W3C home > Mailing lists > Public > www-tag@w3.org > March 2012

Re: either strengthen or retract httpRange-14(a)

From: Kingsley Idehen <kidehen@openlinksw.com>
Date: Sun, 25 Mar 2012 17:51:24 -0400
Message-ID: <4F6F935C.70601@openlinksw.com>
To: www-tag@w3.org, "public-lod@w3.org" <public-lod@w3.org>
On 3/25/12 5:13 PM, Jonathan A Rees wrote:
> I would have hoped that any change proposal coming in would at least
> cover the billions of RDF statements where a URI is used to refer to
> the resource whose instances come from the given URI, and not just any
> old information resource - for example, where there's no RDF that says
> anything to the contrary. I've been harping on this inadequacy of
> httpRange-14(a) for years, and repeatedly suggesting ways to fix it.
> Yet the proposals haven't done this - many of them repeat the useless
> "is an information resource" phrase that's in the baseline document.
>
> The change is pretty simple, you just replace the "is an information
> resource" language with your favorite formulation of "has the
> retrieved representation as an instance", in those situations (if any)
> where it applies.
>
> I have not submitted this as a change proposal, since it does not
> address the problem that the proposals were supposed to address
> (ISSUE-57, performance and deployment of discovery). But it would be
> easy to incorporate into most of the proposals submitted so far.
>
> Jonathan
>
>

With regards to (ISSUE-57, performance and deployment of discovery) 
i.e., reducing the challenges that 303 redirection posses to some 
systems, you have the following items which are already covered in 
section 4.1 of your doc [1]:

200 OK
Link: <http://example.com/uri-documentation>; rel="describedby"

303 See Other
Location: http://example.com/uri-documentation>

Either of the above introduce heuristics that existing Linked Data 
compliant user agents can choose to implement.

In the specific case of "Link:" you have the added benefit of a 
heuristic that works for Web developers that are interested in 
structured data and even high-fidelity linked data modulo anything to do 
with R-D-F. Basically, those that don't have any interested in 
processing HTTP response metadata simply have the option to process 
identical <link/> relations in the <head/> section of (X)HTML resources.

Links:

1. http://www.w3.org/2001/tag/doc/uddp-20120229/ -- Understanding URI 
Hosting Practice as Support for URI Documentation Discovery .

-- 

Regards,

Kingsley Idehen	
Founder&  CEO
OpenLink Software
Company Web: http://www.openlinksw.com
Personal Weblog: http://www.openlinksw.com/blog/~kidehen
Twitter/Identi.ca handle: @kidehen
Google+ Profile: https://plus.google.com/112399767740508618350/about
LinkedIn Profile: http://www.linkedin.com/in/kidehen








Received on Sunday, 25 March 2012 21:51:50 GMT

This archive was generated by hypermail 2.2.0+W3C-0.50 : Thursday, 26 April 2012 12:48:46 GMT