Jeni's Musings

To clarify: the overall process begins with some domain-specific XML. We’re looking to add extra elements to that XML. It’s only much further down the line that we will expose at least some of that extra markup as RDFa/XHTML, but the markup needs to be in the original XML for us to expose it in that way.

I saw a paper about ITS somewhere else — I think it must have been Yves Savourel’s presentation at XTech 2007 — and I remember thinking it looked a lot pretty general purpose. Adding attributes/annotation is just one step in automatically marking up text, but the ITS approach does look like a promising one.