W3C home > Mailing lists > Public > www-tag@w3.org > August 2007

RE: ISSUE-58: Scalability of URI Access to Resources

From: Pat Hayes <phayes@ihmc.us>
Date: Mon, 27 Aug 2007 12:08:49 -0500
Message-Id: <p06230903c2f8b443ca1d@[10.100.0.28]>
To: "Booth, David (HP Software - Boston)" <dbooth@hp.com>
Cc: "Chimezie Ogbuji" <chimezie@gmail.com>, <www-tag@w3.org>

>  > From: Chimezie Ogbuji
>>  [ . . . ]
>>  2. The dereference problem is scheme independent
>>
>>  The second part of this particular point assumes there will
>>  *inevitabely* be a need to dereference these (insert your
>>  favorite other scheme here) URIs.  This is not always true,
>>  especially when the URIs in question are RDF URIs.  RDF URIs
>>  and their use have a model-theoretic mechanism for making
>>  claims about the world.  In most cases, these claims (very
>>  mathematical in nature) are meant to be much more authoritative
>>  than what representation you might get from dereferencing
>>  the URIs themselves especially when the claims are subject to
>>  much more fine-grained constraints through the use of a formal
>>  (OWL) ontology.
>>  [ . . . ]
>
>I think it is reasonable to assume that a URI consumer will inevitably
>want to find out what resource that URI is intended to denote, i.e., to
>obtain its URI declaration[1].

Well, no, it's not reasonable to assume this, since there is 
absolutely no reason to suppose that such a 'declaration' will exist. 
They don't exist in the currently accepted SWeb design anywhere, and 
no specification or standard yet written, or even contemplated, comes 
near to specifying such a thing.

>  The URI declaration *is* the
>authoritative information about what resource the URI denotes.

It might be if there were any, but one of the reasons that there 
aren't any is that this idea of there being a single authoritative 
information source is itself rather at odds with the SWeb design.

>  So it
>sounds like you are describing a situation in which the representation
>obtained via follow-your-nose[2] from the URI is *not* the
>(authoritative) URI declaration,

Yes, but...

>  but the (authoritative) URI declaration
>is provided in some other way.

...no. What can 'declaration' mean for an assertional language? Bear 
in mind that OWL has no way to express a formal definition (other 
than by making an assertion of equivalence.)

Pat

>  If that is what you mean, then I do not
>think that would be a good practice, because unless the representation
>somehow explicitly indicates that it is *not* the (authoritative) URI
>declaration, it is likely to mistaken for one.
>
>I'm not sure this impacts the point you were trying to make regarding
>Issue-58 though.  It sounds like you are saying that in some cases it is
>more efficient to bundle together a number of URI declarations in a
>single document, such that the client will not have to individually
>dereference each URI to find its declaration.  I think that's quite a
>reasonable desire.  At present, if HTTP URIs are used and
>follow-your-nose[2] is assumed, we only have a way to do that with hash
>URIs.  (URIs with the same racine[3] but different fragIDs all resolve
>to the same document.)  It would be good to be able to do it with
>303-URIs also, which is why in [1] I suggested that it would be helpful
>to have a way to explicitly express URI declarations.  If a URI
>declaration predicate like dbooth:declares were available, then when the
>first URI of a related bunch is dereferenced to find its declaration,
>the resulting document could declare all of the other URIs in that
>bunch, and a client would then know that it did not need to dereference
>each one to find its declaration.
>
>References
>1. URI declarations: http://dbooth.org/2007/uri-decl/
>2. Follow-your-nose: http://esw.w3.org/topic/FollowYourNose
>3. Racine: http://esw.w3.org/topic/HTML/Glossary
>
>
>David Booth, Ph.D.
>HP Software
>+1 617 629 8881 office  |  dbooth@hp.com
>http://www.hp.com/go/software
>
>Opinions expressed herein are those of the author and do not represent
>the official views of HP unless explicitly stated otherwise.
>


-- 
---------------------------------------------------------------------
IHMC		(850)434 8903 or (650)494 3973   home
40 South Alcaniz St.	(850)202 4416   office
Pensacola			(850)202 4440   fax
FL 32502			(850)291 0667    cell
phayesAT-SIGNihmc.us       http://www.ihmc.us/users/phayes
Received on Monday, 27 August 2007 17:09:04 GMT

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