Better Web Application Framework

For a web application that I am developing as part of a research project I have decided on using [Python]( for the business logic. The reasons being that the application will use fixed-point binary arithmetic, units, and data output as line-graphs. Python seems to have the libraries that I’ll need for this and is better suited for rapid development than Java.

Part of the requirements that I have been researching this last week have been concerned with the identification of a suitable web application framework for Python. I started out this week assuming that this would be [Django]( and even bought the [book](! However, because this is research, I have the luxury of time so I’ve done some additional research and decided that the alternative frameworks [TurboGears]( and [Pylons]( were also worth a look. Searching Google for getting started screencasts and videos I discovered a [wonderful video]( of a web application framework comparison presentation by Sean Kelly. In this presentation, Sean uses J2EE (Servlets and JSPs), Rails, DJango, TurboGears, Zope/Plone and even Enterprise Java Beans (JBoss) to develop a simple time-tracker application and reports his experiences. Rails, Django and TurboGears come out (roughly in that order), but the big surprise is that [Zope]([Plone]( comes out top! So, maybe I need to rethink!

I’ve embedded the video here as it’s not only a useful and timely comparison of web application frameworks, it’s also an example of just how useful and inspirational a simple presentation can be!

Powered by [ScribeFire](

10 Replies to “Better Web Application Framework”

    1. Paulo, for very simple use-cases, for example aggregating some RSS feeds, doing some filtering, removing duplicates, pipes is very simple. For more complex manipulations, I think you do need more expertise and some programming experience. I’m very much a beginner!

      As I mentioned in my post, Tony Hirst (@psychemedia on twitter) is my mentor in this!

    1. Tony

      One of the problems with the simple pipe I came up with was the OPML parser block itself. Although Stephen Downes’ OMPL is available as a URL, and therefore in a sense always up to date, the parsing into feeds seems to take too long and the pipe often times out before it can generate results. I need access to the raw data that is being used to create the PLENK2010 daily I think!

      But thanks for the tips, I’ll follow them up.

  1. Your screencast was awesome, Chris. I created a dashboard in NetVibes, added it to my Symbaloo homepage and now I’m in the RSS feed business. Thanks so much!

    1. You are very welcome! I have used NetVibes in the past but always went back to Google Reader. Using it to create a dashboard for an on-line course seems a very good idea though and I’ll be experimenting further.

  2. Very interesting! I want try to use an aggregator. But to complicated right now. I just added the link to The Daily as a bookmark on my igoogle, to keep the course separated from the rest in the RSS-reader.

  3. This is great work. It is just what I was looking for. I was about to ask you for this kind of help in the plenk moode but got your blog info instead. Much easier to connect here.

    Thanks much,

Comments are closed.