W3C home > Mailing lists > Public > www-talk@w3.org > March to April 2002

Re: "resolution mechanism"

From: Simon St.Laurent <simonstl@simonstl.com>
Date: 12 Apr 2002 14:00:24 -0400
To: Mark Baker <distobj@acm.org>
Cc: www-talk@w3.org
Message-Id: <1018634432.2632.136.camel@localhost.localdomain>
On Fri, 2002-04-12 at 13:50, Mark Baker wrote:
> > Sure - DDDS lets you query for URI metadata.
> > 
> > http://www.ietf.org/html.charters/urn-charter.html
> > 
> > "Tell me about this URI..."
> DDDS appears to be analagous to HTTP HEAD; "tell me about this
> resource", whereas HTTP GET does what HEAD does, but also returns
> a representation of the resource itself.
> Does that make better sense?  If so, HTTP does both, whereas DDDS
> only does one.

I'm not convinced that doing both is exactly an advantage.  HEAD isn't
exactly a highly-praised verb in the REST pantheon in any event.

The response I seem to be getting when I ask about URI resolution is:
Despite years of recommending the use of URIs and insisting on the
brilliance of their design, we have absolutely zero clue how to process
URIs in the general case.  Although the reason for URI diversity is that
no one approach to identification is appropriate in every case, we
strongly recommend that developers use HTTP because it's the only kind
of URI whose resolution is understood.

Rather than make good on our promises about URIs in general, we think
it's clear that HTTP magically solves every problem you can present, and
you should just drop other notions of how to resolve URIs.
I'm afraid this is just downright laughable, not to mention a direct
contradiction to an enormous number of claims about the nature of URIs
and HTTP URIs in particular over the course of the namespaces debate. 
HTTP URIs come with an enormous amount of baggage, baggage that not
everyone wants to carry.

I'd have been happy if Namespaces in XML had recommended the use of HTTP
URIs and specified RDDL.  Instead, we've mucked around for years in bad
promises about URIs only to be told that HTTP is the magic answer.

I hope you at least appreciate why some of us are pretty bitter about
the current state of URI discussion.

Simon St.Laurent
Ring around the content, a pocket full of brackets
Errors, errors, all fall down!
Received on Friday, 12 April 2002 13:55:15 UTC

This archive was generated by hypermail 2.3.1 : Tuesday, 6 January 2015 21:33:04 UTC