This coming Monday there is going to be a meeting in my company on whether to allow the deployment of Ruby and RoR. Its very easy for me to show dramatic produtivity improvements comparing RoR with Java web apps. But, Python is entrenched and its advocates can also show an improvement over Java. Supporting Ruby across enterprise won''t happen if all I can show is Ruby is roughly equivalent to the Python that is already supported. Keep in mind I''m talking about web apps with Rails, not general Ruby programming. Is there anyone out that whose has come to Rails development from Python who could share the comparative insights? Obviously hard numbers would be best, but I doubt that any exist. So, just your story, insights, and opinions are more than welcomed. Thanks (in advance), Curt _______________________________________________ Rails mailing list Rails-1W37MKcQCpIf0INCOvqR/iCwEArCW2h5@public.gmane.org http://lists.rubyonrails.org/mailman/listinfo/rails
Python isn''t a framework and Rails isn''t a language. The two cannot be compared. -- -- Tom Mornini On Oct 19, 2005, at 11:24 AM, Curt Hibbs wrote:> This coming Monday there is going to be a meeting in my company on > whether to allow the deployment of Ruby and RoR. Its very easy for > me to show dramatic produtivity improvements comparing RoR with > Java web apps. > > But, Python is entrenched and its advocates can also show an > improvement over Java. Supporting Ruby across enterprise won''t > happen if all I can show is Ruby is roughly equivalent to the > Python that is already supported. Keep in mind I''m talking about > web apps with Rails, not general Ruby programming. > > Is there anyone out that whose has come to Rails development from > Python who could share the comparative insights? Obviously hard > numbers would be best, but I doubt that any exist. So, just your > story, insights, and opinions are more than welcomed. > > Thanks (in advance), > Curt > _______________________________________________ > Rails mailing list > Rails-1W37MKcQCpIf0INCOvqR/iCwEArCW2h5@public.gmane.org > http://lists.rubyonrails.org/mailman/listinfo/rails >
Curt Hibbs wrote:> Is there anyone out that whose has come to Rails development from Python > who could share the comparative insights? Obviously hard numbers would > be best, but I doubt that any exist. So, just your story, insights, and > opinions are more than welcomed.This post isn''t going to help your case.. But here goes anyway. We use Python where I work. Primarily for build tools (based on SCons) and data processing tools. We make video games for the PSP. Building a game like this (250k SLOC in 3k source files, plus 11k data files) is complicated, and SCons mostly works well. Because SCons is in Python, it''s been really easy to include more Python functionality, for instance, to parse textual data at build time, and output game-ready data as a result. Backing up a bit, a few years ago we all discovered Python and really liked its rich library, relatively clean syntax, and ease of productivity. However, as time passed, I became frustrated - I''d coded Perl for a LONG time. And, well, try dealing with XML and RegEx in Python, and it just felt clunky and awkward. Earlier this year, I discovered Ruby, and Rails. I converted a few scripts from Python (specifically, for calculation of Glicko2 ratings) as an experiment. They weren''t that different in the end, but the Ruby one was easier and more fun to code. So, I plan to integrate Ruby into some of our tools this development cycle. Why? It just feels right - the elegance factor is extremely high, which makes programming fun. Keeping programming fun == sanity. For me, Ruby''s stdlib is "rich enough" (Python''s is actually a little out of control!), and the language is beautiful. Best thing you could do would be to ask your coders to at least TRY the language before dismissing it. THEN you can compare and contrast the two. You need to ask everyone to open their minds, and leave the religion at home! Like I said in the beginning, this isn''t going to help you. But, that''s my story! --Steve
On 10/19/05, Tom Mornini <tmornini-W/9V78bTXriB+jHODAdFcQ@public.gmane.org> wrote:> Python isn''t a framework and Rails isn''t a language.Tom, Curt is very likely to be aware of this more than most people who have any awareness of Rails at all. His PHBs aren''t. More than that, the request is a useful one. He''s pushing for Rails development at his company, and he''s fighting against Python -- and its myriad frameworks of varying quality. It would be ideal if he could convince the PTB that Ruby is worthwhile in and of itself, and allow for Rails development that way, but the very task set before him is Sysiphean.> The two cannot be compared.Yet ... they can. They can be compared via direct comparisons and failings. That is, Rails can be pointed out that, since it''s written in and on Ruby, it is more likely to be something that the design staff can use effectively without having to worry about space indentation issues as one finds in most Python web situations. That''s just off the top of my head. Of course, I''m only responding because I know Curt -- and know he''s been around Ruby longer than I have. He knows the difference. -austin -- Austin Ziegler * halostatue-Re5JQEeQqe8AvxtiuMwx3w@public.gmane.org * Alternate: austin-/yODNl0JVVCozMbzO90S/Q@public.gmane.org
Tom Mornini wrote:> Python isn''t a framework and Rails isn''t a language.I''m pretty sure Curt knows that, but I was also pretty sure any sentient being could figure out his actual question.> > The two cannot be compared.Yeah, well, let''s try anyway. Those arguing for Python can bring up one or another toolkit or framework, such as Zope or Django or whatever, as their ''Rails''. Unless they already have experience with this tool there will be a learning curve. Probably less of a curve than learning Ruby on Rails, though, which requires learning the Rails DSL and enough Ruby to be useful. I think a straight up comparison (learn new tool Y for known language X, or learn new tool Y'' for new language X'') is a bad move. There needs to be something persuasive to show that using Rails, learning curve and all, will provide better results: faster time to delivery, more maintainable code, code that is easier to fix/enhance/extend. Sadly, I have no background to argue for this. :( The devil may be in the details. Are you building apps where you can create the DB schema from scratch (and so have no issues going with Rails conventions)? How much say do you have in Web server options? Can you get support for fastcgi or scgi, or will the choice be mod_python vs Ruby CGI? It may very well be that Python is the better choice for a given set of circumstances. Pushing for Rails under unfavorable conditions may just lead to a big disappointment. James -- 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 Stuff http://www.jamesbritt.com - Playing with Better Toys
Would it be possible to take some of the Python frameworks (TurboGears for example) and compare time/lines of code written to code the same simple project? If you take the TurboGears video (http://turbogears.org) for example you can see the difference. -Kenny On 10/19/05, Austin Ziegler <halostatue-Re5JQEeQqe8AvxtiuMwx3w@public.gmane.org> wrote:> > On 10/19/05, Tom Mornini <tmornini-W/9V78bTXriB+jHODAdFcQ@public.gmane.org> wrote: > > Python isn''t a framework and Rails isn''t a language. > > Tom, > > Curt is very likely to be aware of this more than most people who have > any awareness of Rails at all. His PHBs aren''t. More than that, the > request is a useful one. He''s pushing for Rails development at his > company, and he''s fighting against Python -- and its myriad frameworks > of varying quality. It would be ideal if he could convince the PTB > that Ruby is worthwhile in and of itself, and allow for Rails > development that way, but the very task set before him is Sysiphean. > > > The two cannot be compared. > > Yet ... they can. They can be compared via direct comparisons and > failings. That is, Rails can be pointed out that, since it''s written > in and on Ruby, it is more likely to be something that the design > staff can use effectively without having to worry about space > indentation issues as one finds in most Python web situations. That''s > just off the top of my head. > > Of course, I''m only responding because I know Curt -- and know he''s > been around Ruby longer than I have. He knows the difference. > > -austin > -- > Austin Ziegler * halostatue-Re5JQEeQqe8AvxtiuMwx3w@public.gmane.org > * Alternate: austin-/yODNl0JVVCozMbzO90S/Q@public.gmane.org > _______________________________________________ > 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 Oct 19, 2005, at 11:53 AM, Austin Ziegler wrote:> On 10/19/05, Tom Mornini <tmornini-W/9V78bTXriB+jHODAdFcQ@public.gmane.org> wrote: > >> Python isn''t a framework and Rails isn''t a language. > > Curt is very likely to be aware of this more than most people who have > any awareness of Rails at all. His PHBs aren''t. More than that, the > request is a useful one. He''s pushing for Rails development at his > company, and he''s fighting against Python -- and its myriad frameworks > of varying quality. It would be ideal if he could convince the PTB > that Ruby is worthwhile in and of itself, and allow for Rails > development that way, but the very task set before him is Sysiphean.You don''t fight ignorance with ignorance. My point is that Curt didn''t even frame the question correctly here, so how can he hope to present the answer properly to his PHBs.>> The two cannot be compared. > > Yet ... they can. They can be compared via direct comparisons and > failings. That is, Rails can be pointed out that, since it''s written > in and on Ruby, it is more likely to be something that the design > staff can use effectively without having to worry about space > indentation issues as one finds in most Python web situations. That''s > just off the top of my head. > > Of course, I''m only responding because I know Curt -- and know he''s > been around Ruby longer than I have. He knows the difference.And yet, they can''t. You can compare Python and it''s frameworks to Ruby on Rails, but not Python to Rails. And that''s the exact way he should compare them. -- -- Tom Mornini
Thanks, I''ll take a look at the video and see if I can extract any useful observations. Curt On 10/19/05, Kenneth Parnell <k.parnell-Re5JQEeQqe8AvxtiuMwx3w@public.gmane.org> wrote:> > Would it be possible to take some of the Python frameworks (TurboGears for > example) and compare time/lines of code written to code the same simple > project? If you take the TurboGears video ( http://turbogears.org) for > example you can see the difference. > > -Kenny > > On 10/19/05, Austin Ziegler < halostatue-Re5JQEeQqe8AvxtiuMwx3w@public.gmane.org> wrote: > > > > On 10/19/05, Tom Mornini < tmornini-W/9V78bTXriB+jHODAdFcQ@public.gmane.org> wrote: > > > Python isn''t a framework and Rails isn''t a language. > > > > Tom, > > > > Curt is very likely to be aware of this more than most people who have > > any awareness of Rails at all. His PHBs aren''t. More than that, the > > request is a useful one. He''s pushing for Rails development at his > > company, and he''s fighting against Python -- and its myriad frameworks > > of varying quality. It would be ideal if he could convince the PTB > > that Ruby is worthwhile in and of itself, and allow for Rails > > development that way, but the very task set before him is Sysiphean. > > > > > The two cannot be compared. > > > > Yet ... they can. They can be compared via direct comparisons and > > failings. That is, Rails can be pointed out that, since it''s written > > in and on Ruby, it is more likely to be something that the design > > staff can use effectively without having to worry about space > > indentation issues as one finds in most Python web situations. That''s > > just off the top of my head. > > > > Of course, I''m only responding because I know Curt -- and know he''s > > been around Ruby longer than I have. He knows the difference. > > > > -austin > > -- > > Austin Ziegler * halostatue-Re5JQEeQqe8AvxtiuMwx3w@public.gmane.org > > * Alternate: austin-/yODNl0JVVCozMbzO90S/Q@public.gmane.org > > _______________________________________________ > > 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 > > >_______________________________________________ Rails mailing list Rails-1W37MKcQCpIf0INCOvqR/iCwEArCW2h5@public.gmane.org http://lists.rubyonrails.org/mailman/listinfo/rails
On 10/19/05, Kenneth Parnell <k.parnell-Re5JQEeQqe8AvxtiuMwx3w@public.gmane.org> wrote:> Would it be possible to take some of the Python frameworks (TurboGears for > example) and compare time/lines of code written to code the same simple > project? If you take the TurboGears video ( http://turbogears.org) for > example you can see the difference.Anyone can rehearse and write a basic wiki/blog/todolist in 20-2(!) minutes. Django is very nice though. Work with it a bit and you will see a proven platform for deploying a publishing oriented sites and a superb generated admin interface. But, you also miss things like integrated unit tests baked into the framework, migrations and database-agnostic schemas, routes, and a clean syntax for specifying your domain model. Let''s pit two competing taggable modules head to head, shall we? https://simon.bofh.ms/cgi-bin/trac-django-projects.cgi/wiki/AbstractTagging http://dema.ruby.com.br/articles/2005/09/03/tagging-on-steroids-with-rails I was perfectly happy coding in Python before, in a similar MVC web framework. Hell I was even content with ASP.Net (especially with 2.0 on the horizon). But, I get a better vibe from Rails. Don''t just show them code though, have them spend a day or so on it. I admit, the first time I saw ruby syntax I cringed a bit (the blocks were freaky), and totally ignored the first whisperings of Rails that I came across. -- rick http://techno-weenie.net
Rick, you are just the person I need to talk to (since you have actually done web programming in Python). What I am going to be faced with is convincing management types (not programmers) that RoR has enough of a productivity advantage of some (any) Python web framework to justify the (perceived) support costs of deploying Ruby/Rails on our servers across the enterprise. Python is already a supported solution. The stated goal of corporate IT is to *reduce* the variety of installed software, not increase it. They are going to look at the RoR request and say: "You can do this in Python, we don''t see any reason to add Ruby". Programming language features and even framework features won''t make any difference, except for when they increase productivity. So, what can you tell me about your web development experiences in Python frameworks vs Rails. What are the frustrations you were glad to leave behind.\? What is it about Rails that makes you more productive? Thanks, Curt On 10/19/05, Rick Olson <technoweenie-Re5JQEeQqe8AvxtiuMwx3w@public.gmane.org> wrote:> > On 10/19/05, Kenneth Parnell <k.parnell-Re5JQEeQqe8AvxtiuMwx3w@public.gmane.org> wrote: > > Would it be possible to take some of the Python frameworks (TurboGears > for > > example) and compare time/lines of code written to code the same simple > > project? If you take the TurboGears video ( http://turbogears.org) for > > example you can see the difference. > > Anyone can rehearse and write a basic wiki/blog/todolist in 20-2(!) > minutes. > > Django is very nice though. Work with it a bit and you will see a > proven platform for deploying a publishing oriented sites and a superb > generated admin interface. > > But, you also miss things like integrated unit tests baked into the > framework, migrations and database-agnostic schemas, routes, and a > clean syntax for specifying your domain model. > > Let''s pit two competing taggable modules head to head, shall we? > > > https://simon.bofh.ms/cgi-bin/trac-django-projects.cgi/wiki/AbstractTagging > http://dema.ruby.com.br/articles/2005/09/03/tagging-on-steroids-with-rails > > I was perfectly happy coding in Python before, in a similar MVC web > framework. Hell I was even content with ASP.Net (especially with 2.0 > on the horizon). But, I get a better vibe from Rails. Don''t just > show them code though, have them spend a day or so on it. I admit, > the first time I saw ruby syntax I cringed a bit (the blocks were > freaky), and totally ignored the first whisperings of Rails that I > came across. > > -- > rick > http://techno-weenie.net > _______________________________________________ > 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
> Python is already a supported solution. The stated goal of corporate IT is > to *reduce* the variety of installed software, not increase it. They are > going to look at the RoR request and say: "You can do this in Python, we > don''t see any reason to add Ruby".Fair enough.> Programming language features and even > framework features won''t make any difference, except for when they increase > productivity.Well, again, fair enough, but that''s could be a tough thing to talk about without knowing the kinds of projects you''re working on, how much tested code you typically reuse in your organization and how many custom tools you''ve built for your own environment and the sorts of things you usually take on.> So, what can you tell me about your web development experiences in Python > frameworks vs Rails. What are the frustrations you were glad to leave > behind.\? What is it about Rails that makes you more productive?Using Python I''ve coded and deployed apps with Webware and CherryPy, using SQLObject for persistence. I''ve fooled around with Django, too, and I like it a lot. For how new Django is, it''s already got a lot of polish, with plenty more likely to come. I started using Rails, though, because I got about three times the amount of work done per workday than with Python and its frameworks (although I''ve never put it up against Django). That, I will admit, is a completely subjective measurement. But I think it has a lot to do with the fact that Rails just seems to cover all the bases where Web applications are concerned (in no particular order): * fully integrated testing I love this. Using the testing features of Rails along with the Rails console, debugging apps has been cut WAY down. The time savings there alone has been, well, more than enough justification. It''s nothing that can''t be done in other languages or with other frameworks, but because Rails provides it cleanly, I find it easier to work with from app to app. With Python, I found myself tweaking my project setups from app to app, which was fine but also wasteful since time was spent with playing around with and optimizing setups rather than coding and getting the job done. * very slick ajax integration Again, because the calls are Just There, and since they''re mostly what I need, I spend more time getting cool interfaces written rather than scouring docs, playing with setups, messing with my own approaches, and admiring my own brilliance. Don''t get me wrong, I like to play with code and come up with my own approaches (and admire my own brilliance), but I''ve got plenty of other projects to play around with rather than getting sidetracked with client work. * a common templating engine, but easy to add others, if necessary It *is* easy to add other templating engines, but I haven''t found it necessary. In the Python world the Kid templating framework has gained a lot of ground, and proponents of Kid will level some solid shots at eruby. Whatever. I find eruby solid and I think it''s worth pointing out that there''s really no way that I''ve seen to avoid including code in templates. You can disguise it as special HTML tags which validate, but at some level it''s still code. Cheetah templates for Python are just Velocity templates from Java, but they''re really good, too. Again, multiple approaches. I find the fact that eruby''s included with Rails simplifies matters for me. I''ve had no complaints so far, either from other developers or clients. * activerecord''s convention-based approach I like ActiveRecord a lot. If you''re used to SQLObject, then ActiveRecord is slam dunk easy. A little configuration, five minutes, you''re done. Everything else is handled for you, from transactions to validations to model relationships, once you''ve added a few lines to your model files. Sweet. Ask any Python programmer used to SQLObject to give ActiveRecord a go, and tell you if it isn''t a productivity improver. Other things, too ... * routes, include named routes Django''s got something like routes, based on regexes, but it isn''t nearly as slick. * api building framework A feature that I haven''t found talked about much with Rails, but which I''ve found useful recently. * layouts * other cool javascript support: prototype and scriptaculous * ruby God, the list goes on. Just look at the API docs and then show me the Python framework that has *all* the stuff Rails has right here and now. Why not try to convince the powers that be to adopt one or two test projects? Let the results speak for themselves. Just my .02. Best of luck, Greg
Curt Hibbs wrote:> > What I am going to be faced with is convincing management types (not > programmers) that RoR has enough of a productivity advantage of someWarning: this has NOTHING to do with Rails. :) One more thing Curt. Have you read [How to Argue and Win][1] by [Gerry Spence][2]? If not, I suggest you digest it before your pitch. I read it years ago and found it valuable. Though, it looks like he has updated it just this year, now called [Win Your Case: ...][3]. I''ll have to pick that up! Also usually highly recommended, but one I haven''t gotten to myself is called [Getting to Yes: ...][4] and perhaps [Getting Past No: ...][5], both by [William Ury][6]. Finally, this post, [How to Sell Your Boss][7] on [Working Smart][8] seems like very sound advice! HTH! --Steve [1]: http://www.amazon.com/exec/obidos/tg/detail/-/0312144776/qid=1129782156/sr=8-1/ref=pd_bbs_1/104-1033675-8397553?v=glance&s=books&n=507846 [2]: http://en.wikipedia.org/wiki/Gerry_Spence [3]: http://www.amazon.com/exec/obidos/tg/detail/-/0312338813/ref=pd_sim_b_2/104-1033675-8397553?%5Fencoding=UTF8&v=glance [4]: http://www.amazon.com/exec/obidos/tg/detail/-/0140157352/qid=1129782349/sr=2-1/ref=pd_bbs_b_2_1/104-1033675-8397553?v=glance&s=books [5]: http://www.amazon.com/exec/obidos/tg/detail/-/0553371312/ref=pd_sim_b_1/104-1033675-8397553?%5Fencoding=UTF8&v=glance [6]: http://www.amazon.com/exec/obidos/search-handle-url/ref=pd_sim_b_1/104-1033675-8397553?%5Fencoding=UTF8&search-type=ss&index=books&field-author=William%20Ury [7]: http://michaelhyatt.blogs.com/workingsmart/2004/11/how_to_sell_you.html [8]: http://www.michaelhyatt.com/workingsmart/
-----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 On Oct 19, 2005, at 3:56 PM, Curt Hibbs wrote:> Rick, you are just the person I need to talk to (since you have > actually done web programming in Python). > > What I am going to be faced with is convincing management types > (not programmers) that RoR has enough of a productivity advantage > of some (any) Python web framework to justify the (perceived) > support costs of deploying Ruby/Rails on our servers across the > enterprise.There shouldn''t be much support costs for Ruby / Rails. The Rails stuff would be packaged along with the application. Ruby should be installed on all the machines anyways. :-)> Python is already a supported solution. The stated goal of > corporate IT is to *reduce* the variety of installed software, not > increase it. They are going to look at the RoR request and say: > "You can do this in Python, we don''t see any reason to add Ruby". > Programming language features and even framework features won''t > make any difference, except for when they increase productivity. >Hype unit testing. Hype functional testing. Hype ajax helpers. Hype the english-like AR (belongs_to :project, etc). Hype the fact that Rails forces you to develop using MVC. Hype the fact that all Rails applications are structured very similarly (reduces variation!). Hype easy web service integration (so that any application can use the AR models). Bullet point: I added integration to the company''s single web sign-on system in Instiki in just a dozen or so lines of code. Might be worth a mention that some of the big, important Java people are switching to Rails.> So, what can you tell me about your web development experiences in > Python frameworks vs Rails. What are the frustrations you were glad > to leave behind.\? What is it about Rails that makes you more > productive? >Can''t help you here, never used a Python framework. Joe> On 10/19/05, Rick Olson <technoweenie-Re5JQEeQqe8AvxtiuMwx3w@public.gmane.org> wrote: On > 10/19/05, Kenneth Parnell <k.parnell-Re5JQEeQqe8AvxtiuMwx3w@public.gmane.org> wrote: > > Would it be possible to take some of the Python frameworks > (TurboGears for > > example) and compare time/lines of code written to code the same > simple > > project? If you take the TurboGears video ( http:// > turbogears.org) for > > example you can see the difference. > > Anyone can rehearse and write a basic wiki/blog/todolist in 20-2(!) > minutes. > > Django is very nice though. Work with it a bit and you will see a > proven platform for deploying a publishing oriented sites and a superb > generated admin interface. > > But, you also miss things like integrated unit tests baked into the > framework, migrations and database-agnostic schemas, routes, and a > clean syntax for specifying your domain model. > > Let''s pit two competing taggable modules head to head, shall we? > > https://simon.bofh.ms/cgi-bin/trac-django-projects.cgi/wiki/ > AbstractTagging > http://dema.ruby.com.br/articles/2005/09/03/tagging-on-steroids- > with-rails > > I was perfectly happy coding in Python before, in a similar MVC web > framework. Hell I was even content with ASP.Net (especially with 2.0 > on the horizon). But, I get a better vibe from Rails. Don''t just > show them code though, have them spend a day or so on it. I admit, > the first time I saw ruby syntax I cringed a bit (the blocks were > freaky), and totally ignored the first whisperings of Rails that I > came across.Joe Van Dyk http://www.pinkpucker.net -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.4.1 (Darwin) iD4DBQFDVzR8sWh6/7z1gt4RAv/1AJjv8o+ZRwGIbbunAbcov0ZfiovsAKCAt5+N D+0bXXO8Em4Qn+41he1pJw==9DQp -----END PGP SIGNATURE-----
> Python is already a supported solution. The stated goal of corporate IT is > to *reduce* the variety of installed software, not increase it. They are > going to look at the RoR request and say: "You can do this in Python, we > don''t see any reason to add Ruby". Programming language features and even > framework features won''t make any difference, except for when they increase > productivity.This is not an issue of whether so-and-so "can be done" with Rails or any Python framework. This is an issue of programmer preference and that, usually, and perhaps especially in the case of Rails, depends on the "feel" of working with the language. To use a stupid analogy, you can drive a nail through a board of wood using a shoe, it "can be done". So why use a hammer? Not because it adds some new possibility, but because it makes the job simpler, easier, more straightforward. It makes it more fun, it makes you feel better because you''re working with better tools. It makes you enjoy your job. That''s the argument you need to make, not that Rails does something that some other tool can''t (because it doesn''t). Hope that helps. Best regards, Tomas Jogin
On 10/20/05, Tomas Jogin <tomasj-Re5JQEeQqe8AvxtiuMwx3w@public.gmane.org> wrote:> > > Python is already a supported solution. The stated goal of corporate IT > is > > to *reduce* the variety of installed software, not increase it. They are > > going to look at the RoR request and say: "You can do this in Python, we > > don''t see any reason to add Ruby". Programming language features and > even > > framework features won''t make any difference, except for when they > increase > > productivity. > > This is not an issue of whether so-and-so "can be done" with Rails or > any Python framework. This is an issue of programmer preference and > that, usually, and perhaps especially in the case of Rails, depends on > the "feel" of working with the language. > > To use a stupid analogy, you can drive a nail through a board of wood > using a shoe, it "can be done". So why use a hammer? Not because it > adds some new possibility, but because it makes the job simpler, > easier, more straightforward. It makes it more fun, it makes you feel > better because you''re working with better tools. It makes you enjoy > your job. That''s the argument you need to make, not that Rails does > something that some other tool can''t (because it doesn''t). > > Hope that helps.Its a good analogy. I agree the "it can be done" argument won''t lead anywhere and I wouldn''t use it. The main selling point is going to be increased productivity, backed with reasons which that increase exists. I still need to collect more anectdotal evidence for this. Come on... there''s got to be more ex-Python hackers out there in the Rails crowd... tell us your story! Curt _______________________________________________ Rails mailing list Rails-1W37MKcQCpIf0INCOvqR/iCwEArCW2h5@public.gmane.org http://lists.rubyonrails.org/mailman/listinfo/rails
> Just my .02. Best of luck, > GregWow, I think Greg summed up what I was going to say already. I really think that the included unit testing is a *huge* plus for Ruby on Rails. Sure, Python has its own wonderful unit testing suites, but Django does not set it up for you. It''s easy to fall into the trap of thinking you''ll set it up later. Also, functional testing in Rails is superb. There''s nothing like deploying an app and knowing that it''s well tested. I don''t know how much more I can add without coming off as a total rails fanboy (well, more than I already do). -- rick http://techno-weenie.net
Dnia Wed, 19 Oct 2005 13:24:23 -0500, Curt Hibbs napisał(a):> Is there anyone out that whose has come to Rails development from Python who > could share the comparative insights? Obviously hard numbers would be best, > but I doubt that any exist. So, just your story, insights, and opinions are > more than welcomed.I have been working with Pythonic frameworks for some years. I have also experience with Rails (some months). I would not repeat good arguments for Rails. Instead of it I would like to share light other pro i cons not mentioned before. - Pythonic frameworks are generall significantly faster than Rails. Python is in general faster than Ruby. E.g. Myghty (www.myghty.org) is much faster than Rails. It is even faster than php5.1. - If you want to prepare offline (CD-only) version of your site (for instance: for marketing guys) every Pythonic builed httpd server is much faster than Webrick. In our company we had to prepare special (CD-only) edition of our web sites. We used Webware and CherryPy and it worked fine. Webware and CherryPy and others can use threads. It may be important for windows platform where there is no forks. Now almost every pythonic framework can work as WSGI and it is no problem to use it as threade or forked server. - If you want to use different multilangugal environment, Python has excelent unicode library. Ruby has very poor unicode support and it's iconv uses is useless for character conversion since it deletes texts when it finds some problematic characters. Eg. there is no way for iconv to conversion from cp1250 to iso-8859-2 without problems. Python is able to exchange such problematic characters (I mean: characters without latin2 equivalent) to xml entity so it can be displayed correctly. - But for intranets based on win32 when sql server plays main role, it is difficult for Python to uses ORM. There is no good ORM for Python when you have to deal with mssql. SQLObject has serious bugs in supporting mssql server. PyDO2 says it can support mssql, but I have no idea how, because there is almost no documentation at all. Django has no mssql support at all either. On the other side, Active Record (used in Rails) works fine with mssql (although its scaffolding seems to be broken when tables do not follow Rails name conventions) - Python has more and betters IDE tools than Ruby. They have better code completion than any other IDE for Ruby. Although there is maybe one good IDE for for Ruby & Rails: rachno IDE. It is commercial and it has no utf8 support yet, but I like it very much. -- Jaroslaw Zabiello _______________________________________________ Rails mailing list Rails-1W37MKcQCpIf0INCOvqR/iCwEArCW2h5@public.gmane.org http://lists.rubyonrails.org/mailman/listinfo/rails
On 10/20/05, Rick Olson <technoweenie-Re5JQEeQqe8AvxtiuMwx3w@public.gmane.org> wrote:> > > Just my .02. Best of luck, > > Greg > > Wow, I think Greg summed up what I was going to say already. I really > think that the included unit testing is a *huge* plus for Ruby on > Rails. Sure, Python has its own wonderful unit testing suites, but > Django does not set it up for you. It''s easy to fall into the trap of > thinking you''ll set it up later. Also, functional testing in Rails is > superb. There''s nothing like deploying an app and knowing that it''s > well tested. > > I don''t know how much more I can add without coming off as a total > rails fanboy (well, more than I already do).Rick, I''m glad that you think Greg was spot-on in his assessment, but I have one request. Could you at least make another post and corroborate Greg''s 3x productivity improvement (or whatever number you feel is appropriate). The reason I ask this is because I am collecting URLs (from gmane.org <http://gmane.org>) of postings that support my assertions. Its great that Greg provided my with solid support, but he''s just one guy relaying one set of experience on the productivity advantages of Rails. That''s good. But not as good as two guys, or even more! Really all I would need would be how much more productive you feel you are in Rails, and what have you used in Python to validate your productivity opinion. Thanks so much, Curt _______________________________________________ Rails mailing list Rails-1W37MKcQCpIf0INCOvqR/iCwEArCW2h5@public.gmane.org http://lists.rubyonrails.org/mailman/listinfo/rails
Curt Hibbs wrote:> This coming Monday there is going to be a meeting in my company onWell.. How did it go? --Steve
On 10/25/05, Stephen Waits <steve-g8GSkY9QmIteoWH0uzbU5w@public.gmane.org> wrote:> Curt Hibbs wrote: > > This coming Monday there is going to be a meeting in my company on > > Well.. How did it go?The meeting was moved to today, I think.
On Oct 25, 2005, at 11:31 AM, Joe Van Dyk wrote:> On 10/25/05, Stephen Waits <steve-g8GSkY9QmIteoWH0uzbU5w@public.gmane.org> wrote: > >> Curt Hibbs wrote: >> >>> This coming Monday there is going to be a meeting in my company on >>> >> >> Well.. How did it go? >> > > The meeting was moved to today, I think.Ok, how about now? --Steve
On 10/26/05, Stephen Waits <steve-g8GSkY9QmIteoWH0uzbU5w@public.gmane.org> wrote:> > > On Oct 25, 2005, at 11:31 AM, Joe Van Dyk wrote: > > > On 10/25/05, Stephen Waits <steve-g8GSkY9QmIteoWH0uzbU5w@public.gmane.org> wrote: > > > >> Curt Hibbs wrote: > >> > >>> This coming Monday there is going to be a meeting in my company on > >>> > >> > >> Well.. How did it go? > >> > > > > The meeting was moved to today, I think. > > Ok, how about now?As I pretty much expected, they were not convinced that there was a compelling reason to change the status quo. Ruby (and Rails) will currently not be a standard, supported component of the server side environment at Boeing. Exceptions will be granted on request if the requestor is willing to fund it. There already is one such exception, and I expect a seocnd to be granted next week. The strategy now is to continue to build internal demand and pressure until it can no longer be ignored (plus I''m trying to get myself into positions where I''ll have more influence) -- this will take time. Curt _______________________________________________ Rails mailing list Rails-1W37MKcQCpIf0INCOvqR/iCwEArCW2h5@public.gmane.org http://lists.rubyonrails.org/mailman/listinfo/rails
On 10/26/05, Curt Hibbs <curt.hibbs-Re5JQEeQqe8AvxtiuMwx3w@public.gmane.org> wrote:> On 10/26/05, Stephen Waits <steve-g8GSkY9QmIteoWH0uzbU5w@public.gmane.org> wrote: > > > > On Oct 25, 2005, at 11:31 AM, Joe Van Dyk wrote: > > > > > On 10/25/05, Stephen Waits <steve-g8GSkY9QmIteoWH0uzbU5w@public.gmane.org> wrote: > > > > > >> Curt Hibbs wrote: > > >> > > >>> This coming Monday there is going to be a meeting in my company on > > >>> > > >> > > >> Well.. How did it go? > > >> > > > > > > The meeting was moved to today, I think. > > > > Ok, how about now? > > As I pretty much expected, they were not convinced that there was a > compelling reason to change the status quo. Ruby (and Rails) will currently > not be a standard, supported component of the server side environment at > Boeing. Exceptions will be granted on request if the requestor is willing to > fund it. There already is one such exception, and I expect a seocnd to be > granted next week. > > The strategy now is to continue to build internal demand and pressure until > it can no longer be ignored (plus I''m trying to get myself into positions > where I''ll have more influence) -- this will take time. > > Curt > > > > _______________________________________________ > Rails mailing list > Rails-1W37MKcQCpIf0INCOvqR/iCwEArCW2h5@public.gmane.org > http://lists.rubyonrails.org/mailman/listinfo/rails > > >I realize it''s a late response, but I''d include the screen shot of the two pragmatic books at the top of the Amazon Computer best-seller list. http://blogs.pragprog.com/cgi-bin/pragdave.cgi/Tech/Ruby/TopTen.html Nick
On 10/26/05, Curt Hibbs <curt.hibbs-Re5JQEeQqe8AvxtiuMwx3w@public.gmane.org> wrote:> On 10/26/05, Stephen Waits <steve-g8GSkY9QmIteoWH0uzbU5w@public.gmane.org> wrote: > > > > On Oct 25, 2005, at 11:31 AM, Joe Van Dyk wrote: > > > > > On 10/25/05, Stephen Waits <steve-g8GSkY9QmIteoWH0uzbU5w@public.gmane.org> wrote: > > > > > >> Curt Hibbs wrote: > > >> > > >>> This coming Monday there is going to be a meeting in my company on > > >>> > > >> > > >> Well.. How did it go? > > >> > > > > > > The meeting was moved to today, I think. > > > > Ok, how about now? > > As I pretty much expected, they were not convinced that there was a > compelling reason to change the status quo. Ruby (and Rails) will currently > not be a standard, supported component of the server side environment at > Boeing. Exceptions will be granted on request if the requestor is willing to > fund it. There already is one such exception, and I expect a seocnd to be > granted next week. > > The strategy now is to continue to build internal demand and pressure until > it can no longer be ignored (plus I''m trying to get myself into positions > where I''ll have more influence) -- this will take time.*shakes fist at silly managers*