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

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

by Esteban A. Maringolo

2014-04-28 23:48 GMT-03:00 Sean P. DeNigris <sean(a)clipperadams.com>: > Esteban A. Maringolo wrote >> Plays well with "choose your favorite text editor" (Sublime, Vim, >> etc.) and IDEs (RubyMine, etc.), with source control systems (any file >> based system), with unix in general (several cli commands), has >> binding for any major/mainstream library* (databases, network, etc.). > > But again these boil down to community size/interest > - To use "your favorite text editor", Craig Latta serves Smalltalk via > WebDav [1], but who has jumped at this opportunity? > - source control - now that there is community interest, progress on git > support has been moving ahead rapidly with minimal resources This is my point, who uses WebDAV as its file support? It is a "workaround" to enable file based development. A VERY CLEVER one, but still. > - unix in general - with FFI and OSProcess, what can't you do? Are we > talking about the lack of cool Ruby backtick syntax? While definitely cool, > that special-purpose syntax is the kind of cognitive load Smalltalk > overcomes. All those little syntactical twists and turns to remember lead > away from "syntax on a T-shirt" to manuals with hundreds of pages I'm not talking about syntax, and I wouldn't trade Smalltalk syntax for anything else. FFI and OSProcess IS NOT unix interoperability, are ways to get "out" of the image (the island in Byte's '81 cover). Think about this... Windows and Unix/Linux culture. One started from an "all windows" architecture, whilst the other was the other way around. Windows is still thriving to get a command line culture for many of its products. Only in the last years they've been adding more and more features through command line, because of the demand of scripting and who knows what else. And maybe that "command line" need comes from people coming from the unix world. And, to me, this is "playing well with others", not being able to call "external" programs via any mechanism. I think that one big part of the success of Mac notebooks in the developer community, is the fact is has an underlying unix. Oversimplified example: You don't need a version of gzip with bindings to tar in order to have both working together. > - bindings - again, obviously just a question of community size and interest Here I agree. That's why I asked if you would recommend Pharo today to somebody who has current needs. > So the "play well with others" is a self-fulfilling prophecy. There are no > bindings because there are no people to write them because there are no > bindings... I have to agree with you about the self-fulfilling prophecy. > At inception, Ruby (and every other language) didn't have those bindings either. Ruby got to be what it is because of Rails. Because of 37signals. Pharo doesn't have that. It could have happened with Seaside. But it didn't. It worries me that when an outsider pinpoints so clearly what are our weak points we can't think outside of our box. Proportions aside, it remembers me when you talk with a 100% minded PHP programmer about something, and they can only think inside of their box, and instead reply that you can do OO in PHP and you don't need an interactive debugger.:) Having all that said, I still bet on Pharo. :) Regards, Esteban A. Maringolo

April 29, 2014

Re: [Pharo-dev] [Pharo-users] [Pharo3] New Sources file starting 30841/30842

by Norbert Hartl

Am 29.04.2014 um 14:19 schrieb Sven Van Caekenberghe <sven(a)stfx.eu>: > Tested on Mac OS X: > > $ mkdir pharo3 > $ cd pharo3/ > $ curl get.pharo.org/vm | bash > $ curl get.pharo.org/30 | bash > $ ./pharo-ui Pharo.image > Just a side note. You know there is a get.pharo.org/30+vm ? Norbert > Loaded some configurations, ran some tests, recompiled all. All OK. > > Super ! > > On 29 Apr 2014, at 12:53, Marcus Denker <marcus.denker(a)inria.fr> wrote: > >> Hi, >> >> We have a sources file for Pharo3! >> >> http://files.pharo.org/sources/PharoV30.sources.zip >> http://files.pharo.org/sources/PharoV30.sources >> >> >> The image that goes with it is 30 842 or newer. >> zipped image is now 5MB smaller due to the empty changes. >> >> http://files.pharo.org/image/30/ >> >> this now needs some careful testingâ¦ the build is green >> >> https://ci.inria.fr/pharo/ >> >> >> >> >> >> Marcus >> >> >> >> > >

April 29, 2014

Re: [Pharo-dev] Amber-Pharo based coding dojo in a coworking space

by kilon alios

I was talking mainly about the IDE and other areas where Pharo and Amber overlap. I think Amber is awesome for what it is and should be kept developed independently. "If your ass is being kicked by AngularJS <https://angularjs.org/> or Meteor<https://www.meteor.com/> because you reach the goal faster using those, your next contract is not to be powered by Amber." is this not the cases if you cannot beat them join them ? Do you seriously believe Amber or Pharo can compete AngularJS , correct me if I am wrong, when behind this framework is Google itself ? "In regard to Pharo, becoming Amber friendly is âattracting javascript people to the smalltalk worldâ are you joking? is the best marketing strategy Pharo can dream about!" Did I said anything against that ? I agree that is great. "With some add ons Amber can do the things Meteor does in the front end." Thats all great but again I ask why not leverage the power of Meteor.js for Amber and Pharo in the first place and sprinkle it with some smalltalk libraries to bring it closer to Smalltalk philosophy ? Isnt this the whole point of doing OOP, to reuse code ? Why should it be Amber / Pharo vs the Others ? Instead as "Amber a way to make your favorite javascript libraries even easier and more accessible / dynamic" Do you seriously believe people will give JS for Amber alternatives ? No thats not a best marketing strategy.

April 29, 2014

Re: [Pharo-dev] Amber-Pharo based coding dojo in a coworking space

by Sebastian Sastre

On Apr 29, 2014, at 9:35 AM, kilon alios <kilon.alios(a)gmail.com> wrote: > I think that Amber and Pharo should be merged, maybe not as code bases or not as projects or websites or whatever but as communities. The way I see it both work towards the same goal, making the Smalltalk voice heard loud and clear. I think a closer integration is for the benefit of both communities. The communities are too small to afford to spread thin. > > Inventing the future is all cool and great but this world runs on money , not on progress and innovation. In most cases money is a big obstacle to progress. > > "I love progress , I just dont like change" (sorry dont remember where I got the quote from) Well Iâd keep Amber free to use other backends. Specially node. But you bring very important points there. If your ass is being kicked by AngularJS or Meteor because you reach the goal faster using those, your next contract is not to be powered by Amber. So do fucking something about that, right? Okay.â¦ In regard to Pharo, becoming Amber friendly is âattracting javascript people to the smalltalk worldâ are you joking? is the best marketing strategy Pharo can dream about! With some add ons Amber can do the things Meteor does in the front end. With some add ons, Pharo can also do the backend things that Meteor does in its backend. Meteor is the new cool kid in the block and is proven itself worth using. With a little code in the back and in the front we can catch up with Meteor (which be honest! is currently kicking our ass big time).

April 29, 2014

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

by Esteban A. Maringolo

2014-04-29 4:42 GMT-03:00 Frank Shearar <frank.shearar(a)gmail.com>: > On 29 April 2014 03:48, Sean P. DeNigris <sean(a)clipperadams.com> wrote: >> So the "play well with others" is a self-fulfilling prophecy. There are no >> bindings because there are no people to write them because there are no >> bindings... At inception, Ruby (and every other language) didn't have those >> bindings either. > > That's largely true, in the sense that someone needs to grab a shovel > to dig that trench. > > But if you start out with an external text editor, with an external > version control system, with (only) stdout/stderr/stdin, you end up > building a different system than if you're already in an insular > environment and want/need to learn to "play well with others". This is what I meant by built from the ground (cli) up. And sticking with unix philosophy. > Ruby plays well with other - interfaces well with external systems - > precisely because it didn't have that integrated environment. It NEEDS external tools, because there isn't a "Ruby" toolkit (as with any smalltalk distro). The concept of IDE is clearly separated from the language and vm. Of course it has its drawbacks. Regards!

April 29, 2014

Re: [Pharo-dev] Amber-Pharo based coding dojo in a coworking space

by kilon alios

People introduced to amber is highly unlikely to be aware of Pharo but that does not change the fact that if you as Amber tell this coder "hey I also offer a very powerful IDE" instead of "hey I also offer a very basic IDE" it will be obviously a big plus for him. I just dont think diverting resources to creating an IDE for Amber is optimal. Anymore than Pharo ignoring Amber and creating its own complex solution to access Javascript libraries would be optimal. I think that Amber and Pharo should be merged, maybe not as code bases or not as projects or websites or whatever but as communities. The way I see it both work towards the same goal, making the Smalltalk voice heard loud and clear. I think a closer integration is for the benefit of both communities. The communities are too small to afford to spread thin. Inventing the future is all cool and great but this world runs on money , not on progress and innovation. In most cases money is a big obstacle to progress. "I love progress , I just dont like change" (sorry dont remember where I got the quote from) So I think it would be wiser to be practical and take advantage of what you already have and take it one step further at a time. On Tue, Apr 29, 2014 at 2:56 PM, Sebastian Sastre < sebastian(a)flowingconcept.com> wrote: > > On Apr 29, 2014, at 5:58 AM, Markus Fritsche <mfritsche(a)reauktion.de> > wrote: > > > 1. The most kick ass thing I can think of would go something like this... > > > The most "kick ass thing" I can think of is... > > - Have a Pharo World being interfaced to an amber frontend, so you can use > your image through your browser. > > ... on second thought... that might be cool but after five minutes it'll > make me think "and now?" > > > Yes, what now? > > I agree the other way around is way more attractive. Both ways is awesome. > > But even that is invisible for most people. > > What they need is to look like ass kickers in the face of their clients: *people > that buys the experience of the app not how easily hard they debug*. > > In the other hand, if they can debug better and use this tools to get to > the goal faster, that counts as ass kicking sure. > > What is happening now is that theyâre using the javascript universe of > packages and make the right salad. > > There is a market opportunity in designers that shyly try to get something > non-static done for their clients. > > We should be listening to these guys. > > Using wordpress and adding plugins is profitable but it has limits and > they know that. > > So they are using the first thing thatâs elegant and work with a conf or > two for a start. > > Express js is a good fit for them because is easy for them to grasp the > basics > > What I see is the opportunity of thinking beyond engineers and computer > scientists and being an option for designers. > > What I donât see is us being prepared or caring about that huuuuuuuuuuuge, > even gigantic market. > > If we keep being reactive and defensive on this we will distance ourselves > of: > > Design Principles Behind Smalltalk > The purpose of the Smalltalk project is to provide computer support for > the creative spirit in *everyone*. > > > Letâs say eveyone is too much. Lets strategically focus in the two > audiences that are influential the most: > 1. computer scientists > 2. designers > > They are the innovators and disruptors. > > They are inventing the future > > In which ways are we making their lives easier? (or staying at the margin > from that) > >

April 29, 2014

Re: [Pharo-dev] [Pharo-users] [Pharo3] New Sources file starting 30841/30842

by Sven Van Caekenberghe

Tested on Mac OS X: $ mkdir pharo3 $ cd pharo3/ $ curl get.pharo.org/vm | bash $ curl get.pharo.org/30 | bash $ ./pharo-ui Pharo.image Loaded some configurations, ran some tests, recompiled all. All OK. Super ! On 29 Apr 2014, at 12:53, Marcus Denker <marcus.denker(a)inria.fr> wrote: > Hi, > > We have a sources file for Pharo3! > > http://files.pharo.org/sources/PharoV30.sources.zip > http://files.pharo.org/sources/PharoV30.sources > > > The image that goes with it is 30 842 or newer. > zipped image is now 5MB smaller due to the empty changes. > > http://files.pharo.org/image/30/ > > this now needs some careful testingâ¦ the build is green > > https://ci.inria.fr/pharo/ > > > > > > Marcus > > > >

April 29, 2014

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

by Sean P. DeNigris

S Krish wrote > labelling ( falsely ), that will give no impact / or add any value to the > perception of people. This is the crux of it. It perfectly illustrates two misunderstandings: 1. The idea that "Smalltalk-inspired" is "false". This is a classic "Blind men and an elephant" problem [1]. As I explained in my OP, Smalltalk is an overloaded term (ST-80 vs. continually evolving dynabook software), so "Smalltalk" is as "false" to the 99.9% of developers as "Smalltalk-inspired" is "false" to us 0.1%. But neither is false. Each tells a different part of the same story to a different audience. I'm describing the elephant's tail, and you're feeling the trunk, arguing that "that's not what an elephant is" ;) 2. That the exact words aren't that important. If we look at the $87 billion spent globally on advertising and marketing [2], and the U.S. political system, we see that money and power disagrees. And, Esteban just gave a real world example of how calling Pharo "Smalltalk" has a very real negative impact. I've made these arguments a few times as we've discussed this topic. Even though it doesn't seem that important, I've taken the time because we are a budding community and it seems extra important to be united. At the same time, we've gone around and around with this. And previous responses after I've made what I consider IMHO to be a logical case, have been something like, "yeah but it's a lie. It *is* Smalltalk!" which doesn't speak at all to my two main points. And, I may be wrong!!! And if I am, I want to know! So will someone who believes "we shouldn't call it Smalltalk-inspired" do me the great honor of refuting the above instead of stating that "Smalltalk-inspired" is a false statement, which I've addressed in point #1. Thanks ;-P [1] http://en.wikipedia.org/wiki/Blind_men_and_an_elephant [2] http://www.reportlinker.com/ci02379/Advertising-and-Marketing.html ----- Cheers, Sean -- View this message in context: http://forum.world.st/a-Pharo-talk-from-a-ruby-conference-tp4756805p4756995… Sent from the Pharo Smalltalk Developers mailing list archive at Nabble.com.

April 29, 2014

Re: [Pharo-dev] Amber-Pharo based coding dojo in a coworking space

by Sebastian Sastre

On Apr 29, 2014, at 5:58 AM, Markus Fritsche <mfritsche(a)reauktion.de> wrote: >> >> 1. The most kick ass thing I can think of would go something like this... >> > The most "kick ass thing" I can think of is... > > - Have a Pharo World being interfaced to an amber frontend, so you can use your image through your browser. > > ... on second thought... that might be cool but after five minutes it'll make me think "and now?" > > Yes, what now? I agree the other way around is way more attractive. Both ways is awesome. But even that is invisible for most people. What they need is to look like ass kickers in the face of their clients: people that buys the experience of the app not how easily hard they debug. In the other hand, if they can debug better and use this tools to get to the goal faster, that counts as ass kicking sure. What is happening now is that theyâre using the javascript universe of packages and make the right salad. There is a market opportunity in designers that shyly try to get something non-static done for their clients. We should be listening to these guys. Using wordpress and adding plugins is profitable but it has limits and they know that. So they are using the first thing thatâs elegant and work with a conf or two for a start. Express js is a good fit for them because is easy for them to grasp the basics What I see is the opportunity of thinking beyond engineers and computer scientists and being an option for designers. What I donât see is us being prepared or caring about that huuuuuuuuuuuge, even gigantic market. If we keep being reactive and defensive on this we will distance ourselves of: Design Principles Behind Smalltalk The purpose of the Smalltalk project is to provide computer support for the creative spirit in everyone. Letâs say eveyone is too much. Lets strategically focus in the two audiences that are influential the most: 1. computer scientists 2. designers They are the innovators and disruptors. They are inventing the future In which ways are we making their lives easier? (or staying at the margin from that)

April 29, 2014

[Pharo3] Size of image and sources...

by Marcus Denker

After Condense Changes, writing a new source file and cleaning the image a bit: -> Size of image zip file down to 9MB (from 17). Uncompressed: 21MB -> sources .zip is 5.5 MB. Uncompressed: 26.5MB So not that badâ¦ considering it contains quite a lot new code and still a lot of duplication. Marcus

April 29, 2014