- From: Dave Beckett <dave.beckett@bristol.ac.uk>
- Date: Mon, 12 Sep 2005 14:47:44 +0100
- To: Bjoern Hoehrmann <derhoermi@gmx.net>
- Cc: Mark Birbeck <mark.birbeck@x-port.net>, public-rdf-in-xhtml-tf@w3.org, dean@w3.org
On Mon, 2005-09-12 at 09:31 +0200, Bjoern Hoehrmann wrote: > * Dave Beckett wrote: > >It's an amusing idea for this to obsolete RDF/XML, but incomplete and > >non-optimal compared to RDF/XML, for writing some RDF triples (XML > >literals for example) and a solution for XHTML2 only. > > It's disappointing that you describe ideas of this task force, which > has an important charter to fulfill, as "amusing". As you can see from my quote, I was amused at it replacing RDF/XML, not the work of the task force. The charter of the task force as I understood it was to better embed RDF in (X)HTML, not to obsolete RDF/XML as a recommended RDF syntax. The results so far seems like it will succeed in allowing much better semantic descriptions in HTML markup but that's not the same thing as writing a new, complete and recommended RDF transfer syntax to replace the existing one. It's an additional syntax (or syntaxes) not a replacement. Replacement syntaxes would likely have many additional requirements and considerations and the charter for that work would have to be very carefully written. I think I've personally experimented with around 3-4 alternate syntaxes so far with differenet approaches and a couple of white papers on it with no final conclusions yet. > >People are very recently creating new formats that are difficult or > >impossible to validate with XML schemas or DTDs. For example, Atom 1.0 > >which only has a RELAXNG schema. Just like the RDF/XML REC has > >(informative) and for very similar reasons - flexibility in inclusion of > >XML and allowing validation specified by namespaces; which is > >trickier/impossible in WXS. > > If you mean that making formats that are difficult to validate using > DTDs and XML Schema might be acceptable, I must urge you to read e.g. > <http://infomesh.net/2002/rdfinhtml/#embedNoValidate> which you will > find convincing with respect to why it is impossible for XHTML 2.0 to > allow for use of RDF/XML syntax. You'll find that informatively cited in the RDF/XML REC from http://www.w3.org/TR/2004/REC-rdf-syntax-grammar-20040210/#ref-rdf-in-xhtml very close to the informative relaxng schema. I just brought up that point for information, and note that new document formats are appearing which have similar design spaces to RDF/XML in terms of extensibility by namespaces not known in advance and schema languages used to validate them. Those were informed choices. Embedding Atom in XHTML would be similarly hard. Dave
Received on Monday, 12 September 2005 13:47:54 UTC