Re: Proposal to amend the httpRange-14 resolution

Hello David

Now that this conversation has turned a bit less noisy :)
What I have written recently is along the lines of  the distinction you
propose between definition and description, and the process you are
envisioning[1].
Kingsley has an amazing and enthusiastic faith in the power of the Web's
architecture, but this is not only technical, it is about a social process.
Agreed there is no way to desambiguate once for all a URI, but like in
natural languages there is a neverending quest towards accuracy and
disambiguation.
And indeed this has to start with the URI owner providing the first
description of the resource, acting as a definition, further descriptions,
either provided by the URI owner or other sources can be compared to the
definition to figure if they bring extra information, or other perspectives
on the resources, or if they are inconsistent with what the URI owner
asserts in the definition.

The fact that, as Pat Hayes and others have correctly pointed, over 99,9%
of URIs on the Web do not provide such definitions does not prevent to push
provision of such definitions as a best practice.

If you are a URI owner, and if you want your URIs to play nicely and be a
reliable reference in the Semantic Web, don't take the risk to see third
parties provide various and probably inconsistent descriptions of what your
URIs mean, based or not on debatable and various interpretations of the
semantics of HTTP GET answers.

Best

Bernard

[1] http://blog.hubjects.com/2012/03/beyond-httprange-14-addiction.html


Le 4 avril 2012 03:00, David Booth <david@dbooth.org> a écrit :

> Hi Kingsley,
>
> On Tue, 2012-04-03 at 15:01 -0400, Kingsley Idehen wrote:
> > On 4/3/12 1:46 PM, David Booth wrote:
> [ . . . ]
> > > This use of URI definitions helps to anchor the "meaning" of the URI,
> so
> > > that it does not drift uncontrollably.
> [ . . . ]
> >
> > But once on the Web the user really [loses] control. There is not such
> > thing as real stability per se. Only when you have system faults can one
> > at least pivot accordingly. Thus, you only get the aforementioned
> > behavior in the context of a specific system and its associated rules.
>
> I think you're right that we can never get total semantic stability in
> an absolute sense.  But if we establish a commonly followed convention
> in which the URI owner's URI definition is used when making statements
> involving a URI, then the semantic drift will at least be substantially
> limited.  Again, this does not require *everyone* to follow the
> convention.  But the more that do follow it, the more effective it
> becomes in making the web a sort of self-describing dictionary.
>
>
> --
> David Booth, Ph.D.
> http://dbooth.org/
>
> Opinions expressed herein are those of the author and do not necessarily
> reflect those of his employer.
>
>
>
-- 
*Bernard Vatant
*
Vocabularies & Data Engineering
Tel :  + 33 (0)9 71 48 84 59
 Skype : bernard.vatant
Linked Open Vocabularies <http://labs.mondeca.com/dataset/lov>

--------------------------------------------------------
*Mondeca**          **                   *
3 cité Nollez 75018 Paris, France
www.mondeca.com
Follow us on Twitter : @mondecanews <http://twitter.com/#%21/mondecanews>