- From: Martin McEvoy <martin@weborganics.co.uk>
- Date: Fri, 16 Jul 2010 14:56:37 +0100
- To: RDFa Working Group WG <public-rdfa-wg@w3.org>
- CC: RDFa Working Group Issue Tracker <sysbot+tracker@w3.org>
Just some feedback that may be helpful when considering this Issue. On 16/07/2010 13:22, RDFa Working Group Issue Tracker wrote: > ISSUE-36 (Default vocab specification): Should Profile documents allow the specification of a default vocabulary? [RDFa 1.1 Core] > > http://www.w3.org/2010/02/rdfa/track/issues/36 > > Raised by: Manu Sporny > On product: RDFa 1.1 Core > > Markus proposed that Profile documents should allow the creation of a default vocabulary: > > http://lists.w3.org/Archives/Public/public-rdfa-wg/2010May/0041.html > > This has a number of implications: > > 1. How does one declare a default vocabulary via the profile document? link rel="profile" X/HTML5+ documents. head @profile for XHTML 1.X and HTML4.X documents. > 2. How does the declaration affect the @vocab attribute, would it override it and if so, in what order? and @prefix It shouldn't @profile should be of least importance in the food chain, an author may not have control of the document of @profile, where he does over the source document that contains @prefix and @vocab. > 3. How does the declaration affect the CURIEs like ":next"? not so sure about that one ;) my preference is RDFa1.1 should grandfather @rel values so none at all :P > The biggest question is the value of this feature? What is the use case that we are attempting to support? Does this overly-complicate RDFa without much payback? > From experience I have to run two parsers one to produce a list of prefix mappings (whether the profile has been cached or not) these prefix mapings (all of them) then have to be injected into a second instance in order to parse the containing RDFa of the referring page. I have found downloading a list of prefix mappings from http://prefix.cc/popular/all.file.txt and processing that faster at resolving prefixes than processing @profile. In short I'd rather not :) Best wishes -- Martin McEvoy
Received on Friday, 16 July 2010 13:57:29 UTC