[Date Prev]
| [Thread Prev]
| [Thread Next]
| [Date Next]
--
[Date Index]
| [Thread Index]
RE: [xml-dev] My specification writing project has failed :-(
- From: Michael Hopwood <michael@editeur.org>
- To: David Lee <dlee@calldei.com>, Victor Porton <porton@narod.ru>, "LenBullard (Len.Bullard@ses-i.com)" <Len.Bullard@ses-i.com>
- Date: Fri, 29 Jun 2012 16:27:14 +0100
I'm not totally sure I follow this discussion but I thought I would note something that sounds similar (not identical):
Full details at http://www.doi.org/VMF/ - the Vocabulary Mapping Framework is a centrally-managed set of *authorised and agreed* mappings of various namespaces (well, subsets in some cases) into one, highly-generalised hub schema. Each has been done with the cooperation and approval of whomever actually maintains the namespace. It uses RDF.
If you map your namespace (and you need to be sure that it is yours to authorise or it's a waste of time) to the VMF, you can reference the mapping along with the namespace and so in theory other people could reference this standard equivalence (or otherwise) in other transformations.
-----Original Message-----
From: David Lee [mailto:dlee@calldei.com]
Sent: 29 June 2012 16:04
To: Victor Porton; xml-dev@lists.xml.org
Subject: RE: [xml-dev] My specification writing project has failed :-(
Thanks for the explanation.
A problem I see is fundimental. What transformation to do is really a property of the application not the namespace.
Example. Suppose you have MathML inside XHTML. There are many ways this could be transformed, and it is not up to either the MathML or XHTML namespaces (or more accurately the publisher of the URIs) to decide for the app what the transformation should be like.
I think you could use RDF to describe a mapping of namespaces to say XSLT code (or a simplier one to one mapping) but it should reside in the application layer, not be associated globally as a property of the namespaces.
[Date Prev]
| [Thread Prev]
| [Thread Next]
| [Date Next]
--
[Date Index]
| [Thread Index]