Going the other direction by removing URI language from spec?

It has probably been proposed before (and may be too 
drastic), however, I was wondering about a variant on 
option #3 -- modifying the namespace spec to remove 
*all* text regarding "URI" and "URI Reference".

> OPTION 3: TREAT NAMESPACE NAMES AS LITERALS. Namespace Names 
> are just Text Strings 

cut "whose value should be expressed in URI Reference form"

> This is the behavior intended by the Namespace Working Group.
>
> +: All names remain stable , permitting reliable namespace-aware document
> processing.
> +: Names can be compared via a simple string comparison.

The above points would not change.

> +: It would support those users who have been using Namespaces as a way of
> binding to external references, relying on the fact that Namespaces 1.0
> permitted relative syntax; they would retrieve the string and the base URI
> and perform the absolutize operation ("absolution"? No, let's not go
> there...) in application code.

This change would become "application specific" behavior.  If Microsoft
has 100% control of the XML documents it uses... then it is more than
welcome to make the target a URI or URI Reference or what ever else
they fell might be necessary.  It just is... the spec won't support
the assumption that it is a URI.  Anyone is more than welcome to layer
another spec on top of namespaces which specifies a particular 
behavior for the attribute value.  I don't think it would be such
a slick idea; but that is my opinion.  This way it stays out of the
spec... kinda walking around the issue. 

> -: According to one subset of the community, this choice does violence to
> the concept of URIs, or at least to the concept that everything in the
> network-universe can be named via a URI Reference.

By eliminating the URI talk from the namespace spec; this crowd 
should no longer be offended; since the spec no longer abuses
URIs by only going half-way.

> ^:The wording of the Namespace spec would have to be cleaned up to clarify
> that these are not URIs, but strings in URI syntax.

Wording should be cleaned up to remove all occurances of
"URI" or "URI reference" 

> ^: Some of the other specs (eg XPath?) which reference Namespaces would
> need to be rewritten to reflect this understanding. Others already have
> this behavior.

I still agree here.  The XPath spec has no need to "resolve" the
URI (so far), so the absolutizing language should be eliminated.

...

On another note; I'd also strike all the stuff about partitions
from the namespace spec as well... there are not many namespace
dependent specs or applications yet; and a bulk of the ones that
do exist *ignore* the partition stuff anyway... such as XPath.

If the W3C isn't going to use the partition stuff in *all* of 
its namespace related specifications; then that language should
be dropped as it causes implementation confusion and hinders 
interoperability.  Especially if the partition explanation 
that details how it could be implemented (i.e. information
model stuff) is "not normative".

If the user community isn't implementing it beacuse it is
confusing and "perhaps not such a good idea in retrospect"
then it should just be crossed out before it ends up like
the TAB character in Makefiles.

Best Wishes,

Clark

Received on Thursday, 25 May 2000 16:42:47 UTC