Re: RFC 2616 vs. AWWW

As usual, I like your explanation very much. It borders on sophistry,
but that doesn't bother me much, or won't until my next conversation
with someone who's upset about the use of http: URIs to refer to
things that aren't network resources.

I've copied your email to the wiki page, reformatted a little bit.
Hope that's OK. I also softened my conclusions a bit.

The httpRange-14 resolution [1] is about identification (of a thing
by/to an http server), not reference. If we translate it from the
language of AWWW to the language of RFC 2616, we get

 * If a [network] resource responds to a GET request with a 303 (See
Other) response, then the thing identified [to the HTTP server] by
that URI could be any thing;

which is absurd since the thing identified by the URI is the resource
that's responding to the GET. If I understand you correctly you would
change the 'identified [to the HTTP server]' to 'referred to [by
someone]', yes? It's not clear who or what is doing the referring -
maybe RDF documents on the same server, or something like that.

(The text [to the HTTP server is mine, but I believe is a correct
interpretation of the RFC.)

The 200 rule would be

 * If a [network] resource responds to a GET request with a 2xx
response, then the thing identified [to the HTTP server] by that URI
is a [network] resource;

which would be tautologous (unless different agents are doing the two
acts of identifying??), requiring a similar change 'identified [to the
HTTP server]' to 'referred to [by someone]'.

Oddly, this rule doesn't tell you which network resource is
referenced; it could be one that's unrelated to the network resource
that responds to the GET.

(Aside: As I have said before, the text also needs to be changed to
say that the 200 only ''states'' that the referenced resource is a
network resource, not that it ''is'' one, since the server could be
wrong.)

httpRange-14 screws citation - we can't use URIs identifying
200-responding network resources to refer to things like journal
articles, which are (ontologically) not network resources even if
they're published online - but that's another story (yes, I'm baiting
you).

[1] http://lists.w3.org/Archives/Public/www-tag/2005Jun/0039.html

Received on Friday, 12 October 2007 03:12:19 UTC