W3C home > Mailing lists > Public > www-tag@w3.org > May 2002

Re: [rdfmsQnameUriMapping-6] Algorithm for creating a URI from a QName in RDF Model?

From: Brian McBride <bwm@hplb.hpl.hp.com>
Date: Thu, 23 May 2002 10:52:34 +0100
Message-Id: <5.1.0.14.0.20020523103404.033115e0@15.144.25.13>
To: Tim Bray <tbray@textuality.com>
Cc: www-tag@w3.org
At 13:26 22/05/2002 -0700, Tim Bray wrote:

>One question that you don't address, but should, is whether this issue is 
>important and whether we ought to invest work in it.

With my RDFCore chair hat on, I say that RDFCore has resolved this issue to 
its satisfaction.

Sans chapeau: the web is using two independent naming schemes, URI's and 
qnames and that is asking for trouble.  My personal vote would be to fix it 
and soon.  How that stacks up against the TAG's other priorities I don't know.


>>>It wouldn't be that hard to write a simple rule for mapping qnames to 
>>>URIs.  A little thought shows that it the mapping would have to be reversible,
>>It would be helpful if you were to document the argument that leads to 
>>the conclusion that the mapping must be reversible.
>
>One of the design goals of namespaces is to support software dispatching 
>based on the namespace name.  If you can't figure out what the namespace 
>name is, you can't do this.

Do you have an example to illustrate this requirement?  I find this 
statement a bit weird.  There is currently no concept of a namespace in a 
URI, so how could  namespaces have a goal to dispatch on it?

Would it be a goal to associate an unambiguous namespace with all 
URI's?  What role do you see the IETF playing in this process?

Brian
Received on Thursday, 23 May 2002 05:53:19 GMT

This archive was generated by hypermail 2.2.0+W3C-0.50 : Thursday, 26 April 2012 12:47:07 GMT