[VM] small thoughts on "Basic Principles for Managing an RDF Vocabulary"

[trying to show some progress here.  Are we having a telecon today?
In 20 mins?  Ref. [1][VM] Telecon - skip Aug 2, hold Aug 16?
[1] http://lists.w3.org/Archives/Public/public-swbp-wg/2005Jul/0069.html ]

Re:

   "Basic Principles for Managing an RDF Vocabulary"
   http://www.w3.org/2001/sw/BestPractices/VM/principles/20050705

   [[
   3. Articulate your Maintenance Policies

   ... the developers/maintainers of an RDF vocabulary should publish a
   maintenance policy for that vocabulary.  The maintenance policy should
   articulate whether or not change is allowed, and the way that change is
   managed.
   ]]

My intuition is that there is a substantitive difference between changes
that only add new terms to a vocabulary and changes that affect the
interpretation of previously-defined terms.  Does this intuition hold
in practice?  (I.e. are there important tools that expect that they 'know'
all the terms in a given namespace without dynamically querying
that namespace?)

If this intuition does hold in practice, then I think it's worth talking about
in this "Basic Principles" document.

   [[
   5. Publish a Formal Schema

   ... the header field 'accept=application/rdf+xml' 

(minor typo: that should be 'Accept: application/rdf+xml')

   against that URI should return an RDF/XML serialisation of an
   RDF graph that includes a description of the denoted resource.

It seems to me that this is an excellent place to document good
practice for using [2]rdfs:isDefinedBy.

   [2] http://www.w3.org/2000/01/rdf-schema#isDefinedBy
   [3] http://www.w3.org/TR/rdf-schema/#ch_isdefinedby

We might document that URI string manipulation to attempt to
heuristically locate a namespace is _not_ good practice; the
proper method to determine the namespace given a Property
or Class name is to do the GET operation and expect to find
an (?s rdfs:isDefinedBy ?n) statement with ?s being the
Property or Class of interest and ?n giving the namespace URI.

Received on Tuesday, 16 August 2005 12:48:38 UTC