Like many, I’ve been watching the development of the semantic web with interest but also with a degree of scepticism. Interest, because it just makes sense that expressing in a machine-readable way what we humans already know — why one piece of information is linked to another — is a good thing to do (the arguments in favour being so apparent and well articulated elsewhere that I don’t need to state them again here), but also scepticism because most methods of doing this to date are just too darned difficult for the majority of us. To me at least it has seemed that the semantic web and it’s underlying language, RDF, is one spoken only by the initiated few.
Well that was until my 2-day workshop at Talis this week. My brain is now full of graphs and triples with their subjects, properties and values. The trouble is, this little insight into how linked data is helping to shape the semantic web has made it even more frustrating for me. I left the workshop fired up to add machine-friendly meaning to all my information in future, only to get back home, fire up my blog, and realise that if I want to add RDFa, the in-line annotated version of RDF that can be embedded in any web-based document, then I’d have to add it by hand by editing the source code of my post. Ok. Maybe I just haven’t looked hard enough. But no, after a couple of days of searching I can’t find an easy way of adding RDFa to a WordPress post. No matter, this is still, ahem, early days of the semantic web and maybe the user-friendly tools are coming RSN (real soon now). But at least if I accept the pain of adding RDFa by hand, it’ll be worth it, my posts will enter the web of linked data. The only trouble is, I’m not sure if that is true. I can’t find how my RDFa-enabled posts can be used to extract their meaning. Am I simply adding this meaning now for future consumption by a semantically-aware search engine, or have I missed something?
By the way I tried to add valid RDFa inside this post but WordPress kept changing it, I guess out of the box it doesn’t like RDFa, or perhaps I was just doing it wrong.
Maybe I should have written a recipe instead.