Re: Determining what a URI identifies

Paul Prescod writes:

>> You want the processor to infer the type of an 
>> object from the syntax of its URI? What happened 
>> to opacity of URIs?

I've thought for awhile that the whole question of URI opacity might be 
something that the TAG would do well to clarify.  At some level, opacity 
is indeed an important, and the need to treat URI's as opaque is often 
misunderstood.  At other levels, it's clear that URIs are not opaque after 
all:  retrieval strategy is often (but not necessarily, I think?) keyed to 
the scheme;  RFC 2396 makes clear that hierarchical names are supported 
architecturally with "/" syntax.  In short URI's are not nearly as opaque 
as, for example, GUIDs.

So, I think the TAG would do the community a service if it would, in the 
course of dealing with the other URI-related issues on the table, provide 
a little more clarity on the senses in which URIs are and are not opaque. 
(For the record, I'm not suggesting anything is broken -- just that going 
back to the first time I saw TimBL's articulation of the opacity principle 
I've felt that a bit of clarification would be helpful.)  Thanks.


------------------------------------------------------------------
Noah Mendelsohn                              Voice: 1-617-693-4036
IBM Corporation                                Fax: 1-617-693-8676
One Rogers Street
Cambridge, MA 02142
------------------------------------------------------------------

Received on Wednesday, 6 November 2002 08:53:24 UTC