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

Re: [Pharo-project] MongoTalk

by Lucio Santi

Great news! Thanks to everyone. :-) On Fri, Jan 27, 2012 at 9:04 PM, Esteban Lorenzano <estebanlm(a)gmail.com>wrote: > Hi, > > I just uploaded a couple of fixes I made... with those fixes, I have an > application in production, and working really well :) > > best, > Esteban > > El 27/01/2012, a las 8:47p.m., Milan Mimica escribiÃ³: > > AFAIR, the next think you'll have to fix is sending #first to aCharacter. > Just remove the message send and off you go. > > > On 27 January 2012 23:28, Lucio Santi <lukius(a)gmail.com> wrote: > >> Hi! >> >> These days I've been playing around with MongoDB and it was a nice >> surprise to discover the availability of a Smalltalk driver: MongoTalk. >> When I tried to install it on Pharo v1.3, however, every test failed >> because of a message sent to a nonexistent class. This class was GRCodec, >> if I recall well. >> >> Does anyone know something about this? >> >> Thanks in advance! >> >> Lucio >> > > > > -- > Milan Mimica > http://sparklet.sf.net > > >

Jan. 28, 2012

Re: [Pharo-project] FSFileSystem: modification date

by DÃ©bora Fortini

ok, Thank you! ;) On 27 January 2012 21:42, Camillo Bruni <camillo.bruni(a)inria.fr> wrote: > Hi Deby, > > Open a bug report :) I think the memory FS should handle the modification / creation dates exactly as it would on disk... > > > BTW: > > file := (fs/'hello') withExtension: 'txt' > > can be written as > > file := (fs/'hello') , 'txt' > > best > camillo > > On 2012-01-27, at 23:33, DÃ©bora Fortini wrote: >> While using FileSystem I found that the memory storage doesn't save >> the modification date. Here is an example of what is failing: >> >> Â Â Â fs := FSFilesystem memory. >> Â Â Â file := (fs/'hello') withExtension: 'txt'. >> Â Â Â file fileStreamDo: [ :stream | stream nextPutAll: 'foo' ]. >> Â Â Â file entry modification Â ---> 1901-01-01T00:00:00-03:00 >> >> This is a little awkward, because I want to use the memory storage to >> write tests... >> Should I open a bug entry? isn't it a bug? >> >> Thank you in advance! >> >> Deby >> > >

Jan. 28, 2012

Re: [Pharo-project] Artifacts Mirroring

by Camillo Bruni

On 2012-01-27, at 20:43, Sven Van Caekenberghe wrote: > On 27 Jan 2012, at 17:31, Camillo Bruni wrote: >> ./googlecode_upload.py --help >> Usage: googlecode-upload.py -s SUMMARY -p PROJECT [options] FILE >> >> Options: >> -h, --help show this help message and exit >> -s SUMMARY, --summary=SUMMARY >> Short description of the file >> -p PROJECT, --project=PROJECT >> Google Code project name >> -u USER, --user=USER Your Google Code username >> -w PASSWORD, --password=PASSWORD >> Your Google Code password >> -l LABELS, --labels=LABELS >> An optional lis > > I had a quick look at http://code.google.com/p/support/source/browse/trunk/scripts/googlecode_upl… > > should be doable with zn+zdc as wellâ¦ you know I still have this inner fear of creating scripts in pharo :D and I immediately back out and use ruby / bash :P I'll have a look, for now I'm fine with a minimal effort solution :). BTW: with the Monkey we already did a complete google API authentification which works perfectly :)

Jan. 28, 2012

Re: [Pharo-project] FSFileSystem: modification date

by Camillo Bruni

Hi Deby, Open a bug report :) I think the memory FS should handle the modification / creation dates exactly as it would on disk... BTW: file := (fs/'hello') withExtension: 'txt' can be written as file := (fs/'hello') , 'txt' best camillo On 2012-01-27, at 23:33, DÃ©bora Fortini wrote: > While using FileSystem I found that the memory storage doesn't save > the modification date. Here is an example of what is failing: > > fs := FSFilesystem memory. > file := (fs/'hello') withExtension: 'txt'. > file fileStreamDo: [ :stream | stream nextPutAll: 'foo' ]. > file entry modification ---> 1901-01-01T00:00:00-03:00 > > This is a little awkward, because I want to use the memory storage to > write tests... > Should I open a bug entry? isn't it a bug? > > Thank you in advance! > > Deby >

Jan. 28, 2012

Re: [Pharo-project] MongoTalk

by Esteban Lorenzano

Hi, I just uploaded a couple of fixes I made... with those fixes, I have an application in production, and working really well :) best, Esteban El 27/01/2012, a las 8:47p.m., Milan Mimica escribiÃ³: > AFAIR, the next think you'll have to fix is sending #first to aCharacter. Just remove the message send and off you go. > > > On 27 January 2012 23:28, Lucio Santi <lukius(a)gmail.com> wrote: > Hi! > > These days I've been playing around with MongoDB and it was a nice surprise to discover the availability of a Smalltalk driver: MongoTalk. When I tried to install it on Pharo v1.3, however, every test failed because of a message sent to a nonexistent class. This class was GRCodec, if I recall well. > > Does anyone know something about this? > > Thanks in advance! > > Lucio > > > > -- > Milan Mimica > http://sparklet.sf.net

Jan. 28, 2012

Re: [Pharo-project] MongoTalk

by Milan Mimica

AFAIR, the next think you'll have to fix is sending #first to aCharacter. Just remove the message send and off you go. On 27 January 2012 23:28, Lucio Santi <lukius(a)gmail.com> wrote: > Hi! > > These days I've been playing around with MongoDB and it was a nice > surprise to discover the availability of a Smalltalk driver: MongoTalk. > When I tried to install it on Pharo v1.3, however, every test failed > because of a message sent to a nonexistent class. This class was GRCodec, > if I recall well. > > Does anyone know something about this? > > Thanks in advance! > > Lucio > -- Milan Mimica http://sparklet.sf.net

Jan. 27, 2012

Re: [Pharo-project] MongoTalk

by Sven Van Caekenberghe

Lucio, On 27 Jan 2012, at 23:28, Lucio Santi wrote: > Hi! > > These days I've been playing around with MongoDB and it was a nice surprise to discover the availability of a Smalltalk driver: MongoTalk. When I tried to install it on Pharo v1.3, however, every test failed because of a message sent to a nonexistent class. This class was GRCodec, if I recall well. > > Does anyone know something about this? > > Thanks in advance! > > Lucio GRCodec is part of Grease, a Smalltalk compatibility layer, used by Seaside. Either install Seaside, or just the Grease part from Seaside. Sven

Jan. 27, 2012

[Pharo-project] FSFileSystem: modification date

by DÃ©bora Fortini

While using FileSystem I found that the memory storage doesn't save the modification date. Here is an example of what is failing: fs := FSFilesystem memory. file := (fs/'hello') withExtension: 'txt'. file fileStreamDo: [ :stream | stream nextPutAll: 'foo' ]. file entry modification ---> 1901-01-01T00:00:00-03:00 This is a little awkward, because I want to use the memory storage to write tests... Should I open a bug entry? isn't it a bug? Thank you in advance! Deby

Jan. 27, 2012

[Pharo-project] MongoTalk

by Lucio Santi

Hi! These days I've been playing around with MongoDB and it was a nice surprise to discover the availability of a Smalltalk driver: MongoTalk. When I tried to install it on Pharo v1.3, however, every test failed because of a message sent to a nonexistent class. This class was GRCodec, if I recall well. Does anyone know something about this? Thanks in advance! Lucio

Jan. 27, 2012

Re: [Pharo-project] Enthousiasm is the main currency among developers

by Eliot Miranda

On Fri, Jan 27, 2012 at 1:53 PM, Guido Stepken <gstepken(a)googlemail.com>wrote: > I see quite a difference between "doing things right" and "doing the right > things" ! :-) > Agreed. Don't let the perfect be the enemy of the good and all that. But we're talking at different levels here. I want to hear what Marcus thinks to my reply to his post. That's where this thread comes from. > Am 27.01.2012 22:50 schrieb "Eliot Miranda" <eliot.miranda(a)gmail.com>: > > >> >> On Fri, Jan 27, 2012 at 11:51 AM, Guido Stepken <gstepken(a)googlemail.com>wrote: >> >>> Hi Elliot! >>> >>> When I rethink, why new programming languages came up from zero to a >>> significant market share, like PERL, PHP, Python, Ruby, JAVA, C# (.net) >>> Visual Basic, Visual C++ and others died out, like Delphi, >>> TurboBasic/Pascal/C I could name different reasons: >>> >>> - Free license vs. expensive >>> - Wrong payment model (per developer, per runtime, both) >>> - Good, free support on websites vs. "Bronze/silver/gold" >>> paystupid-support >>> - Attractiveness of one "killer app" that made programmers change to >>> another language >>> - Portability of code onto other platforms >>> - Mightyness of libraries >>> - Missing standards, protocols, support of hardware >>> - Good vs. bad marketing, deciders not convinced that product will >>> survive/missing timeline, visions, lack of money in background >>> - Subcritical mass of programmers using product, lack of professionals >>> >>> That was in former times. >>> >>> Today, new criterias play a far more relevant role, hat haven't really >>> existed just 3 years ago: >>> >>> - Has it (the OS,the programming language and GUI framework) an >>> appstore/plugin concept to let free, creative brains being able to >>> participate, earn money with? >>> - Barrier - free payment model included (mobile payment, card, bank >>> account)? >>> - Free use with sponsoring by ads possible (programmers payed from >>> multiple resources, not user alone) >>> - Cryptographic prevention of missuse included? >>> - Free and matured SDK available? >>> - Connections to social software like facebook/twitter/Google+/Groupon >>> included (API access, programming language and all protocols supported) >>> - GUI designed for desktop as well usable for touch and self adapting to >>> different screen/touch sizes? >>> - Touch gestures possible and lib avail? >>> - Microsofts Kinect hardware/video recognition of faces, hand/face mimic >>> gestures possible and supported in libs? >>> - Voice recognition supported? >>> - Mobile ready? (touch, GPS, compass, barometer, gyro, hardware OpenGL) >>> - Rockstable? >>> - Fast, running in low power devices? Joule per clock cycle ratio??? >>> - Critical mass of users already reached, increasing? >>> - Critical number of apps there to raise interest? >>> ... >>> >>> So, the Pharo developers might now decide, what to invest their >>> brainpower into! :-) >>> >>> Just my 2ct. >>> >> >> OK, that looks like a great list. But don't you agree that criticism (in >> the sense of something that leads to quality software engineering) >> underlies several of these, such as Rockstable, Fast, running on low-power >> devices, etc? To me, being critical doesn't mean being uncreative or >> conservative; it means thinking about what you're doing, and doing a good >> job. >> >>> Guido Stepken >>> Am 27.01.2012 19:46 schrieb "Eliot Miranda" <eliot.miranda(a)gmail.com>: >>> >>> >>>> >>>> On Fri, Jan 27, 2012 at 5:33 AM, Marcus Denker <marcus.denker(a)inria.fr>wrote: >>>> >>>>> >>>>> On Jan 27, 2012, at 6:13 AM, dimitris chloupis wrote: >>>>> >>>>> > This article is really encapsulates the attitude and what is wrong >>>>> with programming in general. The attitude of superiority and intelligence >>>>> that seems to plague coders and being the biggest obstacle to progress. >>>>> >>>>> Yes! The "Everyone is dumb but me" phenomenon... >>>>> >>>>> What those "intelligent" people don't get is that complexity is >>>>> inherently exponential. So even if you are >>>>> 10 times more intelligent than me (very well possible), it is >>>>> *completely* irrelevant considering that complexity >>>>> grows non-linearly. >>>>> >>>>> If you combine this with the notion of Evolution: that it is >>>>> impossible to creat "the perfect" out of nothing, yet >>>>> entropy grows when you incrementally improve things... than this has >>>>> some very serious consequences. >>>>> >>>>> > For me the main problem with is the whole aura of "elitism" , what >>>>> better example than Lisp, where beginners are attacked and be excluded. >>>>> >>>>> We had the same effect in Squeak at the end. No progress, every >>>>> improvement was actively fighted against, if needed with the nice argument >>>>> that >>>>> one can do it even better, and only "the best" is worth for Squeak. >>>>> >>>>> Another thing that "intelligent" people don't get is that critizising >>>>> is trivial: You can *always* do better, there is no perfection. It's an >>>>> endless process. >>>>> This implies that one has to accept and embrace imperfection if one >>>>> wants to have a future. Else you end up never finishing anything, the death >>>>> of any >>>>> incremental progress. >>>>> >>>> >>>> But criticism is essential. How does one identify a mistake if not by >>>> criticising? There's a huge difference between constructive criticism >>>> (analysis, testing, comparison, evaluation, measurement) and negativity >>>> (denial, fear, slander). How can one engineer without measurement, without >>>> thought? Being agile doesn't imply being random. Evolution measures, and >>>> most harshly; the weaker don't survive. >>>> >>>> >>>>> Pharo was started with the explicit goal to do as many mistakes as >>>>> possible, as fast as possible. >>>>> >>>>> Marcus >>>>> >>>>> -- >>>>> Marcus Denker -- http://marcusdenker.de >>>>> >>>>> >>>>> >>>> >>>> >>>> -- >>>> best, >>>> Eliot >>>> >>>> >> >> >> -- >> best, >> Eliot >> >> -- best, Eliot

Jan. 27, 2012