And now I've finished leading the RDFa + schema.org codelab that I've been stressing over and refining for about a month at the American Library Association annual conference Practical linked data with open source preconference. Long story short, most people got about as far as I expected (part-way through the first exercise), but they all got through the initial hurdles and learned enough to keep learning on their own. My hopes are that this leads to:
the implementation of structured or even linked data in existing systems, for those that at least have systems that give them the ability to edit their HTML templates
the addition of linked data to library web pages the next time they get refreshed or redesigned (it happens pretty often!)
some patterns of implementation, so that we hopefully arrive at a relatively standard way of marking up the same metadata (given the many alternatives that we have just within schema.org for something like a publisher)
when tweaking templates for display or design purposes, to avoid mangling existing structured data that a system like Evergreen, Koha, or VuFind publishes by default
more awesomeness in the world of library metadata!
Oh, and for posterity, I temporarily marked up this page to link to our pizza order form as a really lame short URL service, and as I did that impishly polluted the schema.org vocabulary with the new type PizzaOrderPreferences. I don't think that's going to make it into the official vocab though! The code was:
<p vocab="http://schema.org/" typeof="PizzaOrderPreferences">
And <a href="http://doodle.com/exampleblahblah" property="url">order pizza here</a>.
<p>If our pizza order doesn't get gamed, that just shows how few people visit my blog!
Thanks for this Dan. I've always had an interest in linked data but frankly didn't know where to start. It's great to have someone who knows what they're talking about say, "you should dive in here..." (RDFa in this case).
I'm working my way through the codelab and so far it's revealed the complexity of this stuff -- in terms of scope etc -- which I never really appreciated before.