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

Re: URIs vs. URIviews (was: Agenda for RDFCore WG Telecon 2002-02-15)

From: Pat Hayes <phayes@ai.uwf.edu>
Date: Thu, 14 Feb 2002 18:02:51 -0500
Message-Id: <p05101429b891f08d893d@[192.168.0.6]>
To: Aaron Swartz <me@aaronsw.com>
Cc: w3c-rdfcore-wg@w3.org
>I notice that Brian seems ready to close all the little naggling issues. I
>think this is great but I don't want to see some issues drop thru the
>cracks. Particularly, I'm worried about the URI-vs-URIviews issue, which I
>thought we agreed to put on the issues list, but I don't seem to see it.
>
>Specifically in:
>
>>  16: Issue rdfms-fragments
>>
>>  Propose:
>>
>>  o The WG resolves that the meaning of absolute
>>    URI's with fragment ID's is a matter of web architecture and
>>    beyond the scope of this WG and that this issue be closed.
>>
>>
>>  See:
>>  http://www.w3.org/2000/03/rdf-tracking/#rdfms-fragments
>
>I really can't agree with this. It's our problem that RDF uses this
>non-standard piece of Web architecture, and in doing so has incurred all
>sorts of problems. If we're going to be the Resource Description Framework,
>we need we're actually describing resources. My ideal resolution would look
>like:
>
>  o The WG resolves that the use of absolute URIs with fragment IDs is a
>    to identify Web resources is relatively incompatible with current Web
>    architecture.

In what way??

>  o We recommend that RDF users refrain from using '#' in their Resource
>    identifiers and namespaces.


Aaargh!!  No, we don't. Such use is ESSENTIAL. How else can one 
ontology use the names used in another ontology??

>RDF developers and tool creators may present
>    a warning to the user when using resource identifiers with '#' in them.
>
>  o We understand the need to identify portions of Web entities (data used to
>    describe a resource, such as the data returned when making an HTTP GET
>    request). We recommend that they identify such Resources using something
>    along the lines of:
>
>_:x rdf:type web:Fragment .
>_:x web:resource <http://www.w3.org/2001/tag/ilist> .
>_:x web:fragID "w3cMediaType-1" .
>_:x dc:date "2002-02-14T13:03Z" .
>
>My goal is to:
>  a) raise awareness about the problem while
>  b) maintaining backwards-compatibility but
>  c) lay the ground work for future WGs to fix this bug

My goal is prevent URIs becoming completely unusable in web ontologies.

>
>[...later...]
>>  (d) choose namespace names that end in non-xml-name-characters
>>  such as / # ?
>
>I think perhaps we should provide some warning about using # in namespace
>names, dependent on the resolution of rdfms-fragments.
>
>you're-not-getting-off-that-easy-'ly yrs,

Wait a minute. Whats the 'bug' here? If RDF (and DAML and OWL and...) 
are required to use urirefs  for all identifiers, and if urirefs have 
no way to be [a name as used inside another ontology which is in turn 
identified by a URI], then the entire semantic web is dead in the 
water. This absolutely MUST be possible somehow. Syntactic problems 
with URI formats are a comparatively unimportant technical issue. I 
would be very unhappy indeed if we were to go on record as 
deprecating the use of # if there isn't anything else to take its 
place. Without that, or something equivalent to it, web ontology use 
is impossible. And it had better be reasonable to use. That 
four-triple monstrosity is unusable, in particular.

Pat


-- 
---------------------------------------------------------------------
IHMC					(850)434 8903   home
40 South Alcaniz St.			(850)202 4416   office
Pensacola,  FL 32501			(850)202 4440   fax
phayes@ai.uwf.edu 
http://www.coginst.uwf.edu/~phayes
Received on Thursday, 14 February 2002 18:02:51 EST

This archive was generated by hypermail pre-2.1.9 : Wednesday, 3 September 2003 09:45:11 EDT