Re: Are *relative* URIs as namespace nemes considered harmful?

-----Original Message-----
From: John Cowan <jcowan@reutershealth.com>
To: Tim Berners-Lee <timbl@w3.org>; xml-uri@w3.org <xml-uri@w3.org>
Date: Wednesday, May 17, 2000 12:35 PM
Subject: Re: Are *relative* URIs as namespace nemes considered harmful?


>Tim Berners-Lee wrote:
>
>> On the contrary, there is a serious one: you get things which were
>> (identical, not identical)
>> when compared as strings turning out to generate (not identical,
identical)
>> objects when
>> considered as URIs.  This seems to me to be untenable.
>
>However, we do not (yet) have any Recommendations that insist on treating
>namespace names as anything but strings.  There is still time to get it
>right, for suitable values of "right".


Actually, we do have XPath and RDF which are Recommendations which treat the
namespace
as a URI. XPath explicitly calls for absolutization before comparison.  For
RDF, every property
has a URI which is the concatenatoin of the namespace URI and the localname
of the
XML element which denotes it.

We have systems like XSV and Microsoft's code which actually use the
Namespace
URI to get the schema.

So there is a body of thought that absolutizing is right. And some code. And
some specs.
The same applied to literal comparison.  The difference is that is we settle
for URIs not
being URIs, then RDF comes down like a house of cards, wheras if we go for
absolutization
then we probably don't have any real documents which fail.


>> The XML specs have to go one way or the other: commitment to or
separation
>> from URIs.
>
>Compromise positions are possible.  For example, one might:
>
> 1) have two classes of namespace names, those which are syntactic
> absolute URIs and those which are not;
> 2) define identity as string identity;
> 3) say that non-URI names have no meaning outside the current document.

I don't see that you gain anything by this over just banning relative URIs.

Tim BL

Received on Thursday, 18 May 2000 18:36:43 UTC