Re: Are MGET descriptions workable/necessary?

On Monday, Nov 24, 2003, at 12:29 Europe/Helsinki, ext Phil Dawes wrote:

> Hi Patrick,
>
> Patrick Stickler writes:
>>
>> [...]
>>
>> I'm sorry, but I don't see any difference. Cool URIs don't change,
>> period,
>> whether one uses them to interact with representations or 
>> descriptions.
>>
>> If a resource "moves", it gets a new URI, and the same redirection 
>> that
>> occurs for GET requests would have to happen for MGET requests.
>>
>
> Am I mis-understanding what you mean by resource?
> (I'm failing to come to the same conclusions as you, which probably
> means there's some misunderstanding)
>
> I was taking your use of 'resource' to mean any/every rdf term used in
> the semantic web. I.e. subjects, predicates and objects.
>
> If that's the case then I don't think GET requests for arbitrary
> representations are like MGET description requests at all, simply
> given the orders of magnitude involved.
>

MGET, MPUT and MDELETE are only relevant for URIs which are meaningful
to the HTTP protocol (or any other protocol for which bindings are
defined).

So there will be URIs used in RDF statements which will not be usable
with MGET, etc. because they are not dereferencible via HTTP.

For these cases, the URIQA model also defines a service interface, which
works for any arbitrary URI, but without question, the utility for such
URIs is significantly reduced.

IMO, though, folks should be using http: URIs to denote all resources,
abstract or otherwise, web documents or otherwise, and only use other
URI schemes when absolutely necessary and justified.

URNs, DOIs, and many other strongly promoted URI schemes are simply
unnecessary. One can achieve all the requirements/goals put forth for
such schemes using http: URIs and domain management.

Cheers,

Patrick


> Many thanks,
>
> Phil
>
>

Received on Tuesday, 25 November 2003 12:42:29 UTC