Re: Uniform access to descriptions

Pat Hayes wrote:
> At 1:29 AM +0100 4/12/08, Xiaoshu Wang wrote:
>> <snip>
>>>>
>>>> Question 4: Is an HTTP-URI =  HTTP+URI?
>>>
>>> I have no idea what this means.
>> What I mean is this:
>> HTTP-URI is simply an HTTP URI.
>> HTTP+URI is when the HTTP URI is bound to the HTTP transportation 
>> protocol.
>> Hence, the question can be rephrased as such:
>> Is what a URI denotes the same thing as what the URI is dereferenced?
>
> Sometimes but also sometimes not. http-range-14 says that when the 
> response code is 200, the answer is yes. As I say, I don't like this 
> much either; but I can't see any feasible other way to answer the 
> question/ at all/ for a given URI.
>
> I take it that your answer would also be: maybe, maybe not; but that 
> you would want the decision to depend not on an http code, but instead 
> on some RDF assertions which would be accessible from the URI (in a 
> way I confess to not following yet, but ...) Is that right?
Yes.  That is my point.  But I am not stubborn and unwilling to accept 
any other model. It is because I don't see other models that can give me 
a clear and objective way to answer the four questions that I asked.

I think TAG's httpRange-14 is the following logic.

Representation=Resource if HTTP=200.

But Conneg breaks either the "equal" sign or the if clause. 

Xiaoshu