Re: httpRange-14 Adjunct: 302 is Valid for Non-Information Resources

On 4 Dec 2007, at 18:00, Sean B. Palmer wrote:

> On Dec 4, 2007 5:29 PM, Richard Cyganiak <richard@cyganiak.de> wrote:
>
>> Yes. But the vast majority of HTTP URIs are for traditional web
>> pages, and those don't return any RDF. We are left to guess what
>> they identify. Web pages? People? Things?
>
> Well yes, but httpRange-14 doesn't fully resolve this.
[snip]
> But you can see that in general it only helps when the resource
> returns a 200, because 303 doesn't mean <URI> a [ owl:complementOf
> web:InformationResource ], it means <URI> a rdfs:Resource!

You are right. As I see it, the value of httpRange-14 is that it  
clarifies the meaning of all those billions of 200-returning URIs out  
there. It does *not* tell us what a 303 URI identifies.

>> httpRange-14 axiomatically declares that for all those URIs, the
>> "naïve" interpretation is correct: They identify "the Google home
>> page"; "Richard's homepage"; "the TAG blog"; and so on.
>
> And you're saying that for the more irregular case of when a 303 is
> returned, generally that question isn't going to be asked
> anyway--especially if the page returns 303?
>
> That does actually make some sense because in the use case above with
> the [FAIL], "Identify the tool Amaya; the URI returns a 303", why is
> the page returning a 303? Quite probably because it wants to identify
> the tool Amaya and as a result it'll be giving you some 200'd RDF/XML
> on the other side of the 303.

Yes, exactly. People set up 303s to redirect to a document that has  
explicit (RDF) statements about the URI. The point of the 303 response  
is to enable that practice.

> Are you *sure* that people won't end up having the same questions for
> 303'd resources? And does it matter even if they don't?

No, I'm not sure. I just observe that the traditional Web contains  
billions of URIs that return 200, and a *much* smaller number of URIs  
that return 303 (and the 303 typically is returned as a response to  
POST requests, not normal GETs).

I guess it's fair to say that httpRange-14 has only pushed the problem  
into a smaller corner of the web architecture.

Richard



>
>
> (Cf. the other thread I just started where I ask exactly the same
> thing... I didn't expect these to converge within three emails!)
>
> -- 
> Sean B. Palmer, http://inamidst.com/sbp/

Received on Wednesday, 5 December 2007 09:57:19 UTC