It's interesting. Django seems to be a more CMS oriented solution - similar to Drupal/Civicspace - not surprising given it's news oriented roots. That's not going to suit everyone though - me for starters. If I had to use Django I'd be pulling out the admin stuff from the start as it doesn't really fit in with our requirements. There's soon going to be a huge explosion of web application diversity and not all models are going to be based on CMS paradigm.

I personally think Kevin has made some good choices with TurboGears that deliver a lot of flexibility and power. CherryPy's a great framework, Kid's a very cool templating system that does The Right Thing with regards to not encouraging code and logic integration into the template. It's XML oriented approach is also going to become more important as time goes on. SQLObject is a great ease of entry ORM.

There seems to be quite a lot of grizzling about Kevin's choices too which, quite frankly, aren't helping matters much. If Kevin and Co. had been less clear minded about TurboGears it could have actually destroyed the whole effort. If the choice of components seems disagreeable the why not build your own? Fragmenting the TurboGears effort isn't going to help.

One of the main reasons that I think RoR has been so successful is because it's a very cohesive and consistent bunch of components that work really well together. That and endless fervent evangelism. :-)

TurboGears seems to be heading the way of supporting lot's of different components. Is this going to fragment the effort? Would it have been better to stick with Kevin's choice of best of breed components and make them better still?


http://www.oreillynet.com/pub/wlg/9168?wlg=yes

I am *not* posting this as a call to arms, but rather as means of gathering end-user input.

My comments:

1,2: With full WSGI multi-app support from the bottom up we will retain praised simplicity while adding even more powerful application composition.

3: It would be nice to support Django templates now they decoupled them somewhat.
We really need to improve error reporting for Kid.

4: WSGI + Ian's excellent middleware

5: Some work into alternative mapping schemes would be good.
How are we on Routes integration?

6: we are getting there with ToolBox



Cheers,
Simon



Reply via email to