W3C home > Mailing lists > Public > w3c-rdfcore-wg@w3.org > June 2002

Re: namespaces as punctuation, was: new semantics initiative

From: Patrick Stickler <patrick.stickler@nokia.com>
Date: Thu, 13 Jun 2002 10:54:18 +0300
To: ext Jonathan Borden <jonathan@openhealth.org>, RDF Core <w3c-rdfcore-wg@w3.org>
Message-ID: <B92E2A5A.16A76%patrick.stickler@nokia.com>

On 2002-06-12 15:42, "ext Jonathan Borden" <jonathan@openhealth.org> wrote:

> 
> Dave Beckett wrote:
>> 
>> 
>>>>> patrick hayes said:
>> 
>> <snip amount="lots"/>
>>> (6) Does this require any changes to syntax/ test cases/ Ntriples/
>>> datatyping/ whatever?
>>> A: No.
>> 
>> Until you had used the word 'namespace' in the bit I've cut from
>> above, I would say no.  When namespace appears in the MT, I'm
>> worried.
>> 
>> I just went and checked in the MT WD, and you do use it a few more
>> times than I expected.  Hmm!
>> 
>> Patrick Stickler has said elsewhere that "namespaces are punctuation"
>> which is a bit strong, but for RDF/XML that is mostly correct.  RDF
>> does not have namespaces in the model (theory).
>> 
> 
> Yes, well I don't agree that XML namespaces are just punctuation, but
> that is not important for this discussion -  I agree with what you say
> here. I suggest that the RDF MT can provide for sets of asserted and
> unasserted triples, or even for any number sets of triples colored by
> whatever name one wishes. For the purposes of RDF/XML _syntax_ the
> namespace is a perfectly reasonable way to distinguish which triples are
> intended to go into which bucket in an unambiguous and monotonic -- even
> monotonous -- fashion. I don't suggest that namespaces are the only way,
> simply that for the purposes of RDF/XML they can provide a _purely
> syntactic_ way for a parser to flip triples into buckets ... and with
> perhaps minimal, but nonzero, changes to the syntax.

I appreciate your view that namespaces are more than punctuation (though
I do disagree) but because namespaces are a point of strong contention,
I feel it is unwise to use them as a basis for identifying unasserted
triples in the RDF/XML syntax.

Also, it precludes

(a) using the same term in both asserted and unasserted triples
(b) using only some terms grounded in a given namespace in unasserted
    triples and others in asserted triples

Now, it may be that both of the above will seldom if ever be needed,
but the preclusion seems arbitrary and needless, simply due to the
coarse resolution of and global scope of the namespace prefix.

I feel that a generic mechanism, such as my proposed rdf:Notation
element, would serve the need just as well without imposing the
above restrictions nor requiring any vocabulary specific definitions.

Regards,

Patrick

--
               
Patrick Stickler              Phone: +358 50 483 9453
Senior Research Scientist     Fax:   +358 7180 35409
Nokia Research Center         Email: patrick.stickler@nokia.com
Received on Thursday, 13 June 2002 03:50:06 EDT

This archive was generated by hypermail pre-2.1.9 : Wednesday, 3 September 2003 09:49:15 EDT