Being a Python/Django guy, I can’t really recommend one framework over the other, but I can tell you this … go with whichever one has the larger/more active community. You’ll need them. For a business application, you’ll want timely response time to questions and ideally an well populated and active IRC channel. What I find with many of these “alternative” frameworks especially in the early days is they get to the equivalent of version 2 (depending on their versioning scheme) and say “oh, you know we really should make [dramatic change] to [core component]” and you’re left holding the bag. Pick a framework that has already had that epiphany and done that work.

I’ve spent the last year+ doing Django contract work and one of the biggest objections I hear from clients is that it’s prohibitively hard to get Django developers. If your intention is to never have developers leave/get sick/die then perhaps you can build your own center of excellence for an early-adopter framework, but otherwise you’ll need to find them somewhere or retrain internally. Business-wise it’s always better to have a larger gene pool to pick from (look at the popularity of Java … don’t get me started).

So, if is just a learning exercise, pick whatever you like. But if you plan on making production code from it, I would suggest you go with the belle-of-the-ball.

-Sandy

Yehuda Katz here, core developer of the Merb framework. I’d say it depends on whether your developers will be *Ruby* developers or developers using a DSL to make websites. That, of course, depends on the complexity of your application. If your application will require real programming, and not just glue for various pieces of web technologies, Merb is a much better fit.

On the other hand, Rails is perfect for cases where you merely want to use Ruby to simplify some common tasks (templating, database access, routing), etc.

If you have any additional questions, I’d be happy to chat via GTalk (wycats@gmail.com) or IRC (wycats on freenode).