- From: Christoph LANGE <ch.lange@jacobs-university.de>
- Date: Tue, 1 Dec 2009 12:02:32 +0100
- To: Mark Birbeck <mark.birbeck@webbackplane.com>
- Cc: Ivan Herman <ivan@w3.org>, RDFa Developers <public-rdf-in-xhtml-tf@w3.org>
- Message-Id: <200912011202.32644.ch.lange@jacobs-university.de>
Hi Mark, 2009-12-01 11:31 Mark Birbeck <mark.birbeck@webbackplane.com>: > But Christoph seems to be saying that his language needs to make lots > of statements about the document itself, resulting in a common pattern > of: > > <x id="y" about="#y"> > ... > </x> > > I don't have any immediate suggestions about how to work around this, > although I'm sure we could come up with something. But I would caution > against using @id in a language, to automatically generate a subject, > since I think it will cause problems for that language in the longer > term. Indeed automatic generation of subjects may be tricky, and that's also why I didn't pursue this any further. The actual OMDoc situation is that certain _elements_ denote formal concepts just by their existence. E.g. <definition><link rel="dc:creator" resource="x:me"/>...</definition> denotes a definition that has some creator. While it is strongly recommended that such elements carry an @xml:id so that we can refer to them by URI, it is not strictly required. Now think of the above XML as "there exists a definition created by x:me", and think of the correspondence of bnodes to existentially quantified variables. Compatibility to other RDFa processors aside, if we had this automatic subject generation, the way of doing it "right" would be: If an @xml:id "i" exists, make #i the subject; otherwise generate a bnode _:someid and make _:someid the subject. Not sure if that is nice, though, it is certainly a bit too far away from RDFa compatibility. Cheers, Christoph -- Christoph Lange, Jacobs Univ. Bremen, http://kwarc.info/clange, Skype duke4701
Received on Tuesday, 1 December 2009 11:08:08 UTC