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

Re: Editor's Draft of ISSUE-57 URI Usage Primer

From: David Booth <david@dbooth.org>
Date: Tue, 09 Oct 2012 12:32:30 -0400
To: "Henry S. Thompson" <ht@inf.ed.ac.uk>
Cc: Jeni Tennison <jeni@jenitennison.com>, www-tag <www-tag@w3.org>
Message-ID: <1349800350.18784.12397.camel@dbooth-laptop>
Hi Henry,

On Tue, 2012-10-09 at 10:55 +0100, Henry S. Thompson wrote:
>  . . . The landing-page/thing-described
> ambiguity is a problem of web architecture and for web architecture.
> The other problems (which are not problems which I recognise as what I
> believe linguistics or philosophy of language call *ambiguity*, but
> rather are problems of *vagueness*.  'Everest' is not ambiguous, it's
> vague.  But that requires another post) are problems which appear to
> me to be shared by virtually _all_ human-engaged naming systems
> (i.e. not purely computational ones, such as programming-language
> identifiers).
> So the first (landing-page/thing-described) _must_ be addressed by web
> architecture -- it's _our_ problem.  But the second (vagueness of the
> thing identified) is just a fact of life, which by-and-large works to
> our advantage.  It may give rise to difficulties in some, particularly
> formal deductive, circumstances, but as such it _isn't_ just our
> problem, and doesn't make sense to me to expect or seek a
> web-architecture-specific solution.

I think not, but let's suppose for a moment that it is, and examine this
more deeply.  Bearing in mind that the context of this issue is the
RDF / Semantic Web use case:

1. What exactly do you mean by ambiguity, and how does it differ from

2. Why exactly do you think this landing-page-versus-thing-described
ambiguity is specially important to Web architecture?   What exactly
would break if it were not resolved?  

I can see that some applications would not work properly with
landing-page-versus-thing-described ambiguity, such as those that need
to understand what license is associated with what content.  But this is
true of *any* kind ambiguity.  For example, an application that needs to
count the relative incidence of each strain of influenza would not work
properly with a URI that (ambiguously) denotes the influenza virus but
fails to distinguish between different strains.  Why exactly do you see
the landing-page-versus-thing-described case as fundamentally different
and important to Web architecture?

David Booth, Ph.D.

Opinions expressed herein are those of the author and do not necessarily
reflect those of his employer.
Received on Tuesday, 9 October 2012 16:33:05 UTC

This archive was generated by hypermail 2.3.1 : Wednesday, 7 January 2015 15:33:17 UTC