Re: IETF XML formats use lots of URNs for namespace names

I agree with the concerns raised by Dan and Henry.  Just to add a bit of 
detail relating to XML Schemas in particular,  RFC 3688 says:

"schema -- XML Schemas [W3C.REC-xmlschema-1] are also identified by a URI 
but their contents are machine parseable.  The IANA registered document 
will be the XML Schema file.  The URN the IANA assigns can be used as the 
URI for the schema and is of the form 'urn:ietf:params:xml:schema:<id>'."

This is at best a misuse of XML Schema terminology and perhaps somewhat 
more deeply flawed as well. 

Background on terminology from the schema recommendation:

schema document:  an XML document conforming to the schema for schemas and 
containing declarations (part of) one namespace.  Schema documents can 
include and redefine each other, so one namespace may be "declared" in 
multiple files. 

schema:  informally, the declarations you need for validation.  More 
specifically, a collection of information modeled as schema components 
that meets certain constraints set out in the schema recommendation. 
Typically, but not always, that information comes from a combination of 
multiple schema documents, along with certain built in declarations.

schema file:  not a term used by the recommendation.

Concerns:

Schema documents on the web are like all Web resources identified by URI; 
by contrast, schemas are abstractions which are often but not always 
implied by a combination of such documents.  The schema WG believes it 
would be desirable to have a URI for schemas as well as for schema 
documents, but contrary to what is stated above, there is in general no 
such URI today, and there is no agreed media type that represents such a 
composed schema. 

I suspect but can't quite prove that beyond the somewhat sloppy use of 
terminology lies troubling ambiguity or misunderstanding about what is 
actually being named in the IANA registry.  Dan's and Henry's concerns are 
more fundamental, IMO, but even if those were dealt with, I think more 
clarity is needed regarding what is being named in the specific case of 
XML Schemas.

--------------------------------------
Noah Mendelsohn 
IBM Corporation
One Rogers Street
Cambridge, MA 02142
1-617-693-4036
--------------------------------------








Dan Connolly <connolly@w3.org>
Sent by: www-tag-request@w3.org
03/15/05 11:29 AM

 
        To:     "Henry S. Thompson" <ht@inf.ed.ac.uk>
        cc:     "Martin J. Duerst" <duerst@w3.org>, www-tag@w3.org, (bcc: Noah 
Mendelsohn/Cambridge/IBM)
        Subject:        Re: IETF XML formats use lots of URNs for namespace names



On Tue, 2005-03-15 at 11:17 +0000, Henry S. Thompson wrote:
> Dan Connolly <connolly@w3.org> writes:
> 
> > [what's this use of URNs for namespaces in IETF work ???]
> >   targetNamespace="urn:ietf:params:xml:ns:pidf:geopriv10
> > [from] http://ietfreport.isoc.org/idref/draft-ietf-geopriv-pidf-lo/
> 
> So it turns out this is based on something called the IETF XML
> Registry (RFC 3688) [1], which seems to me to interact with various
> aspects of Web architecture.

Yes, re-reading it, I find...

"ns -- XML Namespaces [W3C.REC-xml-names] are named by a URI.  They
      have no real, machine-parseable representation."

vs.

"Another benefit of using URIs to build XML namespaces is that the 
namespace URI can be
used to identify an information resource that contains useful information, 
machine-usable
and/or human-usable, about terms in the namespace. This type of 
information resource is
called a namespace document.

...

The owner of an XML namespace name SHOULD make available material intended 
for people
to read and material optimized for software agents in order to meet the 
needs of those
who will use the namespace vocabulary."

 -- http://www.w3.org/TR/webarch/#namespace-document

Bjorn points out
  http://www.iana.org/assignments/xml-registry/

which means each urn:ietf:... resource has 2 names, which goes against..

"Good practice: Avoiding URI aliases

A URI owner SHOULD NOT associate arbitrarily different URIs with the
same resource."

 -- http://www.w3.org/TR/webarch/#avoid-uri-aliases



>   It provides for 'registering' assigned
> URNs for namespaces, DTDs, W3C XML/RDF Schemas and public identifiers.
> 
> It mandates the creation of a server by IANA (not yet in existence, I
> don't think), which will serve the definitive definition of the
> registered things.  It also says that you shouldn't count on the
> server to give you DTDs or schema documents . . .
> 
> It says nothing whatsoever about versioning. . .
> 
> This whole thing makes me quite nervous . . .
> 
> ht
> 
> [1] http://ietfreport.isoc.org/idref/rfc3688/
-- 
Dan Connolly, W3C http://www.w3.org/People/Connolly/
D3C2 887B 0F92 6005 C541  0875 0F91 96DE 6E52 C29E

Received on Wednesday, 16 March 2005 03:30:50 UTC