I’m considering giving Django another go, mostly because I’m getting somewhat burnt out by building Rails-based sites, but also because I’m interested in seeing how Django handles the few areas where Rails still sucks. These are:

• Localization: I’m very happy about the Unicode stuff that was added to Rails (or rather, ActiveSupport) a while ago, but once you actually need to translate stuff, you’re still forced to pick one of the many localization plugins, most of which are either horrible or just plain broken. The one I usually end up with is Ruby-GetText, which, well, works. Just don’t look at its source code. Ungh.

• Admin UI generation: granted, this is something you don’t really need in most applications, but what Django does here is really pretty f’ing sweet. Just not having to bother with building code for making your application data editable by admins may make Django worth using in a lot of applications. Rails, of course, has Scaffolding, which a lot of people confuse with an admin system generator and thus compare it (not favourably) with Django’s stuff, but that’s another rant really.

• Performance: well, this is a touchy subject. Usually, when discussing the performance and scalability of Rails applications, my stance is that 99% of scaling problems are due to developers being stupid (for example, by not using database indices or caching), and not the framework putting insurmountable hurdles in their way. However, Ruby is slow, and yes, much slower than Python. I’d like to know how well Django performs compared to Rails, and how it tackles typical scalability issues.

However, I’m still expecting Django to piss me off at some point. No migrations? No Capistrano? No HAML/SASS? No ActiveRecord (yes, I’m a fan — the new named_scopes rock)? No Ruby? Ungh! A very high price to pay for using another framework, no matter how more grown-up it feels.

Any comments? Should I give it a go?