Re: A use case for anon nodes - action from telecon

At 12:25 PM 7/26/01 -0700, pat hayes wrote:
>>I think that's one of the great unresolved issues of URIs in general.  I 
>>think either way is sustainable and results in the same models, but 
>>different folks seem to think differently.
>
>Maybe. I think a global unique-name model is sustainable only in a very 
>fragile sense: if everyone cooperates and nobody ever makes a mistake or 
>gets confused, then it will work. But one error, anywhere, could make the 
>entire edifice crumble, since nobody could then rely on the global 
>assumption being valid.  In fact, I will undertake to be the fly in the 
>ointment, and will offer a universal unique-name-wrecking service by 
>generating copies of RDF webpages with just the names changed.

Well, some folks' thinking as I understand it would have it that you have, 
by definition, created a distinct set of resources that happen to deliver 
the same set of entities in response to the same queries.

This is sort-of what I meant by saying they result in the "same model" 
(which is probably not exactly right), in the sense that the resource 
itself is a nearly invisible intermediary between a request that names or 
references the resource (by means of its URI) and a thing one gets back in 
response to the request.  RFC2396 talks of resources as "conceptual mapping"s.

This is all very woolly, and I'm not going to try and defend either view 
right now.  OTOH, if you (or someone) can come up with an irrefutable 
argument why one of the approaches won't work, forcing a choice could help 
to avoid a whole minefield of potential misunderstandings.

I think you're right that there is fragility in some of the assumptions 
about URIs.  I guess the theory is a lot about the way the Web SHOULD work.

#g


------------------------------------------------------------
Graham Klyne                    Baltimore Technologies
Strategic Research              Content Security Group
<Graham.Klyne@Baltimore.com>    <http://www.mimesweeper.com>
                                 <http://www.baltimore.com>
------------------------------------------------------------

Received on Thursday, 26 July 2001 16:35:57 UTC