W3C home > Mailing lists > Public > www-smil@w3.org > October to December 2007

Re: SMIL 3.0 comment: Normative reference to W3C namespace policy

From: Philippe Le Hegaret <plh@w3.org>
Date: Wed, 31 Oct 2007 20:51:59 +0000
To: tmichel@w3.org
Cc: www-smil@w3.org
Message-Id: <1193863919.8359.37.camel@localhost>

On Tue, 2007-10-23 at 09:08 +0000, tmichel@w3.org wrote:
> Your comment on 10.3.1 Elements and attributes:
> > [[
> > This section is normative.
> > 
> > For the purpose of identifying the version and the language profile
> > used, SMIL host language conformant documents must satisfy the
> > following
> > requirements:
> > [...]
> > In case the SMIL host language conformant language profile has been
> > issued as a W3C Recommendation, the default namespace identifier must
> > satisfy the following requirements: 
> >      1. The URI is constructed conformant to the requirements set
> > forth
> >         by the W3C [W3C-NSURI].
> > ]]
> > 
> > It seems awkward for a W3C Recommendation to normatively enforce and
> > reference the namespaces policy of W3C:
> > 
> > 1. the namespace policy is a moving target. Are you expected
> > implementations to test the identifiers with regards to that policy?
> > 
> > 2. the namespace policy is already enforced any way for W3C
> > Recommendation, so I'm not sure what it brings here except unnecessary
> > text in the document. 
> > 
> > I'd rather leave it to W3C to manage their web space as they want.
> > 
> > Proposal: I would remove "The URI is constructed conformant to the
> > requirements set forth by the W3C [W3C-NSURI]." 
> > 
> > 
> > 
> > [1] http://www.w3.org/TR/2007/WD-SMIL3-20070713/
> 
> 
> Working Group Resolution:
> We have modified definition of xmlns in section "10.3.1 Elements and
> attributes" as follows:
> 
> xmlns
>     The xmlns attribute declares an XML namespace, and is defined in
> "Namespaces in XML" [XML-NS].
> 
> 
> We have removed "and is defined in "Namespaces in XML" [XML-NS]" in the
> above statement.

But my comment has nothing to do with section 10.3.1... The text I
quoted actually appears in section 2.5.3 [1]. Why should the namespace
policy of W3C be a normative reference in a W3C Recommendation? I'm not
talking about the W3C Recommendation "Namespaces in XML". It's a moving
target and can be updated at any time. If the profile has been issued as
a W3C Recommendation, it has to follow the publication requirements set
forth by W3C in any case, but this should have nothing to do with SMIL
3.0. Again, I suggest the text:
[[
The URI is constructed conformant to the requirements set forth by the
W3C [W3C-NSURI].
]]
in section 2.5.3, as well as removing the normative reference in the
reference section.

Thank you,

Philippe

[1]
http://www.w3.org/TR/2007/WD-SMIL3-20070713/smil-modules.html#smilModulesNSSMILHostIdentifierReqs
Received on Wednesday, 31 October 2007 20:52:10 UTC

This archive was generated by hypermail 2.3.1 : Tuesday, 6 January 2015 20:34:28 UTC