On Sat, Jan 31, 2004 at 05:14:12 -0500, Dave Phillips wrote:
IIRC, we also discerned that one of the first tasks
would be the
establishment of a backend database from the existing links
and categories (which I would like to see redefined too). The URLs would
need culled and organized: I don't think this would
be a terribly difficult first step, maybe a good Perl script can do it ?
Is PW not a Perlmeister ?? ;)
RDF! (rdf, rdf, rdf, ...)
http://www.w3.org/RDF/
Full disclosure: You may want to take this with a pinch of salt due to
favourite hammer syndrome; I currently justify my existence by developing
a large-scale (GPL'd of course) RDF server.
Good points:
* this is exactly the kind of thing RDF is designed for
* nice query syntax (like SQL but better)
* very fashionable
* interoperability (metadata can be used, and partly understood, by other
tools because RDF has inherent semantics)
Bad points:
* RDF/XML syntax is horrific (but you hardly ever deal with it)
* not famialiar to people who know LAMP systems
* a bit more resource intensive than PHP+MySQL
* RDF tools not as mature as relational databases
You wouldn't want to use RDF for things like user comments and reviews,
but its ideal for metadata-y things.
I could help by providing things like web form -> RDF/XML CGI scripts and
RDF knowledge-base hosting for betas.
- Steve