RE: Resource definition

I agree.  BTW, I'd wanted an id: scheme for a while, so as to remove the
notion of access methods from identity.  But it turned out not to make a
difference.

Our definition of Web service specifies that it is identified by a URI -
which therefore means it is a Resource.  We could elaborate a bit and say
that "A Web service is identified by a URI - and as such is a Resource -
...."

> -----Original Message-----
> From: [email protected] [mailto:[email protected]]On
> Behalf Of Michael Mealling
> Sent: Tuesday, February 18, 2003 3:05 PM
> To: James M Snell
> Cc: Mark Baker; David Orchard; 'Cutler, Roger ""(RogerCutler)';
> [email protected]; [email protected]
> Subject: Re: Resource definition
>
>
>
> On Tue, 2003-02-18 at 17:47, James M Snell wrote:
> > Hmmm.. the assertion that "All Web services have identity, and are
> > therefore resources" doesn't seem right to this lurking
> observer.  I view
> > "Web Services" as one way of getting to a resource, not as
> the resource
> > itself...
> >
> > A Resource has an identity and a collection of mechanisms a
> resource
> > consumer can use to get at that resource.  Web services are
> just one of
> > those mechanisms.
> >
> > I believe my disconnect here is purely a matter of
> semantics so perhaps
> > you could ellaborate a bit more.  What am I missing?
>
> You're making the assumption that the access methods are
> external to the
> resource when they are actually part of it (depending on the
> URI scheme
> you're using of course). If you use an http URI for some resource you
> are making the statement that the Resource being identified has the
> semantics that the 'http:' scheme gives it, which include things like
> methods 'n such. But its still an abstract concept you're identifying.
> A web service can just as easily be a Resource as anything
> else can be.
> A Resource can be _anything_ with identity: that includes
> cars, people,
> cats, services, emotions, galaxies, atoms, collections of bits sitting
> on a hard drive, nothing, everything, randomness, etc. If any of those
> things become Resources once someone bind a URI to that
> concept then why
> can't  a Web Service be one as well? Since when did a Web
> Service become
> some extra-universal uber concept higher than any other concept?
>
> -MM
>
>