Pedro Graca wrote:
> Or use a database.

I was trying to avoid using a database for now, as I know that if I went
ahead with that, it would end up being so huge in scope that I'd never
finish it :)

> Of course these options assume the 'index' file and the database would
> be updated for every new file ... (and deletions and changes).

I have been considering the index file approach. However, considering
the speed problem is mainly a problem I don't want my visitors to
encounter making the slow "grab the info from all the files" version
could be done on my harddrive and the output of the script I could just
copy and paste into an index page. That way the slow stuff will be done
behind the scenes.

Maybe that's the best option. Hadn't considered that until just now.

Kim Jensen