RE: HTTP Endpoints and Resources

Hello David, 

> 
> I like your recognition that 200 and 303 URIs have something 
> in common, but please don't refer to the 303 case has having 
> an "http endpoint"
> that "responds", because doing so would introduce the 
> unnecessary confusion of having the same URI denote two 
> different things: an "http endpoint" and the thing the URI 
> was intended to denote -- a person, for example.
> 

I think I have to disagree I'm afraid. It is absolutely clear to me that
in the case of non-information resources, that which is accessed is not
that which is denoted by the URI. The URI doesn't denote two different
things. It denotes one thing. HTTP provides the function that allows an
associated http:resource to be accessed. Pat has put it much better in his
response on this thread.

As you point out, in practice it's the server that actually emits all the
responses. But unless an http:resource has been established, 404 is all
that it can emit. So architecturally, I think of the http:resource as
emiting the response, whether that be a 200 or a 303 or whatever. As you
point out, in the case of a non-information resource like a person, it's
not the person that responds. My point is that what responds is the
http:resource associated with the URI that denotes the person. I see no
problem with that.

By the way, I've made the assumption that what I've labelled an
http:resource and what you've called an HTTP endpoint are the same thing,
and I still believe that to be the case. Apologies if I've misunderstood
your use of the term.

Best wishes
Rhys

Received on Thursday, 27 September 2007 03:25:29 UTC