inserts cr into text. > >> > >> So i hacked a lil piece of code to make autocompletion to insert > >> suggested text when you pressing either tab or cr > >> > >> -- > >> Best regards, > >> Igor Stasenko AKA sig. > >> > > > > -- > www.tudorgirba.com > > "Reasonable is what we are accustomed with." > > >

May 5, 2011

Re: [Pharo-project] new Cog VMs uploaded

by Toon Verwaest

> You have it backwards. There is *no* GPL code in OpenQwaq. If there > had been the whole project would have had to become GPL and Teleplace > could not have protected itself against others using the code. By > releasing the code as GPL Teleplace's investors retain the sole > commercial rights to the code. No-one else can use the code without > their contributions becoming GPL. Hence no-one wanting to do > commercial development using OpenQwaq can do so and protect their > contributions. Teleplace's investors, however, /can/ subsequently > sell or exploit commercial rights to the code, which they still hold. A policy that even Richard Stallman sees as good use of GPL...

May 5, 2011

Re: [Pharo-project] new Cog VMs uploaded

by Eliot Miranda

On Wed, May 4, 2011 at 5:33 PM, Igor Stasenko <siguctua(a)gmail.com> wrote: > On 4 May 2011 23:39, StÃ©phane Ducasse <stephane.ducasse(a)inria.fr> wrote: > > Hi chris > > > > > >> You should be able to extract the code and use it in Pharo/Squeak > >> without issue, even if it is GPL. What you can't do is include it in > >> the distribution, since it isn't MIT. You also shouldn't study the > >> code and write your own version of it - I believe that would be a > >> derivative work, which would likely make it a GPL derivative. > > > > so it does not exist and this is better because GPL is viral and we do > not want it. > > > > I finding it really funny that GPL were invented to help open-source > to rise and spread, > and now its just stands in your way, as any other closed-source > proprietary one... > > >> However, if it was extracted, and you brought it in to do profiling, > >> and then removed it afterwards, that shouldn't be an issue at all. > >> Just remember to remove it after you no longer need it - that way the > >> GPL code won't accidentally creep into the Pharo/Squeak code-base. > >> > >> Basically, this would be a great example of a project that should be > >> an external project and not part of core (or dev). > > > > the problem is that this kind of philosophy goes against the spirit of > smalltalk > > of been able to read and learn the code and modify it. > > > I think that Teleplace has not much choice under which license to > release this code, > because as long as you using even portion of GPL-ed code, you are > forced to use GPL as well. > You have it backwards. There is *no* GPL code in OpenQwaq. If there had been the whole project would have had to become GPL and Teleplace could not have protected itself against others using the code. By releasing the code as GPL Teleplace's investors retain the sole commercial rights to the code. No-one else can use the code without their contributions becoming GPL. Hence no-one wanting to do commercial development using OpenQwaq can do so and protect their contributions. Teleplace's investors, however, /can/ subsequently sell or exploit commercial rights to the code, which they still hold. > > >> > >> -Chris > >> > > > -- > Best regards, > Igor Stasenko AKA sig. > >

May 5, 2011

Re: [Pharo-project] Making cr to autocomplete as well as tab (in Pharo 1.3)

by Tudor Girba

Hi, This is not great at all. Please read this before going forward. We had this in OCompletion before and then it got changed to not use CR for a reason. The problem is when using fluent APIs that are best read when written on multiple lines. For example, suppose that I want to type something like this in Glamour: ... a tree display: [ ... ] After typing "a tree", the first proposed item is "treeLayout" (see the attached picture). If I press enter to go to the next line "tree" is replaced with "treeLayout" and this is not the behavior I want. The problem is that CR has a useful meaning when typing a piece of text and this makes it not good as a completion character. Tab on the other hand, is only used in Smalltalk at the beginning of an empty line so there is no danger of overloading its functionality in the middle of a text. Now, if you insist, I still believe there is a place for CR. Completion has two modes: 1. one in which I write and the completion offers me something, and 2. one in which I am after I press the down arrow to select some completion item. For 1. you do not want to have CR as a completion character. For 2. it is probably ok because you enter explicitly in a temporary mode and thus you are not typing anymore and this will not induce the conflict. Cheers, Doru On 5 May 2011, at 17:37, Thierry Lebourque wrote: > That is really great! > Many thanks > > 2011/5/5 Igor Stasenko <siguctua(a)gmail.com>: >> Hi, i just found inconvenient that for autocompletion i should press >> tab all the time >> while i tend to always hit enter, which instead of pasting suggestion, >> inserts cr into text. >> >> So i hacked a lil piece of code to make autocompletion to insert >> suggested text when you pressing either tab or cr >> >> -- >> Best regards, >> Igor Stasenko AKA sig. >> > -- www.tudorgirba.com "Reasonable is what we are accustomed with."

May 5, 2011

Re: [Pharo-project] [squeak-dev] Personalized systems with Treated

by Miguel CobÃ¡

This workaround is valid although not less a workaround. The problem with those other repositories is that they are for specific uses, as Stef said in other post, the Treated repository is for packages integrated in current Pharo. So, of course you could post some package there but it wont belongs to there really. This workaround it could be better implemented if the package with specific changes will be pushed to some dedicated compatibility repository where everyone can push every package and use it in their install procedures. That would be better. Now, lets remember that for Pharo at lease, backwards compatibility is a "great to have" feature not a "must have at all means" feature. So if sometime, with some release, current or future, a break with the past is to be made, it will be made. That is for sure. There can be a few people that will be affected by this (in this case that they can't install something just like they could yesterday, but hey, this is life, as Markus said, no progress is to be dead already) but the majority won't care/will be happier because it is always better for a few persons to fix some problems than the entire project to stall waiting to be all for everyone and always. The changes of core libraries, (collections, kernel, etc) almost sure be ported to pharo from squeak and if some package release can't work because of a missing library in Squeak not yet in Pharo that won't be a big problem. Next release will work, when Pharo has the fixes. We don't worry, one step a time. And nobody has died by waiting a little for some wanted feature. So in the end, I think that this is a non-issue, just a minor effect of the vision of Pharo that is solved with a little of time and patience. Other thing, we really don't like changesets or gofer scripts for installing anything on Pharo (even if we really like gofer). Why? Because they aren't a package management system. The end result could be the same: the packages are installed in the system, that is granted, but the capability of Metacello of declaring which other configurations it depends on and the one of querying the configuration to see the full list of packages to be installed is something that no gofer script, installer script or SqueakMap feature can handle right now without downloading everything to check. Regards El miÃ©, 04-05-2011 a las 21:35 -0500, Chris Muller escribiÃ³: > To support Magma 1.2, I introduced some changes to OrderedCollection > into the Squeak trunk some time ago. To deliver Magma to Pharo users, > I needed to have these same enhancements in Pharo 1.1 and 1.2. But > they were already released, so how was this solved? > > My first instinct was to deliver a change-set; the classical way to > modify the system. But I know dirty packages are unattractive, and > besides that the only way to achieve one-click load of something > involving a change-set + MC-packages in Pharo is with a SAR - overkill > since Magma is just MC packages; no resources. I really wanted to > release Magma 1.2 to Pharo with just a Gofer script. > > So, I decided to import the same changes I made to Squeak's > OrderedCollection into Pharo's and then save that new version to the > Inbox and adjusted my Gofer script for loading Magma to merge that > version of Collections from repositories: Pharo, Inbox, TreatedInbox. > > I knew they would accept it, thank you Pharo, BUT it's nice to know > that I could actually deliver my application without having to wait > for it to be integrated, or worry even if it didn't get accepted; that > chain of repositories should work immediately and forever, no matter > the outcome of acceptance. > > Thanks to MC merging, this is as unintrusive as it can be; in case > someone runs with their own "personalizations" to the Collections > package, loading Magma won't clobber them; nor will updating with > official Pharo-released patches, because those are merged too. > > With the context of this experience, I came to view the "Treated" > repository with a new respect; enabling a non-core developer to > personalize system packages for my application.. > > - Chris > -- Miguel CobÃ¡ http://twitter.com/MiguelCobaMtz http://miguel.leugim.com.mx

May 5, 2011

Re: [Pharo-project] Popularity of Smalltalk in Software Industry

by Toon Verwaest

On 05/05/2011 05:46 PM, Dave Mason wrote: > On May 5, 2011, at 11:32, Toon Verwaest wrote: > >> This is totally unrelated to having an image; it's just a great debugger implementation. 2 completely different things. No reason why this wouldn't work for C; except for the fact that they didn't do it yet (those lazy bastards). > ... and that it's an order of magnitude more difficult! > > Yes, there are certainly other aspects of being image-based that are way cool, but iterative code development and debugging is a big part of it. > > (I am teaching a Java course right now with BlueJ, which is a cool environment (at least if you're not used to an image-based system like Smalltalk). It provides a little UML diagram of your classes, and you can create an object by clicking on the appropriate class, and inspect, and call methods. Objects that you create appear on a "workbench". So I had build up a little example with a few objects, found a bug, and edited the class, hit "compile" and *poof* empty workbench. Yes, in Java they probably could have emulated the image, but it's a lot of work!) > > ../Dave It's more difficult because it's not integrated yet. Suppose that this was already integrated within GDB? Suppose GDB looked like a Smalltalk debugger? Is there anything technical that stands in your way? C is annoying for many other reasons, but not for that one.

May 5, 2011