W3C home > Mailing lists > Public > www-ws-arch@w3.org > December 2003

Re: new resource model

From: Francis McCabe <fgm@fla.fujitsu.com>
Date: Fri, 19 Dec 2003 09:22:15 -0800
Message-Id: <E3A8DFB8-3247-11D8-A4A4-000A95DC494A@fla.fujitsu.com>
Cc: www-ws-arch@w3.org
To: Mark Baker <distobj@acm.org>

Mark and others fishing in the pond ...

On Dec 18, 2003, at 6:08 PM, Mark Baker wrote:

>
> Hi,
>
> On Thu, Dec 18, 2003 at 03:45:49PM -0800, Francis McCabe wrote:
>> This is a revised resource model diagram.
>>
>> I have removed the confusing service model aspect of discovery, and
>> added in agents using discovery services to discover resources.
>
> I wouldn't think that a resource model needs a separate discovery
> service.  Representations facilitate discovery.

Discovery is an essential aspect of resources. Representations can be 
*used* (under some circumstances) to facilitate discovery, but its a 
category error to assert that representations are the means of 
discovering a resource. (Consider a picture of Dan's infamous car)



>
> In general, I'd prefer the removal of service, discovery service, and
> resource description for similar reasons.  If the group wants to
> describe how resources relate to services, you'd need to introduce
> a "Web server" into the diagram I think.  But all resource descriptions
> are representation, and all representations are resource descriptions.
>
>> It does
>> not seem to be necessary to add publishing, but I am open to that.
>
> If the group wants to go there, I'd suggest saying that an agent can
> publish a representation (a Web server could too, depending upon your
> definition of "publish").
>
>> I have added in that resources may have representations. I realize 
>> that
>> services may not have useful representations, but other resources do.
>
> Though I'd prefer service not be mentioned, all resources - even
> services - have useful representations.

Mike has commented on this. Its not obvious that a representation of a 
service has much utility.

>
>> I have also tied in a resource's URI to the resource description: 
>> i.e.,
>> its not a description of the resource if it does not identify it. This
>> is potentially controversial - it disallows descriptions of the form:
>> all resources which are colored red.
>
> As above, I don't see the need to distinguish between resource
> description and representation.

descriptions and representations are quite different. A description is 
a meta-statement and a representation is, if anything, an object-level 
(as opposed to meta-level) copy of the resource.

>
>> A person who owns a resource also set policy on it. That policy may be
>> linked to from the description. Again, there may be others who *also*
>> set policies on resources (the hosting agency for example).
>
> I'll have to think more about that one ...
>
> BTW, I'd also suggest that the agent discovers the URI rather than the
> resource.

No, the agent is discovering the resource. That is what it is 
interested in. The URI is but a means to an end.

Frank
Received on Friday, 19 December 2003 12:58:39 GMT

This archive was generated by hypermail 2.2.0+W3C-0.50 : Tuesday, 3 July 2007 12:25:24 GMT