Re: URI etymology

> I'm quite sure, by the way, that in some given applications,
> URIs will be "misused", i.e. used to identify other things
> that the one resource they theoreticaly identify; and the
> context will allow agents to handle that.

But what resource do they theoretically identify? There is no
definition of that: there is no central repository of "this is what a
URI means", there is no way in which that can be done. What we do
agree upon is, for example, protocols to act upon URIs, such as HTTP,
so that we may get representations of URIs back; but even that isn't
set on stone. If we all decided that HTTP was crud, and that we wanted
to use some new system, then the HTTP specification would be
redundant.

URIs are basically impossible to "misuse" in the context of what
you're saying, because they do not operate on Platonic ideals. People
can say what they want about whatever they want. It is just that very
often you will have some very tight context which everyone agrees
upon - for example the XHTML namespace is the namespace for XHTML.
It's very unlikely that everyone in the world will boycott it and use
it for something totally different, but the possibility still remains.

This is all fundamental thinking behind the essence of what URIs are:
they're just strings with the simple rule that they're unique. Apart
from that, you can't really say anything - URI strings are opaque, and
totally defined by the context of their uses. The "resource" that they
identify is a conceptual mapping between the URI and the entity, so
it's not something that is written in stone somewhere.

--
Kindest Regards,
Sean B. Palmer
@prefix : <http://webns.net/roughterms/> .
:Sean :hasHomepage <http://purl.org/net/sbp/> .

Received on Friday, 15 June 2001 10:55:53 UTC