Social Bookmarking (Web 2.0) using old Web technology

I have an old Linux server that sits, sad and muted, unused in a corner. I decided to blow the dust off of it, and see what I can wrangle out of it.

For the morbidly curious: it’s a Fedora Core 1 install, running on a Dell Poweredge server. It has a crusty old hard drive installed in it, that contains some data (nothing very useful any more, really) that dates back to 1997. It has several installs of Python — ranging from 2.3 to Stackless 2.5. At various points, it’s had Plone, Zope, and even some (gasp) Perl scripts running on it. It itself was originally a RedHat 5.0 install.

So, in deciding what to do with the machine, and the software installed upon it, I re-acquainted myself with one of the little software gems installed — a lovely Web application server called SkunkWeb. It seems to have died a death of neglect, more so than lack of functionality. It’s ability to cache various parts of the rendered web page, along with caching the application’s data are very powerful. And, it’s almost all written in lovely Python, with only a small handful of parts that in C (for example, the hashing code for the cache engine is in C for ultra-performance concerns.)

It didn’t take long to remember how to work with it — I spent a little over a year working with it when it was a closed-source project (I worked for the company that originally developed it). After trying to figure out a project that all the little lovely caching bits of the system would work well for, I settled in on trying to make one of those fancy Web 2.0 social bookmarking sites that everyone’s always raving about.

Since Reddit’s code was already open-sourced, I opted to nick their look & feel. Unlike some of the others that used the vote up/down concept (Vinefire…I’m looking at you), I also implemeted a large amount of the functionalilty. Not just the posting of links to your site, but also the ability to have discussions, and even have group moderation of those discussions.

I have no illusions of success based on the site, but I’m curious how much of beating the server & software can take before their age becomes apparent. The machine is from 2002, the version of SkunkWeb I’m using is from 2004, and the database backend is an older Postgres (since I can no longer use Yum to update and I’m afraid of breaking things if I try to build from source).

Along the way, I also learned alot about AJAX, which was a topic I had been hesitant to learn. That experience is for another blog entry, however. Get thee to Share2Rate and behold the glory of 5+ year old software!

Share and Enjoy:
  • Digg
  • del.icio.us
  • Facebook
  • NewsVine
  • Reddit
  • StumbleUpon
  • YahooMyWeb
  • Google Bookmarks
  • Yahoo! Buzz
  • TwitThis
  • Live
  • LinkedIn
  • Pownce
  • MySpace

Leave a Reply

Security Code: