I've been working with Turbogears2 and following it's development for about 2 years now, and I'll go ahead and lay it out for ya.
The Python vs. Ruby debate is useless, it's going to come down to personal preference more than anything. I prefer Python, but you could go ahead and make alot of the same arguments for Ruby on Rails as for Django (written in Python). Documentation for Turbogears/Pylons has gotten much better over the past year, but it's still not even half way to where it needs to be for a mainstream audience to embrace it. Furthermore when taking into account 3rd party documentation, aka how-to's and video walk-thru's, that's going to be a chicken and the egg scenario...it'll only get better if more people start using TG, but the lack of those things hold back the rate at which that will happen. Personally, I don't see this getting that much better anytime soon due to the small size of the community, so deciding to use TG/Pylons implies you are willing to look elsewhere for documentation/help and capable of figuring things out on your own. I've spent much more time on other sites than I have on here when looking for answers. Next up, the modular design of TG is a double edged sword. On the one hand it's awesome, you have choice and it's just a better development model in general. However, on the other hand it means if you choose to use a component that is not the default, you're going to have an even harder time finding documentation, and once again will be having to figure out alot of the differences all on your own. Another problem is that if one of the core components gets changed between releases, you may have to make some major modifications to your own code to move forward. For example, a few months ago there was a big scare that Genshi was going to be obsoleted and need to be replaced, so if you have almost a hundred pages written for Genshi (like the project I've been working on) that means a major rewrite when you update. And as TG relies on about 30 some odd libraries, any one of them could go at any point in the future. Now I don't mean to sound like I'm hating on Turbogears, it's a great framework, there's just a bit that could be better. By way of the design choices and libraries from which TG is built upon, I'd wager you wouldn't see a service like Twitter "Fail Whaling" quite as often if they had written it with TG/Pylons vs Ruby on Rails as it's currently written in. Just know that it's never a clear cut answer of which is better, it all depends on you (and your team's) development styles/preferences and what type of app you are trying to write. To be quite honest about the whole thing though, I don't realistically see Turbogears dethroning ROR (or Django for that matter) anytime soon, not for a matter of form, function nor marketing, but just simply inertia. Human nature dictates people will continue using what they already know unless there is a huge benefit to changing, and for the majority of web developers ROR is the easier, more well known solution. Nothing to get emotional about, it's just how things are. - Derick On Jul 28, 11:59 pm, waugust <[email protected]> wrote: > The advantages that you cite for Rails are exactly what frustrates me > here. > Rails has a humongous, fanatical, user base. So many podcasts, > screencasts, books, blogs, sites, etc.. etc... > Its marketed as well as Budweiser Beer, for sure. Spuds McKenzie never > made that crap taste any better though. > I agree, 80-20, Rails is fast. Its as fast and versitle (though > arguably not very scalable) due to its fanatic user base. There's gems > for everything out there and generators as well. > Now, Pylons and TurboGears, lack in documentation (it seems you really > have to get a book, unfortunetly), and you may find yourself creating > some things that you were accustomed to in Rails. I haven't yet found > a good source of Paste scripts out there (to rival generators in > Rails), though its pretty easy to write your own. The Python Package > Index (PyPI) does have some things on there though no where near the > amount of gems. > Though the biggest item to argue in favor of the Python platform is > Python. Everything is indeed way decoupled. You can have any ORM you > want (Rails 3 only now offers a different ORM) or any templating > language. You will find that you can also look under the hood and > McGyver things allot easier then you could with Rails. There is no DSL > here, everything is pure sweet, simple Python. Its like trying to fix > a Honda (Rails) vs. a VW (Python platform framework). > One thing that I'll miss is Cucumber... Saw somewhere that it isn't > really feasible with Python due to the lack of real "introspection" in > the language. Though it seems like, the somewhat recently introduced, > decorators would work. > I think you will like the web debugger (I know, small things).. it has > an ajax shell allowing you to run code right there, plus you can > automatically search the exception on the web or submit it to the > forum... pretty neat. > > On Jul 28, 9:21 pm, Guyren G Howe <[email protected]> wrote: > > > > > On Jul 27, 2010, at 12:33 , waugust wrote: > > > > So, I divert the topic from technical questions to this rant to ask > > > you: Why and how is it possible for the industry to allow us to > > > degrade the progress of technology in favor of good marketing? > > > As a long-time Rails user and new TurboGears user, let me make some > > comments in Rails' favor, and to compare and contrast what I've seen of > > TurboGears so far. > > > First: it is much easier to get into Rails because there is an order of > > magnitude (maybe two) of documentation, much of it of very high standard. > > The documentation that there is for TurboGears is of quite good standard, > > but there isn't enough of it, and much of it is out of date or incomplete. > > > Second: you are selling Rails rather short. I can't say yet whether > > TurboGears is better, but Rails is very good. And Rails v3 in particular is > > a very solid and mature product. I can get shit done in Rails 3 very > > quickly, and what I write is concise, readable and easy to maintain. > > > Ruby and Python are both, I think, equally good languages for different > > reasons. Most developers I know who have spent time with both find both > > quite comfortable and effective languages. This is certainly my opinion. > > > From what I've seen, TurboGears is a much more loosely coupled design than > > is Rails (although Rails 3 is now similarly loose), which I like. And I > > like the batteries that are included, particularly ToscaWidgets. > > > As I say, I'm not familiar enough with TurboGears to offer an opinion about > > which is better. But I think that even if TurboGears is better, it cannot > > be by as much as you suggest. Rails really is very good. -- You received this message because you are subscribed to the Google Groups "TurboGears" group. To post to this group, send email to [email protected]. To unsubscribe from this group, send email to [email protected]. For more options, visit this group at http://groups.google.com/group/turbogears?hl=en.
