September 09, 2005
Johannes Ernst is launching a one-man campaign to simplify FOAF. Except actually simplification in this case boils down to "not in RDF".
I keep seeing these RDF vs XML tirades, and I have to say its a bit puzzling. I've never found RDF to be very difficult to work with - in some ways I actually find it rather simpler than XML as a lot of the legwork of interpretation is uneccessary, although the separation of structure and meaning in RDF is a tricky one if you're used to tag scraping. The trick is to start off using a triplestore from the outset - its pretty easy really, or to just choose to create RDF instance with a predictable rigid structure to make life easier for those who want to use a regular XML parsing or simple tag-scraping approach.
Anyway, technology fundamentalism aside, I don't think its a question of really needing to simplify FOAF - its already simple enough to be used - its a question of really using FOAF. Even autogenerating basic FOAF (a la LiveJournal) and linking to it from feeds would take us a very long way.
I already link to my FOAF from my RSS 1.0 feed (although this makes it invalid according to FeedValidator - aaargh!) and my Atom feed using the author.uri element. If this became common practice all kinds of fun could ensue, without the need for more or different specifications being developed.