I have been quite open about my struggles with Linked Data but I have recently decided to focus on the bit of it closer to the web, the RDFa/XHTML stuff that has similarities to the microformat stuff I embraced a few years ago. Pete recently wrote an extremely helpful post over at eFoundations that got me thinking that maybe this element of LD was a way in as god knows I haven’t been doing too well at grasping the big data set side of things.
The fact that Google (with Rich snippets) and Yahoo (with SearchMonkey) are supporting these elements with the possibility of enhanced search results (mainly in their custom search offerings I think though occasionally in ‘proper’ search results) starts to give me something solid to hold on to and the fact I have some familiarity with microformats seems to help as well.
Anyway at the same time as I’ve been reading up on this I have also been doing background reading for my part of the UKOER Call that is coming up and which I wrote about earlier. Anyway I had a bit of a brainwave (that has probably been obvious to everyone else forever!) that using RDFa to markup OERs in a way that Google or Yahoo MIGHT support in the future could allow (a) some pretty enhanced search tools to be built with minimal costs and (b) it would allow collections and aggregators to be more ‘intelligent’ than tools like RSS and Atom allow alone.
I followed this idea down the rabbit hole for a while and am pretty sure I discovered that there is currently no agreed or standard RDFa vocabulary or profile (is that the right terminology?) but that there has been some work in this area and in particular the Creative Commons has developed the ‘beta’ OER search engine called DiscoverEd which encourages the use of RDFa so that, in a very Rich Snippets kind of way, they can display search results that show license information, subject, educational level and author. At the moment this doesn’t spider the open web but relies on collections of OERs being submitted (by adding the RSS feed of the collection) but nonetheless it has the potential to be a really interesting project. That said it has been ‘live’ for 18 months and I couldn’t find much information about how it was going but it is something I’d like to learn more about.
There isn’t the kind of critical mass of content marked up in this way yet for the collection/aggregation element of my idea to work but I really do think RDFa might be a part of the answer to making the discovery of OERs easier and make creating collections more dynamic.
Anyway I’m going to be digging in to this space a bit more in the coming weeks and months; with JISC starting to dip its toe into Linked Data with the Expo call and JorumOpen starting to be a pretty impressive repository of OERs there could be some real opportunities here (then again I might be entirely misreading the situation!). If nothing else I want to learn more about DiscoverEd.