W3C home > Mailing lists > Public > uri@w3.org > May 2003

Re: Resources and URIs

From: james anderson <james.anderson@setf.de>
Date: Thu, 08 May 2003 09:48:55 +0200
Message-ID: <3EBA0ACC.DB9146DC@setf.de>
To: uri@w3.org

Tim Bray wrote:
> Patrick.Stickler@nokia.com wrote:
> > URIs used as XML Namespaces do *not* denote those namespaces.
> >

i am not certain what the denotation of an isolated namespace name is, but a
namespace name taken together with a local part denotes a univaral name just
as much as (+ 1 2) can be said to denote "the result of adding the number 1 to
the number 2." just as

<!ATTLIST e a (value) #FIXED 'value'>

can be said to denote an "empty element named [[e]] with an annotation named
[[a]] with a value [[v]]."

> > XML Namespaces are punctuation.

 ':' is punctuation. '<' is punctuation.

>      They are macros. They are
> > syntactic sugar for XML instances so that URIs can be
> > used as element and attribute names (by a little bit of
> > slight of hand).

they are also not syntactic sugar[0]. they contribute essentially to a form
which has denotation. a prefix is syntactic sugar.

<n:e xmlns:n='ns'/>

is a macro.

in any case, given the context of [1], is it possible to understand this claim
as an overstated distinction between namespace-as-a-set-of-names and
namespace-as-vocabulary (which is a set of temrns and relations). against
which distinction this reader has no argument. 

> You have now expressed this opinion on many occasions dating back quite
> a few months.

indeed, the earliest such expression appears to be about 16 months old[1]. on
the other hand, there were not entirely contrary opinions expressed as long as
almost three years ago[2].

>   I remain unconvinced.  Yes, the main utility of
> namespaces has been to sneak URIs into element/attribute names to
> support dispatching and collision avoidance.

at one point that was not their "main utility", it was their exact purpose.

>   But if we choose to say
> that the URIs thus used do actually identify resources, and provide
> representations of them, this breaks nothing and is actually useful in
> the real world.  So why not?

as long as the claim is "[URL]s thus used [can] actually identify resources,
and provide representations of them, [but these resources should not be
construed to either denote or 'be' the namespace.]" the claim would be
supportable. anything more makes a claim beyond that which a namespace name is
specified to mean.

until there is a revision to "namespaces in xml" which says "URIs thus used do
actually identify resources", and defines the relation between namespaces and
such resource, there are no grounds to say more than one provides
representations for urls.


[0] http://foldoc.doc.ic.ac.uk/foldoc/foldoc.cgi?query=syntactic+sugar
[1] http://lists.w3.org/Archives/Public/uri/2002Jan/0043.html
[2] http://lists.w3.org/Archives/Public/xml-uri/2000May/0016.html
Received on Thursday, 8 May 2003 03:41:02 UTC

This archive was generated by hypermail 2.3.1 : Tuesday, 6 January 2015 21:25:05 UTC