It seems like everyone now has a RoR type framework: Perl: http://dev.catalyst.perl.org/ PHP: http://cakephp.org/ Python: http://www.django.com ASP.NET: http://www.castleproject.org/index.php/Main_Page etc., etc. Has a revolution begun with the birth of these "Web 2.0" frameworks? Its going to be interesting when the tech media begins its mass hype like it did SOAP, XML, etc. a few years back - setting expectations above and beyond. Hopefully by then all the frameworks will be mature enough to live up to any of the CTO''s wild dreams :-) I''ve seen hints in the press of the hype starting... Its a great time to be a developer.
> Python: > http://www.django.comhttp://www.djangoproject.com/ you mean...
Chad Lung wrote:> It seems like everyone now has a RoR type framework: > > Perl: > http://dev.catalyst.perl.org/ >Speaking of which, what do you think of the comparison table here: http://dev.catalyst.perl.org/wiki/VersusRails I''ve been doing comparison research for my company and helped start that page. The goal was to be as fair as possible, but it''s obviously had more attention by Catalyst folks than you guys. Phil
On Aug 17, 2005, at 5:07 PM, Fabien Penso wrote:>> Python: >> http://www.django.com > > http://www.djangoproject.com/ you mean...bandwagon.jump_on("http://www.javaonsails.org")
anyone''s arm getting tired yet? -- ________________________________ toddgrimason*todd[ at ]slack.net
Philip Edelbrock Wrote:>Speaking of which, what do you think of the comparison table here:>http://dev.catalyst.perl.org/wiki/VersusRailsIts a good comparison I think. To be honest, other than RoR, Django, and Catalyst, I don''t see the others keeping up. I find the Perl syntax a bit obtuse really, but there is no denying that it is an industry powerhorse and can scale very, very well. I imagine as usual it will come down to what a person feels comfortable working in.
One of the benefits of creating a new idea from scratch is that there is less tendency for work-arounds in the code. This is a key benefit that I find in RoR - the framework was designed from the ground up with Web 2 notions and ideas. Other frameworks will no doubt atempt the same ease in development, however it become very complicated as libraries and dependencies begin to mount up in them. On 8/17/05, Chad Lung <chad.lung-Re5JQEeQqe8AvxtiuMwx3w@public.gmane.org> wrote:> Philip Edelbrock Wrote: > > >Speaking of which, what do you think of the comparison table here: > > >http://dev.catalyst.perl.org/wiki/VersusRails > > Its a good comparison I think. > > To be honest, other than RoR, Django, and Catalyst, I don''t see the > others keeping up. I find the Perl syntax a bit obtuse really, but > there is no denying that it is an industry powerhorse and can scale > very, very well. I imagine as usual it will come down to what a > person feels comfortable working in. > _______________________________________________ > Rails mailing list > Rails-1W37MKcQCpIf0INCOvqR/iCwEArCW2h5@public.gmane.org > http://lists.rubyonrails.org/mailman/listinfo/rails >-- ~~~~~~~~~~~~~~~~~~~ D''Andrew Thompson http://dathompson.blogspot.com
Philip Edelbrock wrote:> I''ve been doing comparison research for my company and helped start that > page. The goal was to be as fair as possible, but it''s obviously had > more attention by Catalyst folks than you guys.I gave it a brief treatment. We''ll see how long that lasts though. It is on their wiki after all...
Philip Edelbrock wrote:> > > Chad Lung wrote: > >> It seems like everyone now has a RoR type framework: >> >> Perl: >> http://dev.catalyst.perl.org/ >> > > Speaking of which, what do you think of the comparison table here: > > http://dev.catalyst.perl.org/wiki/VersusRails >Um, is Perl 6 and Parrot out yet... if not... then when did... "Catalyst leverages Perl''s OO architecture with multiple-inheritance as well as mix-ins." perl get a worthwhile OO architecture? The rest of the comparison seems fairly objective (except the younger/older part which seems like pointless adjective use) Zach
Zach Dennis wrote:> "Catalyst leverages Perl''s OO architecture with multiple-inheritance as > well as mix-ins."Yea, and when was MI considered an advantage? Especially given all the cool stuff we can do with Module#included et al? Try reading the versions before my modifications for some decent humor...
Zach Dennis wrote:> Um, is Perl 6 and Parrot out yet... if not... then when did...A better question for me is, "Will Perl 6 run my old Perl 4 code without me having to invest any time in reviewing it? :)
Keep an eye out for "lisp on lines" and "scheme on skis" _______________________________________________ Rails mailing list Rails-1W37MKcQCpIf0INCOvqR/iCwEArCW2h5@public.gmane.org http://lists.rubyonrails.org/mailman/listinfo/rails
I''ve seen Lisp on Lines ... how about Guile on Guides? John Knight wrote:> Keep an eye out for "lisp on lines" and "scheme on skis" > > > >------------------------------------------------------------------------ > >_______________________________________________ >Rails mailing list >Rails-1W37MKcQCpIf0INCOvqR/iCwEArCW2h5@public.gmane.org >http://lists.rubyonrails.org/mailman/listinfo/rails > >
On 8/17/05, M. Edward (Ed) Borasky <znmeb-2WxwdZd67h7R7s880joybQ@public.gmane.org> wrote:> I''ve seen Lisp on Lines ... how about Guile on Guides?I''m waiting for Forth on Frames. -- Urban Artography http://artography.ath.cx
On 8/17/05, D''Andrew Thompson <dandrew.thompson-Re5JQEeQqe8AvxtiuMwx3w@public.gmane.org> wrote:> One of the benefits of creating a new idea from scratch is that there > is less tendency for work-arounds in the code. This is a key benefit > that I find in RoR - the framework was designed from the ground up > with Web 2 notions and ideas.Yeah, just give it a couple of years and it''ll be old a crusty just like everything else ;) -- Urban Artography http://artography.ath.cx
for java its "trails" :) https://trails.dev.java.net/
Just wait for Cobol on Crates On 8/18/05, Rob Park <rbpark-Re5JQEeQqe8AvxtiuMwx3w@public.gmane.org> wrote:> On 8/17/05, M. Edward (Ed) Borasky <znmeb-2WxwdZd67h7R7s880joybQ@public.gmane.org> wrote: > > I''ve seen Lisp on Lines ... how about Guile on Guides? > > I''m waiting for Forth on Frames.-- ---------------------------------------------------- http://hronia.blogalia.com
Zach Dennis <zdennis-aRAREQmnvsAAvxtiuMwx3w@public.gmane.org> writes:> Philip Edelbrock wrote:> Um, is Perl 6 and Parrot out yet... if not... then when did... > > "Catalyst leverages Perl''s OO architecture with multiple-inheritance > as well as mix-ins." > > perl get a worthwhile OO architecture? The rest of the comparison > seems fairly objective (except the younger/older part which seems > like pointless adjective use)To be fair, his writeup make no assertions as to perl''s OO''s worthiness, only that it''s there. -- I tend to view "truly flexible" by another term: "Make everything equally hard". -- DHH
Or Ada on Amtrack ;-) On Aug 17, 2005, at 11:54 PM, Rob Park wrote:> On 8/17/05, M. Edward (Ed) Borasky <znmeb-2WxwdZd67h7R7s880joybQ@public.gmane.org> wrote: > >> I''ve seen Lisp on Lines ... how about Guile on Guides? >> > > I''m waiting for Forth on Frames. > > -- > Urban Artography > http://artography.ath.cx > _______________________________________________ > Rails mailing list > Rails-1W37MKcQCpIf0INCOvqR/iCwEArCW2h5@public.gmane.org > http://lists.rubyonrails.org/mailman/listinfo/rails >-Ezra Zygmuntowicz Yakima Herald-Republic WebMaster 509-577-7732 ezra-gdxLOakOTQ9oetBuM9ipNAC/G2K4zDHf@public.gmane.org
PLEASE let this be the last response to this message... The first one or two silly replies may have been funny, but now its starting to look like spam! Ezra Zygmuntowicz <ezra-gdxLOakOTQ9oetBuM9ipNAC/G2K4zDHf@public.gmane.org> wrote: Or Ada on Amtrack ;-) On Aug 17, 2005, at 11:54 PM, Rob Park wrote:> On 8/17/05, M. Edward (Ed) Borasky wrote: > >> I''ve seen Lisp on Lines ... how about Guile on Guides? >> > > I''m waiting for Forth on Frames. > > -- > Urban Artography > http://artography.ath.cx > _______________________________________________ > Rails mailing list > Rails-1W37MKcQCpIf0INCOvqR/iCwEArCW2h5@public.gmane.org > http://lists.rubyonrails.org/mailman/listinfo/rails >-Ezra Zygmuntowicz Yakima Herald-Republic WebMaster 509-577-7732 ezra-gdxLOakOTQ9oetBuM9ipNAC/G2K4zDHf@public.gmane.org _______________________________________________ Rails mailing list Rails-1W37MKcQCpIf0INCOvqR/iCwEArCW2h5@public.gmane.org http://lists.rubyonrails.org/mailman/listinfo/rails --------------------------------- Do you Yahoo!? Yahoo! Mail - Helps protect you from nasty viruses. _______________________________________________ Rails mailing list Rails-1W37MKcQCpIf0INCOvqR/iCwEArCW2h5@public.gmane.org http://lists.rubyonrails.org/mailman/listinfo/rails
To be fair to Django, it was developed around the same time as Rails, it''s not a knockoff. Putting it in the same category as the "me-too''s" isn''t really appropriate. Larry http://www.approachingnormal.com On 8/17/05, Chad Lung <chad.lung-Re5JQEeQqe8AvxtiuMwx3w@public.gmane.org> wrote:> > It seems like everyone now has a RoR type framework: > > Perl: > http://dev.catalyst.perl.org/ > > PHP: > http://cakephp.org/ > > Python: > http://www.django.com > > ASP.NET <http://ASP.NET>: > http://www.castleproject.org/index.php/Main_Page > > etc., etc. > > Has a revolution begun with the birth of these "Web 2.0" frameworks? > Its going to be interesting when the tech media begins its mass hype > like it did SOAP, XML, etc. a few years back - setting expectations > above and beyond. Hopefully by then all the frameworks will be mature > enough to live up to any of the CTO''s wild dreams :-) I''ve seen hints > in the press of the hype starting... > > Its a great time to be a developer. > _______________________________________________ > Rails mailing list > Rails-1W37MKcQCpIf0INCOvqR/iCwEArCW2h5@public.gmane.org > http://lists.rubyonrails.org/mailman/listinfo/rails >_______________________________________________ Rails mailing list Rails-1W37MKcQCpIf0INCOvqR/iCwEArCW2h5@public.gmane.org http://lists.rubyonrails.org/mailman/listinfo/rails
On 8/18/05, Timothy Fisher <trfishermi-/E1597aS9LQAvxtiuMwx3w@public.gmane.org> wrote:> PLEASE let this be the last response to this message... > The first one or two silly replies may have been funny, but now its starting > to look like spam!Actually I enjoy puns & wordplay and I''d like to see more examples of "[Insert Programming Language] on [Insert word that starts with the same first letter as the programming language and refers to some kind of support structure or similar]". Perhaps Algol on Avenues, Basic on Braces, Matlab on Mountings, Pascal on Pegs, Perl on Posts, Python on Props, Scheme on Splints, Tcl on Trusses, VB on Vices... Ok, I''m done... for now ;) -- Urban Artography http://artography.ath.cx
R on Rollerskates? Rob Park wrote:>On 8/18/05, Timothy Fisher <trfishermi-/E1597aS9LQAvxtiuMwx3w@public.gmane.org> wrote: > > >>PLEASE let this be the last response to this message... >>The first one or two silly replies may have been funny, but now its starting >>to look like spam! >> >> > >Actually I enjoy puns & wordplay and I''d like to see more examples of >"[Insert Programming Language] on [Insert word that starts with the >same first letter as the programming language and refers to some kind >of support structure or similar]". > >Perhaps Algol on Avenues, Basic on Braces, Matlab on Mountings, Pascal >on Pegs, Perl on Posts, Python on Props, Scheme on Splints, Tcl on >Trusses, VB on Vices... > >Ok, I''m done... for now ;) > > >
I suppose I should have known better than to suggest we stop.... :) "M. Edward (Ed) Borasky" <znmeb-2WxwdZd67h7R7s880joybQ@public.gmane.org> wrote:R on Rollerskates? Rob Park wrote:>On 8/18/05, Timothy Fisher wrote: > > >>PLEASE let this be the last response to this message... >>The first one or two silly replies may have been funny, but now its starting >>to look like spam! >> >> > >Actually I enjoy puns & wordplay and I''d like to see more examples of >"[Insert Programming Language] on [Insert word that starts with the >same first letter as the programming language and refers to some kind >of support structure or similar]". > >Perhaps Algol on Avenues, Basic on Braces, Matlab on Mountings, Pascal >on Pegs, Perl on Posts, Python on Props, Scheme on Splints, Tcl on >Trusses, VB on Vices... > >Ok, I''m done... for now ;) > > >_______________________________________________ Rails mailing list Rails-1W37MKcQCpIf0INCOvqR/iCwEArCW2h5@public.gmane.org http://lists.rubyonrails.org/mailman/listinfo/rails --------------------------------- Start your day with Yahoo! - make it your home page _______________________________________________ Rails mailing list Rails-1W37MKcQCpIf0INCOvqR/iCwEArCW2h5@public.gmane.org http://lists.rubyonrails.org/mailman/listinfo/rails
How about Visual Basic in Vomit Basin? Oh... we already have that it''s called ASP... Maybe it should have been Visual Basic on Venomous Boa? ;-) That''s not really an Asp, though, is it? :) On 18/08/2005, at 4:54 PM, Rob Park wrote:> On 8/17/05, M. Edward (Ed) Borasky <znmeb-2WxwdZd67h7R7s880joybQ@public.gmane.org> wrote: > >> I''ve seen Lisp on Lines ... how about Guile on Guides? >> > > I''m waiting for Forth on Frames. > > -- > Urban Artography > http://artography.ath.cx > _______________________________________________ > Rails mailing list > Rails-1W37MKcQCpIf0INCOvqR/iCwEArCW2h5@public.gmane.org > http://lists.rubyonrails.org/mailman/listinfo/rails >
Actually, I have needs for something resembling "R on Rollerskates". That is, either a full-scale web application framework like Ruby on Rails implemented in the R language rather than in Ruby, or some way of running R subservient to Ruby on Rails. Taking the first approach, is there anything special in the *semantics* of Ruby that made/makes Rails easier to do in Ruby than it would in Perl, Python, PHP, Tcl/Tk or any other "modern scripting language"? R has everything a "modern scripting language" has. R does have an object system, although it''s quite different from most others. R can be extended with C, C++, and, given its scientific nature, also with Fortran! R has its own package management system, R has interface code for most common databases, R has libraries that can handle Perl-compatible regular expressions, CGI, HTML and XML, etc. R is open source. The one hitch in this is that R for Linux/UNIX/BSD/XWindows and R for Microsoft Windows are two quite different implementations, constructed and maintained using completely different tool sets, having completely different user interfaces, etc. There is a common semantic core between the two, of course, and one can build R as a DLL on Windows and a shared library on Linux. So making a Windows version of R on Rollerskates can probably be done, though it isn''t something I''d want to spend *any* time on. The second approach looks far easier. In any event, where I''m going with all of this is that R contains intensely advanced statistical capability (can you say "data mining on steroids"?) that I have no desire or intention to re-implement in Ruby. M. Edward (Ed) Borasky wrote:> R on Rollerskates? > >-- M. Edward (Ed) Borasky http://www.borasky-research.net/ http://borasky-research.blogspot.com/ http://pdxneurosemantics.com http://pdx-sales-coach.com http://algocompsynth.com
M. Edward (Ed) Borasky wrote:> [...] is there anything special in the *semantics* > of Ruby that made/makes Rails easier to do in Ruby than it would in > Perl, Python, PHP, Tcl/Tk or any other "modern scripting language"?Ruby is strong for metaprogramming, allowing Rails to cleanly embed domain-specific languages for validation, O-R mapping, routing etc. I just found these slides from Glenn Vanderburg''s OSCON presentation, "Metaprogramming Ruby: Domain-Specific Languages for Programmers", which look like a good introduction: http://www.vanderburg.org/Speaking/Stuff/oscon05.pdf regards Justin
On 8/25/05, Justin Forder <justin-zSfPWr5aQuznITO/+xaoB7VCufUGDwFn@public.gmane.org> wrote:> M. Edward (Ed) Borasky wrote: > > > [...] is there anything special in the *semantics* > > of Ruby that made/makes Rails easier to do in Ruby than it would in > > Perl, Python, PHP, Tcl/Tk or any other "modern scripting language"? > > Ruby is strong for metaprogramming, allowing Rails to cleanly embed > domain-specific languages for validation, O-R mapping, routing etc. > > I just found these slides from Glenn Vanderburg''s OSCON presentation, > "Metaprogramming Ruby: Domain-Specific Languages for Programmers", > which look like a good introduction: > > http://www.vanderburg.org/Speaking/Stuff/oscon05.pdfBill Katz had a weblog entry on this same topic: "Could Rails have been built without Ruby?" http://www.billkatz.com/node/42 -- rick http://techno-weenie.net
Rick Olson said:> Bill Katz had a weblog entry on this same topic: "Could Rails have > been built without Ruby?" > http://www.billkatz.com/node/42Mr. Katz doesn''t seem to really answer the question. For those who are still wondering, the answer is "absolutely!" Ruby is not uniquely suited to building a web application development framework with the essential characteristics of Rails. Would another implementation in another language share all of the accidental traits of Rails? Of course not. The future of Rails is not assured: only its place in history is. Frameworks in every conceivable language will emerge from people who have paid careful attention to the strengths and weaknesses of Rails. DHH and the rest of the Rails developers don''t see themselves as building a proof-of-concept prototype, but that''s what Rails may amount to. And if that''s all Rails ever becomes, it will still have been a fantastic success. Regards, Ed -- Transmogrify, LLC * <http://xmog.com>
> For those who are still wondering, the answer is "absolutely!"I agree. In the same ways as trying to build a Ferrari from the spare parts of a Toyota can be called a success when the wheels start turning and you have movement. In other words, if recreating Rails in another language means building a web-application framework that shares some of the same attributes (a steering wheel, engine, etc), then that has already happened. Equality defined through that point of view is a sliding scale. A clone can be more or less like Rails than another clone. So it''s a relative thing. Ed''s scale for when something is close enough to Rails to call it "done in another language" is probably not the same as mine, though. -- David Heinemeier Hansson http://www.loudthinking.com -- Broadcasting Brain http://www.basecamphq.com -- Online project management http://www.backpackit.com -- Personal information manager http://www.rubyonrails.com -- Web-application framework
David Heinemeier Hansson said:>> For those who are still wondering, the answer is "absolutely!" > > [...] > > Ed''s scale for when something is close enough to Rails to call it > "done in another language" is probably not the same as mine, though.David, So what are the essential attributes of Rails? What virtues must a web development framework possess in order to be your metaphorical Ferrari? I ask so that we can have a set of criteria to judge each new wanna-be Rails usurper, a set of criteria that doesn''t shift like the sands as incidental attributes of Rails are trotted out to demonstrate Rails''s superior Rails-ness. Here''s a start: * Doesn''t require configuration when convention will do. * Uses same language for control flow, business logic, and output templates. * Encourages incremental development through fast turnaround of modifications . * Embraces the Web, version 2.0. * Supports model-view-controller approach to development. * Provides tools to capture and re-use abstractions, such as CRUD scaffolding. I''ve probably missed a few things. I''d be willing to bet you 100 USD that in five years, Rails will not be the most popular framework that supports whatever list of criteria we come up with, and whatever framework is most popular will not be written in Ruby. Regards, Ed -- Transmogrify, LLC * <http://xmog.com>
Thanks! Regarding "metaprogramming", the book "S Programming", by Venables and Ripley, has a whole section called "Computing on the Language", which I interpret as "metaprogramming". R is an open-source dialect of S, and is sometimes called "GNU S". I think I saw those slides; del.icio.us put them in my Inbox when someone linked to them. BTW, I have a lot of links to RoR there; try http://del.icio.us/znmeb In any event, R semantics is quite "functional" and owes a lot to Lisp and Scheme, though probably not as much to Smalltalk as Ruby does. Its object philosophy doesn''t look at all like what people normally think of, although it probably resembles the Common Lisp Object System (CLOS). Rick Olson wrote:>On 8/25/05, Justin Forder <justin-zSfPWr5aQuznITO/+xaoB7VCufUGDwFn@public.gmane.org> wrote: > > >>M. Edward (Ed) Borasky wrote: >> >> >> >>>[...] is there anything special in the *semantics* >>>of Ruby that made/makes Rails easier to do in Ruby than it would in >>>Perl, Python, PHP, Tcl/Tk or any other "modern scripting language"? >>> >>> >>Ruby is strong for metaprogramming, allowing Rails to cleanly embed >>domain-specific languages for validation, O-R mapping, routing etc. >> >>I just found these slides from Glenn Vanderburg''s OSCON presentation, >>"Metaprogramming Ruby: Domain-Specific Languages for Programmers", >>which look like a good introduction: >> >> http://www.vanderburg.org/Speaking/Stuff/oscon05.pdf >> >> > >Bill Katz had a weblog entry on this same topic: "Could Rails have >been built without Ruby?" >http://www.billkatz.com/node/42 > > > >-- M. Edward (Ed) Borasky http://www.borasky-research.net/ http://borasky-research.blogspot.com/ http://pdxneurosemantics.com http://pdx-sales-coach.com http://algocompsynth.com
David Heinemeier Hansson wrote:>>For those who are still wondering, the answer is "absolutely!" >> >> > >I agree. In the same ways as trying to build a Ferrari from the spare >parts of a Toyota can be called a success when the wheels start >turning and you have movement. > >In other words, if recreating Rails in another language means building >a web-application framework that shares some of the same attributes (a >steering wheel, engine, etc), then that has already happened. > >What would that be, and what language is it written in?>Equality defined through that point of view is a sliding scale. A >clone can be more or less like Rails than another clone. So it''s a >relative thing. > >Ed''s scale for when something is close enough to Rails to call it >"done in another language" is probably not the same as mine, though. > >Which Ed, me or Watkeys? Or both? :) I suspect the path of least work for what I want to accomplish will be to build the framework using RoR, even though I have zero Ruby knowledge and will have to transfer Perl skills and develop a reading knowledge of Ruby very quickly to pull *anything* off. Then the heavy statistical lifting can be delegated to R, where I do have a few years'' experience and lots of existing software to do the database access and "unsupervised learning". I suspect in the end, the core data moving stuff, which is what most applications spend most of their time doing, is going to end up much more efficient in Ruby than in R anyhow. For this application, the framework doesn''t have to be very complicated anyhow ... it''s essentially ASCII (or maybe Unicode) text processing. -- M. Edward (Ed) Borasky http://www.borasky-research.net/ http://borasky-research.blogspot.com/ http://pdxneurosemantics.com http://pdx-sales-coach.com http://algocompsynth.com
Ed Watkeys wrote:>David Heinemeier Hansson said: > > >>>For those who are still wondering, the answer is "absolutely!" >>> >>> >>[...] >> >>Ed''s scale for when something is close enough to Rails to call it >>"done in another language" is probably not the same as mine, though. >> >> > >David, > >So what are the essential attributes of Rails? What virtues must a web >development framework possess in order to be your metaphorical Ferrari? I >ask so that we can have a set of criteria to judge each new wanna-be Rails >usurper, a set of criteria that doesn''t shift like the sands as incidental >attributes of Rails are trotted out to demonstrate Rails''s superior >Rails-ness. > >Here''s a start: > >* Doesn''t require configuration when convention will do. > >Can be done in R.>* Uses same language for control flow, business logic, and output templates. > >Can be done in R. Of course, the developer needs to be capable/proficient in R, just as a Rails developer needs to be capable/proficient in Ruby.>* Encourages incremental development through fast turnaround of >modifications . > >R is interpretive. R has one basic data type, a list. I don''t remember hashes, but the lists have named and numbered elements. I guess that implies fast turnaround.>* Embraces the Web, version 2.0. > >That would need to be built into libraries for R, I think. Right now, I don''t know how one does "embedded R" in a web page like one does "embedded Ruby" in an rhtml page. And I don''t think R has anything like Ajax yet, so more coding ... :(>* Supports model-view-controller approach to development. > >I don''t see why this can''t be done in R, although R is probably better at modeling -- really advanced modeling -- and views -- really advanced scientific graphics -- than it is at dispatching emails, etc. I think what I''m looking for is the controller in Ruby and the model and view in R.>* Provides tools to capture and re-use abstractions, such as CRUD >scaffolding. > >Creates aren''t necessarily easy in R, but you can do any SQL queries the underlying RDMS supports in one line of code (albeit a call to a DBI or DBD routine.) I''m pretty sure you can build something like the Rails ORM quite easily in R.>I''ve probably missed a few things. I''d be willing to bet you 100 USD that >in five years, Rails will not be the most popular framework that supports >whatever list of criteria we come up with, and whatever framework is most >popular will not be written in Ruby. > >Well ... given what I call "the vast army of inexpensive C/C++ programmers", you''re probably right. However, I think Rails/Ruby has as good a shot at it as any *currently-existing* framework/language, and if the RoR community comes up with "killer features" and "killer apps", who knows? The Perl of today isn''t the same Perl as five years ago, and the same is likely for Ruby. Of course, if the Perl community keeps breaking my old code ... :) -- M. Edward (Ed) Borasky http://www.borasky-research.net/ http://borasky-research.blogspot.com/ http://pdxneurosemantics.com http://pdx-sales-coach.com http://algocompsynth.com
On 8/25/05, Ed Watkeys <edw-tIV1OJqwIcc@public.gmane.org> wrote:> So what are the essential attributes of Rails? What virtues must a web > development framework possess in order to be your metaphorical Ferrari? I > ask so that we can have a set of criteria to judge each new wanna-be Rails > usurper, a set of criteria that doesn''t shift like the sands as incidental > attributes of Rails are trotted out to demonstrate Rails''s superior > Rails-ness.See the docs for ActiveRecord::Associations::ClassMethods, ActiveRecord::Validations::ClassMethods and ActiveRecord::Aggregations::ClassMethods (among others) for a good idea. Yes, you can achieve the end result of these things in any halfway capable language, but it would be both harder and uglier. There _are_ other languages besides Ruby that could achieve something quite similar to Rails, but they are probably _less_ popular than Ruby is these days. (PHP, Perl, Python, etc. are _not_ one of those languages.) If you look at a website created with Ruby on Rails, it is absolutely possible to create an identical (to the end user) web site with PHP, Perl or Python. That''s not what the argument is, though. It''s about how much time it takes to create that website and (more importantly) how easy it is to maintain the code in the long-term. -- Regards, John Wilger ----------- Alice came to a fork in the road. "Which road do I take?" she asked. "Where do you want to go?" responded the Cheshire cat. "I don''t know," Alice answered. "Then," said the cat, "it doesn''t matter." - Lewis Carrol, Alice in Wonderland
On 26.8.2005, at 6.17, John Wilger wrote:> > If you look at a website created with Ruby on Rails, it is absolutely > possible to create an identical (to the end user) web site with PHP, > Perl or Python. That''s not what the argument is, though. It''s about > how much time it takes to create that website and (more importantly) > how easy it is to maintain the code in the long-term....and, most important to me, how much fun you have along the way. //jarkko -- Jarkko Laine http://jlaine.net http://odesign.fi _______________________________________________ Rails mailing list Rails-1W37MKcQCpIf0INCOvqR/iCwEArCW2h5@public.gmane.org http://lists.rubyonrails.org/mailman/listinfo/rails
On 8/25/05, Ed Watkeys <edw-tIV1OJqwIcc@public.gmane.org> wrote:> I''ve probably missed a few things. I''d be willing to bet you 100 USD that > in five years, Rails will not be the most popular framework that supports > whatever list of criteria we come up with, and whatever framework is most > popular will not be written in Ruby.Maybe I''ve missed something, but I thought the "popularity" arguement was bunk, otherwise we''d all be using J2EE or .Net -- Chris Griego
The "popularity" argument is most certainly not "bunk". Availability of inexpensive programmers is a legitimate concern to a project manager in a software development project. And I don''t buy the "Rails is more productive than X" argument myself. I''ve heard it before -- dozens of times. What is more productive is skilled, trained and motivated programmers who are intimately familiar with both the development environment and the application domain. Chris Griego wrote:>On 8/25/05, Ed Watkeys <edw-tIV1OJqwIcc@public.gmane.org> wrote: > > >>I''ve probably missed a few things. I''d be willing to bet you 100 USD that >>in five years, Rails will not be the most popular framework that supports >>whatever list of criteria we come up with, and whatever framework is most >>popular will not be written in Ruby. >> >> > >Maybe I''ve missed something, but I thought the "popularity" arguement >was bunk, otherwise we''d all be using J2EE or .Net > > >-- M. Edward (Ed) Borasky http://www.borasky-research.net/ http://borasky-research.blogspot.com/ http://pdxneurosemantics.com http://pdx-sales-coach.com http://algocompsynth.com
On Aug 27, 2005, at 4:15 AM, Chris Griego wrote:> On 8/25/05, Ed Watkeys <edw-tIV1OJqwIcc@public.gmane.org> wrote: > >> I''ve probably missed a few things. I''d be willing to bet you 100 >> USD that >> in five years, Rails will not be the most popular framework that >> supports >> whatever list of criteria we come up with, and whatever framework >> is most >> popular will not be written in Ruby. >> > > Maybe I''ve missed something, but I thought the "popularity" arguement > was bunk, otherwise we''d all be using J2EE or .NetSo would you like to propose an empirical, objective method of determining when a framework has out-Rails''d Rails? Does this religion have a falsifiability test? Please note that I don''t have a problem with Rails, simply the irrational cheerleading surrounding it. Regards, Ed -- Transmogrify, LLC * <http://xmog.com/>
On 8/27/05, Ed Watkeys <edw-tIV1OJqwIcc@public.gmane.org> wrote:> Please note that I don''t have a problem with Rails, simply the > irrational cheerleading surrounding it.Then again, without the irrational cheerleading, I''d bet that a lot fewer people would even know about Rails and thus would not be able to benefit from it. Certainly a company needs to make a rational, well-informed decision about the platform they use; but _every_ successful technology/framework/whatever has had its share of "irrational cheerleaders" which helped it get the attention it needed to survive among the plethora of options available. So the irrational cheerleading is actually a _good_ thing, as long as you''re smart enough not to pick your platform based solely on the _quantity_ of the cheerleaders. -- Regards, John Wilger ----------- Alice came to a fork in the road. "Which road do I take?" she asked. "Where do you want to go?" responded the Cheshire cat. "I don''t know," Alice answered. "Then," said the cat, "it doesn''t matter." - Lewis Carrol, Alice in Wonderland
M. Edward (Ed) Borasky wrote:> The "popularity" argument is most certainly not "bunk". Availability of > inexpensive programmers is a legitimate concern to a project manager in > a software development project. And I don''t buy the "Rails is more > productive than X" argument myself. I''ve heard it before -- dozens of > times. What is more productive is skilled, trained and motivated > programmers who are intimately familiar with both the development > environment and the application domain.Given two groups of equally skilled, trained and motivated programmers (Rails for one group and pick your poison for the other) who are intimately familiar with their respective development environments and _the same_ application domain, the Rails framework is more productive. I have tested it myself against another framework I hardly knew so that familiarity was not a determining factor, and Rails was much more productive. Other people I respect and believe have put it to much better tests than I, and they came to the same conclusion. You don''t have to buy it. You don''t have to believe that astronauts have been to the moon, either. This one is much easier to prove for yourself though.
On Sun, 2005-08-28 at 10:48 -0400, John Wilger wrote:> On 8/27/05, Ed Watkeys <edw-tIV1OJqwIcc@public.gmane.org> wrote: > > Please note that I don''t have a problem with Rails, simply the > > irrational cheerleading surrounding it. > > Then again, without the irrational cheerleading, I''d bet that a lot > fewer people would even know about Rails and thus would not be able to > benefit from it. Certainly a company needs to make a rational, > well-informed decision about the platform they use; but _every_ > successful technology/framework/whatever has had its share of > "irrational cheerleaders" which helped it get the attention it needed > to survive among the plethora of options available. > > So the irrational cheerleading is actually a _good_ thing, as long as > you''re smart enough not to pick your platform based solely on the > _quantity_ of the cheerleaders.Is it irrational cheer leading because no one is paid to do the cheer leading or by another name "Marketing"? -- Steven Critchfield critch-wQLwMjUOumVBDgjK7y7TUQ@public.gmane.org KI4KTY
Kevin Williams wrote:> Given two groups of equally skilled, trained and motivated programmers > (Rails for one group and pick your poison for the other) who are > intimately familiar with their respective development environments and > _the same_ application domain, the Rails framework is more productive. > I have tested it myself against another framework I hardly knew so > that familiarity was not a determining factor, and Rails was much more > productive. Other people I respect and believe have put it to much > better tests than I, and they came to the same conclusion.Ah ... OK ... how narrow is the "same application domain"?> > You don''t have to buy it. You don''t have to believe that astronauts > have been to the moon, either. This one is much easier to prove for > yourself though.Well ... I guess this depends on the answer to my question above. In any event, I''m using an analogy with a similar phenomenon -- a small group of "irrational cheerleaders" claiming vastly improved productivity -- that occurred in the early days of the move of personal computers from the realm of hobbyists to the multi-billion dollar business they are today. I''ve mentioned this before, so I''d like to expound a little more. I''m talking about the FORTH language, and I was one of the fanatics and cheerleaders. Now, in the context of 8-bit microprocessors with 16-bit real-only addressing, small hard disk drives or even floppy-only systems, Microsoft Basic with its dumbed-down syntax and semantics and few C compilers, FORTH was a highly productive environment. I learned it. I used it. I stayed with it through the evolution to 16-bit DOS systems with hard drives and, for the rich, floating-point co-processors. I ran SwiftForth on my first Windows PC. And I know of at least one web site that was written in FORTH. I think it''s still in FORTH. But today, despite its flexibility, despite the band of fanatics and cheerleaders, FORTH lives on only in the exceptionally high quality products of Forth, Inc., a couple of "free" compilers that run in the GNU environment, in embedded systems and special-purpose microprocessor controlled devices, and, of course, in the hearts and minds of us fanatics and cheerleaders. Could I do "Forth on Frames" and make a world-class web 2.0 application development environment using the GNU FORTH (gforth) package? Hell, yes, I could! I don''t consider that a productive use of my time, nor will it help me learn Ruby and Ruby on Rails. :) But can anyone on this list build a microprocessor-controlled robot in Ruby on Rails? Will Ruby even run on a 6502, Z80, 8051 or 80186? Will the Ruby run-time fit in a few kilobytes? Etc., etc., etc. -- M. Edward (Ed) Borasky http://www.borasky-research.net/ http://borasky-research.blogspot.com/ http://pdxneurosemantics.com http://pdx-sales-coach.com http://algocompsynth.com
John Wilger said:> On 8/27/05, Ed Watkeys <edw-tIV1OJqwIcc@public.gmane.org> wrote: >> Please note that I don''t have a problem with Rails, simply the >> irrational cheerleading surrounding it. > > Then again, without the irrational cheerleading, I''d bet that a lot > fewer people would even know about Rails and thus would not be able to > benefit from it. Certainly a company needs to make a rational, > well-informed decision about the platform they use; but _every_ > successful technology/framework/whatever has had its share of > "irrational cheerleaders" which helped it get the attention it needed > to survive among the plethora of options available. > > So the irrational cheerleading is actually a _good_ thing, as long as > you''re smart enough not to pick your platform based solely on the > _quantity_ of the cheerleaders.That''s a good point. Ed -- Transmogrify, LLC * <http://xmog.com>
Kevin Williams said:> > Given two groups of equally skilled, trained and motivated programmers > (Rails for one group and pick your poison for the other) who are > intimately familiar with their respective development environments and > _the same_ application domain, the Rails framework is more productive. I > have tested it myself against another framework I hardly knew so that > familiarity was not a determining factor, and Rails was much more > productive. Other people I respect and believe have put it to much > better tests than I, and they came to the same conclusion. > > You don''t have to buy it. You don''t have to believe that astronauts have > been to the moon, either. This one is much easier to prove for yourself > though.I don''t think this is what "the other Ed" was trying to say. I think his point was simply that choosing a platform should involve a consideration of the availability of commodity labor competent with it. And there''s no need to be rude by comparing Rails skeptics to flat Earth advocates. Given that Rails has only been around for a year and that Java web development has been around seriously for about seven years, it says a lot that Rails experts beat Java experts at solving many sorts of common web development problems. This seemingly makes less important the relative dearth of talented Rails developers, because recent history indicates that people can be converted into talented Rails developers fairly quickly. That said, you are limiting your talent pool, because some CF, Java, Perl, or whatever developers will not consider Rails development positions. Considering the big picture, Paul Graham might say that it doesn''t matter that you''re choosing a relatively obscure technology as long as you can find a few good nerds. I am partial to this argument, but there are places where a few good nerds just aren''t going to fit in, such as in huge dysfunctional corporations that want to treat programmers as faceless commodities. If you''re stuck in such an environment, a few good nerds may be a recipe for disaster. Regards, Ed -- Transmogrify, LLC * <http://xmog.com>
On Aug 29, 2005, at 4:19 AM, Steven wrote:> On Sun, 2005-08-28 at 10:48 -0400, John Wilger wrote: > >> On 8/27/05, Ed Watkeys <edw-tIV1OJqwIcc@public.gmane.org> wrote: >> >>> Please note that I don''t have a problem with Rails, simply the >>> irrational cheerleading surrounding it. >>> >> >> Then again, without the irrational cheerleading, I''d bet that a lot >> fewer people would even know about Rails and thus would not be >> able to >> benefit from it. Certainly a company needs to make a rational, >> well-informed decision about the platform they use; but _every_ >> successful technology/framework/whatever has had its share of >> "irrational cheerleaders" which helped it get the attention it needed >> to survive among the plethora of options available. >> >> So the irrational cheerleading is actually a _good_ thing, as long as >> you''re smart enough not to pick your platform based solely on the >> _quantity_ of the cheerleaders. >> > > Is it irrational cheer leading because no one is paid to do the cheer > leading or by another name "Marketing"?Are you saying that anyone who isn''t a Rails cheerleader is a slave to Microsoft''s, Sun''s, or Macromedia''s marketing efforts? If you didn''t mean that, please excuse me. I''m having difficulty inferring the argument in your innuendo. Regards, Ed -- Transmogrify, LLC * <http://xmog.com/>
John Wilger wrote:> On 8/27/05, Ed Watkeys <edw-tIV1OJqwIcc@public.gmane.org> wrote: > >>Please note that I don''t have a problem with Rails, simply the >>irrational cheerleading surrounding it. > > > Then again, without the irrational cheerleading, I''d bet that a lot > fewer people would even know about Rails and thus would not be able to > benefit from it.Well, comparing irrational cheerleading versus *no* cheerleading, sure. But Rails offers a lot of reason for rational cheerleading, which (assuming for the sake of argument that this hasn''t actually been the case) might have attracted more people, given that several people have expressed disdain for Rails simply because of alleged fanboy behavior. It''s hard to argue that the noise about Rails hasn''t attracted a great many eyeballs (witness Tim Bray''s recent Web post[0], where he refers to Ruby as "That Language".) But when people criticize the hype they are likely suggesting that one could have achieved the same or better results with fewer irrational claims. (And when Mr. Bray is ready to come around, perhaps I''ll get him a nice new T-shirt: http://www.cafepress.com/javarehab.29832201 or http://www.cafepress.com/pythonrehab.29833445 And, yes, them''s More RubyStuff Items ) James [0] http://www.tbray.org/ongoing/When/200x/2005/08/27/Ruby -- http://www.ruby-doc.org - The Ruby Documentation Site http://www.rubyxml.com - News, Articles, and Listings for Ruby & XML http://www.rubystuff.com - The Ruby Store for Ruby and Rails Stuff http://www.jamesbritt.com - Playing with Better Toys
On 8/29/05, Ed Watkeys <edw-tIV1OJqwIcc@public.gmane.org> wrote:> Considering the big picture, Paul Graham might say that it doesn''t matter > that you''re choosing a relatively obscure technology as long as you can > find a few good nerds. I am partial to this argument, but there are places > where a few good nerds just aren''t going to fit in, such as in huge > dysfunctional corporations that want to treat programmers as faceless > commodities. If you''re stuck in such an environment, a few good nerds may > be a recipe for disaster.Kinda reminds me of how I''ve felt being a FreeBSD user for so long (as apposed to a Linux user). Now I''m a FreeBSD, and a Ruby on Rails user! James
Ed Watkeys wrote:> >Considering the big picture, Paul Graham might say that it doesn''t matter >that you''re choosing a relatively obscure technology as long as you can >find a few good nerds. I am partial to this argument, but there are places >where a few good nerds just aren''t going to fit in, such as in huge >dysfunctional corporations that want to treat programmers as faceless >commodities. If you''re stuck in such an environment, a few good nerds may >be a recipe for disaster. > >Nerds of a feather flock together. A few good nerds will, if lucky, hook up with a good attorney and a good accountant and start a small business. They''ll find a defendable market niche, grow a business, and then at some point they''ll attract the attention of a huge dysfunctional corporation. The HDC will either buy them or turn loose an army of inexpensive nerds to create something a little better and a *lot* less expensive, which they will then sell until NOAF is driven out of business. Unless NOAF is stupid enough to sue HDC. Then NOAF will go out of business *much* faster. :( Wait a minute ... I''m the other Ed? I thought *you* were the other Ed. -- M. Edward (Ed) Borasky http://www.borasky-research.net/ http://borasky-research.blogspot.com/ http://pdxneurosemantics.com http://pdx-sales-coach.com http://algocompsynth.com
On 8/17/05, Philip Edelbrock <phil-RqHDiG/X+WF8uvyFNTHIBg@public.gmane.org> wrote:> > > Chad Lung wrote: > > It seems like everyone now has a RoR type framework: > > > > Perl: > > http://dev.catalyst.perl.org/ > > > > Speaking of which, what do you think of the comparison table here: > > http://dev.catalyst.perl.org/wiki/VersusRails ><blatant_self_publicity> I did a comparison of django and rails a while back that was well recieved (by most!) but I think I forgot to post it: http://www.magpiebrain.com/archives/2005/08/14/rails_and_django </blatant_self_publicity> And does anyone else think it''s kind of cool that Rails, Django and Catalyst all use Trac? I really hope someone doesn''t decide to rewrite it in Rails just because it''s a Python app! Oh, and I vote for Haskell on Horse to be the next port... -- sam http://www.magpiebrain.com/
On Tue, 2005-08-30 at 00:01 -0400, Ed Watkeys wrote:> On Aug 29, 2005, at 4:19 AM, Steven wrote: > > > On Sun, 2005-08-28 at 10:48 -0400, John Wilger wrote: > > > >> On 8/27/05, Ed Watkeys <edw-tIV1OJqwIcc@public.gmane.org> wrote: > >> > >>> Please note that I don''t have a problem with Rails, simply the > >>> irrational cheerleading surrounding it. > >>> > >> > >> Then again, without the irrational cheerleading, I''d bet that a lot > >> fewer people would even know about Rails and thus would not be > >> able to > >> benefit from it. Certainly a company needs to make a rational, > >> well-informed decision about the platform they use; but _every_ > >> successful technology/framework/whatever has had its share of > >> "irrational cheerleaders" which helped it get the attention it needed > >> to survive among the plethora of options available. > >> > >> So the irrational cheerleading is actually a _good_ thing, as long as > >> you''re smart enough not to pick your platform based solely on the > >> _quantity_ of the cheerleaders. > >> > > > > Is it irrational cheer leading because no one is paid to do the cheer > > leading or by another name "Marketing"? > > Are you saying that anyone who isn''t a Rails cheerleader is a slave > to Microsoft''s, Sun''s, or Macromedia''s marketing efforts? If you > didn''t mean that, please excuse me. I''m having difficulty inferring > the argument in your innuendo.No, I didn''t infer anything about any ones leanings, just that the cheerleading isn''t any different from other groups marketing efforts. Very little marketing hype is rational. -- Steven Critchfield critch-wQLwMjUOumVBDgjK7y7TUQ@public.gmane.org KI4KTY
On 8/31/05, Sam Newman <sam.newman-Re5JQEeQqe8AvxtiuMwx3w@public.gmane.org> wrote:> On 8/17/05, Philip Edelbrock <phil-RqHDiG/X+WF8uvyFNTHIBg@public.gmane.org> wrote:> And does anyone else think it''s kind of cool that Rails, Django and > Catalyst all use Trac? I really hope someone doesn''t decide to rewrite > it in Rails just because it''s a Python app!http://collaboa.org/
> >> On 8/17/05, Philip Edelbrock <phil-RqHDiG/X+WF8uvyFNTHIBg@public.gmane.org> wrote: >> > > >> And does anyone else think it''s kind of cool that Rails, Django and >> Catalyst all use Trac? I really hope someone doesn''t decide to >> rewrite >> it in Rails just because it''s a Python app! >>Hi, I didn''t write that (not that I have any huge objections). Just get your quoting right, please. -Phil
On 8/29/05, Ed Watkeys <edw-tIV1OJqwIcc@public.gmane.org> wrote:> That said, you are limiting your talent pool, because some CF, Java, Perl, > or whatever developers will not consider Rails development positions. >Heh. My experience has been the converse of that. My company was recently hiring for a web developer as we begin a ground-up redevelopment of our e-commerce site. I tried really hard to convince management that we should use Rails, but to no avail (so It''ll be PHP5). One of their biggest fears with Rails was that they would not be able to hire people with Rails/Ruby knowledge. Turns out, out of the resumes we received, only about 3 were halfway qualified developers. One didn''t bother to show up for his interview, one called the day before his interview and said he had accepted another position, and we ended up hiring the third guy pretty much by default. On the other hand, I''ve had quite a few well-qualified people tell me that they weren''t even interested in sending us a resume because they were only looking for Rails work now. So much for the "we won''t be able to hire talented developers" FUD. -- Regards, John Wilger ----------- Alice came to a fork in the road. "Which road do I take?" she asked. "Where do you want to go?" responded the Cheshire cat. "I don''t know," Alice answered. "Then," said the cat, "it doesn''t matter." - Lewis Carrol, Alice in Wonderland