W3C home > Mailing lists > Public > w3c-rdfcore-wg@w3.org > October 2003

Re: 2396bis

From: Graham Klyne <gk@ninebynine.org>
Date: Fri, 31 Oct 2003 11:14:07 +0000
Message-Id: <>
To: Brian McBride <bwm@hplb.hpl.hp.com>, Jeremy Carroll <jjc@hplb.hpl.hp.com>
Cc: w3c-rdfcore-wg@w3.org

At 09:20 31/10/03 +0000, Brian McBride wrote:

>I interpret this as a proposal to remove test case
>   http://www.w3.org/2000/10/rdf-tests/rdfcore/xmlbase/test012
>which I suggest we discuss at today's telecon.

I was already thinking of suggesting that this test case be removed, since 
it doesn't really illustrate anything about RDF, but about URIs.

>A possible response to the commentor might be:
>Thank you for pointing out this proposed change to RFC 2396 which we had 
>been unaware of.
>RDFCore have resolved to remove this test case.
>RDFCore would advise implementors that consume RDF that the RDFCore specs 
>are based on RFC 2396, a strict interpretation of which states that URI's 
>with too many ".."'s in their path are an error, though many URI 
>implementations correct that error and RFC 2396bis proposes to require 
>that correction.  Implementors are free to choose whether to strictly 
>comply with RFC 2396 or be more liberal.

I have broad agreement with the thrust here, but the last sentence might be 
interpreted as official permission to be non-standard.  I suggest dropping it.

>RDFCore advises implementors and content creators that produce RDF that 
>creating URI's with too many ".."'s in their path is inadvisable and may 
>lead to interoperability problems.

I agree, with caveat as noted.


Graham Klyne
For email:
Received on Friday, 31 October 2003 06:36:30 UTC

This archive was generated by hypermail 2.3.1 : Wednesday, 7 January 2015 14:54:08 UTC