Re: Announcement: The "info" URI Scheme

On 2003-10-03 15:20, "ext John Cowan" <cowan@mercury.ccil.org> wrote:

> 
> Patrick Stickler scripsit:
> 
>>>> You are aware, of course, that "Shakespeare" is a bunny? :-)
>> 
>> I challenge you to prove to me that that is true, based
>> solely on that stream of bits.
> 
> Nobody can prove anything based solely on a stream of bits.  For all
> you know, you are (and have been since the beginning of your net.life)
> the victim of a man-in-the-middle attack by evil beings from Beta Hydri.
> 
>> What trust can I place in the mnemmonic characteristics
>> of the URI? Perhaps the URI actually denotes a picture
>> of rabbit named "Fred" taken by a local plumber named
>> Shakespeare. Perhaps "Shakespeare" means "this is
>> what's for dinner" in some bizarre language, which
>> is what is really the linguistic basis for the URI.
> 
> Any of that could be true.  A little investigation of context shows
> that "Shakespeare" is the name of the rabbit.

Well, given the situation that SW agents are not quite as
intelligent as most humans, it would be hard to expect
SW agents to be able to deduce that, and even humans may
still be left arguing about such conclusions.

The goal of technologies such as URIQA, based on the use
of HTTP meaningful URIs, is to be able to publish formal
descriptions, in RDF, which say what resources are denoted
by such URIs, in a way that has a precise mathematical
(and thus provable) form of expression. One may still
disagree with the published assertions, but at least it
will be clear what they mean (at least as far as what
the RDF MT can tell us).

Regards,

Patrick

Received on Friday, 3 October 2003 08:41:37 UTC