- From: Patrick Stickler <patrick.stickler@nokia.com>
- Date: Fri, 18 Feb 2005 10:32:55 +0200
- To: "ext Harry Halpin" <hhalpin@ibiblio.org>
- Cc: Norman Walsh <Norman.Walsh@sun.com>, www-tag@w3.org
On Feb 17, 2005, at 18:11, ext Harry Halpin wrote: > > However, it does seem like a namespace document would be a sensible > place to put such a schema, although - of course - it's not required. > One could put the schema anywhere - however, then the question is how > to find it. > DocBook documents could be defined by their schema, and one could > find the schema in the namespace document > (http://www.oasis-open.org/docbook/...) or somewhere else entirely, > like your hard drive. I think the reason people are talking at > cross-hairs is > that they think the pro-"namespace document" camp are saying that it > would be useful to upgrade RDDL or something like it to help with > versioning issues, and the anti-"namespace document" camp thinks that > we want to make having a namespace document a necessary condition of > the namespace URI - which as the spec says and use shows, is obviously > not true. However, as a *optional* place to put things like versioning > info, a namespace document seems just as good as a URI. We need a consistent, standardized, and (potentially) globally ubiquitous solution which provides effective discovery of information about the models needed to properly interpret data instances. Namespace documents fail to provide such a solution. Adding versioning support (however that might be done) will not enable namespace documents to be less ambiguous or more ubiquitous. Patrick > > > -harry > > > On Thu, 17 Feb 2005, Norman Walsh wrote: > >> / Harry Halpin <hhalpin@ibiblio.org> was heard to say: >> | While I agree that Henry is technically correct (technically as in >> | "read the specification"), this giant perma-thread clearly shows >> | that there are simply problems in keeping track of versioning with >> | namespaces. If a namespace can have an infinity of names, then >> | perhaps this should be repaired in a new spec, since having an >> | infinity of names makes versioning difficult. Either one has a >> | separate URI for each version, or a representation is returned by >> | the URI that contains the versioning information. >> >> In the general case, I simply do not believe that there is any >> relationship between the namespace name and the set of terms in the >> namespace. >> >> Consider the case of DocBook. DocBook V5.0 will be in a namespace. I >> do not expect that namespace to change. Ever. It will be the namespace >> for V5.0, V5.1, V6.0, ... V17.3, etc. ad nauseum, of DocBook. >> >> If the document you have in hand validates against the DocBook V5.3 >> schema, it is a DocBook V5.3 document. If it validates (instead of, or >> also) against the V5.0 schema, it is a DocBook V5.0 document. Turning >> that around, as a consequence of the versioning policy of the DocBook >> Technical Committee, I can predict that every V5.0 document will also >> be V5.3 document. >> >> This is not the only possible namespace/versioning strategy, but for a >> bunch of practical reasons, it is the best policy for DocBook and I'd >> resist any attempt to define a general policy for namespace/versioning >> that prevented the DocBook policy. >> >> Be seeing you, >> norm >> >> > > -- > --harry > > Harry Halpin > Informatics, University of Edinburgh > http://www.ibiblio.org/hhalpin >
Received on Friday, 18 February 2005 08:33:47 UTC