- From: Christoph LANGE <ch.lange@jacobs-university.de>
- Date: Tue, 1 Dec 2009 11:46:13 +0100
- To: Ivan Herman <ivan@w3.org>
- Cc: Toby Inkster <tai@g5n.co.uk>, RDFa Developers <public-rdf-in-xhtml-tf@w3.org>
- Message-Id: <200912011146.14142.ch.lange@jacobs-university.de>
Dear Toby, Ivan, all, thanks for your comments! Indeed I agree with Ivan that it makes sense to stick to @about, as normal RDFa processors currently support it that way, and one of our main points in using RDFa at all was the wide support for it. While we will also publish XHTML+RDFa documents _generated_ from our OMDoc documents on the web (will be online in a few days from now), which contain even more RDFa extracted from our markup than those metadata for which we use RDFa in the _input_ syntax (i.e. OMDoc), the OMDoc sources will also be online. Then, we'd risk spreading inconsistencies, if crawlers parsed both the RDFa in OMDoc, not knowing about implicit @about's, and the generated XHTML+RDFa with spec-conforming @about's. Therefore, we are indeed thinking about having editors and other tools support the user in creating those -- seemingly redundant -- @about's. 2009-12-01 10:53 Ivan Herman <ivan@w3.org>: > At the moment, the rdfa distiller has a flag on whether it is XML or > HTML, essentially taking care of things like xml:base and possibly > existing RDF/XML portions. I would hate to have to have all kinds of > different options for different XML applications... My current favorite position on that is: Use @about for now, but hope for some declarative mechanism that allows us to specify custom parsing rules for RDFa in different host languages (for which e.g. @profile has been suggested in the other thread). BTW, another related question: When generating XHTML+RDFa from a formal representation that is already RDF-compatible in itself (e.g. RDF/XML, e.g. OWL, e.g. OMDoc, …), I suppose that both the original representation and the XHTML+RDFa should use the same URIs for the same things. Suppose the original document, e.g. having the URI doc.omdoc, contains a triple with subject #resource, and suppose we generate doc.xhtml with RDFa from it. A naïve translation might create something like <div about="#resource">, i.e. talk about a resource doc.xhtml#resource. But I guess it should rather be <div about="doc.omdoc#resource">, as the formal concept will stay the same, regardless of its presentation. Are there similar experiences from, e.g., generating XHTML+RDFa from OWL? Cheers, Christoph -- Christoph Lange, Jacobs Univ. Bremen, http://kwarc.info/clange, Skype duke4701
Received on Tuesday, 1 December 2009 10:46:31 UTC