[cheesegrits]

Originally Posted by BlackSun

I still work on the feature. I get the infos from wikipedia and strip the HTML but at time I work on a better solution - with a additional text file or database with the keywords - that wont cause so much load for wikipedia.

While I applaud your efforts to be a good net.citizen by caching wiki results, I'm just wondering what method you will use to make sure the cached results don't become outdated? Obviously one of the fundamental principles of wikipedia is that it is constantly changing, with anyone being able to edit the entries.

Maybe a cron job which runs thru your table once a day and refreshes the data?

On the flat file vs database, personally I would vastly prefer a database table.

If you need anyone to help with development or testing, I'd be more than happy to help out in any way I can.

-- hugh