Re: Let's make sure we've got nails and not screws...

[Patrick Stickler, Nokia/Finland, (+358 40) 801 9690, patrick.stickler@nokia.com]


----- Original Message ----- 
From: "ext Miles Sabin" <miles@milessabin.com>
To: "WWW-Tag" <www-tag@w3.org>
Sent: 19 November, 2002 12:14
Subject: Re: Let's make sure we've got nails and not screws...


> 
> Patrick Stickler wrote,
> [snip: lots of stuff I completely agree with]
> 
> > Because a namespace is simply a set of names, all that a
> > representation of a namespace could provide is a list of those names.
> 
> Not even that.

Well, in all fairness the XML Namespaces rec states "An XML namespace is 
a collection of names, identified by a URI reference" (actually, I was
wrong to call them a set).

So I think it's fair to talk of terms belonging to a namespace and
a namespace having (theoretically at least) a representation corresponding
to some sort of enumeration of the terms grounded in that namespace.

> The set of names within a namespace is completely determined by the 
> namespace identifier and the LocalPart production of the Namespaces 
> REC. There's no need, and it's not possible, to list them.

Good point. The membership of terms in a given namespace may be open-ended
and therefore a complete representation may not be obtainable in practice,
though that also does not preclude the existence of representations of
namespaces with known closed membership or partial representations of
namespaces based on known terms.

The key point here, is that a namespace does not assert any semantics for
its member terms and therefore a representation of a namespace should not
include any such semantics if it is to be a valid representation of the
namespace.

Thus, RDDL documents retrieved from namespace URIs violate the very 
fundamentals of the web architecture by being invalid representations
of the namespaces.

Cheers,

Patrick

Received on Tuesday, 19 November 2002 05:53:00 UTC