- From: Danny Ayers <danny666@virgilio.it>
- Date: Sun, 18 Nov 2001 12:15:36 +0100
- To: "Ralph R. Swick" <swick@w3.org>, <www-annotation@w3.org>
1,000 thanks. It all falls nicely into place now. >At 12:54 AM 11/18/2001 +0100, Danny Ayers wrote: >>Hi, >>Can someone please tell me why the annotation initiative seems to have >>skirted XLink and opted for RDF to supply link semantics? > >because there are much richer semantics that other applications >of Annotea will eventually use. The extensibility of RDF makes >it far more straightforward to add properties refining your own >semantics than the XLink role model permits. > >I agree with you that were we only interested in human navigation >links then XLink would have been quite sufficient. But our ambitions >go well beyond the tiny bit of work you see in Amaya and Annotea >right now. And even with what we currently represent in Amaya >annotations, we stress what the simpler XLink models express >in their most convenient forms. > >You could certainly think of Annotea and its protocol as an example >of what XLink calls an 'external link database'. > >Early on in our work we did a whiteboard model of how to express >what we wanted for simple annotations in an 'extended XLink with >locator elements'. The model turned out to have similar, and in >certain respects a little more, complexity as the plain RDF model >we finally chose for Annotea. (Note that other people's models >for annotations -- such as the ILRT one -- have refinements over >ours that are even harder to describe in XLink syntax.) Our early >modelling was the inverse of the exercise described by Ron Daniel >in "Harvesting RDF Statements from XLinks" W3C Note 29 September 2000 >http://www.w3.org/TR/2000/NOTE-xlink2rdf-20000929/ >though we did our modelling independently of Ron's work. > >An interesting exercise would be to repeat the inverse modelling >using the structure in Ron Daniel's NOTE in a more careful >manner. I wouldd be very happy to review drafts of such work if >someone picks up this exercise. > >>(I'm afraid the archive search is "Unable to contact target server >>onatopp.w3.org:11000 after 3 tries."). > >We've had major hardware failures on that machine recently >and as a result our system's group has raised the priority of >a complete replacement for our site search facility. > >-Ralph
Received on Sunday, 18 November 2001 06:18:53 UTC