Re: Occurences of <rules> element in document and linking

Hi Yves, all,


Yves Savourel wrote:
> Hi everyone,
>
> Piggy-backing on some time left at the end of the editors'
> teleconference today, Christian, Felix and I were able to discuss of the
> "Shall ITS allows several <rules> elements in a single document?"
> question as well as the "What the ITS xlink:href link points to
> exactly?" question, that were generated from the Versionning discussion.
>
> And we even reached a consensus that hopefully you will share:
>
> We propose that:
>
> -#1- Only one <rules> element should be allowed per document. We think
> it covers most of the cases, and allowing more than one would result in
> quite a few headaches in processing/conformance while not adding much
> benefits.

I have integreated a text proposal from Yves for the linking mechanism
into the draft, see
http://www.w3.org/International/its/itstagset/itstagset.html#link-external-rules
and please comment.

Note that if we go for our proposal #1, we need to change

1-1: The  rules   element must be part of the content model of at least
one element declared in the schema. It should be in a document model for
meta information, if this is available in that schema (e.g. the head
element in XHTML).

to

1-1: The  rules   element must be part of the content model of one
element declared in the schema. It must not be allowed to occur several
times. It should be in a document model for meta information, if this is
available in that schema (e.g. the head element in XHTML).

Cheers,

Felix

>
> This is a change from the current specification insofar as we didn't
> have any explicit wording forbidding several <rules> to be in a single
> file (and therefore it was implicitely allowed).
>
>
> -#2- The xlink:href attributes of the <rules> element points to an
> arbitrary XML document that contains a <rules> element. That element can
> be the root element or elsewhere in the document (and there is only one
> per #1).
>
> This is a clarification of the linking mechanism rather than a change.
> The next updated specification should include a section describing the
> linking mechanism as we agreed already.
>
>
> If you see any problem with either #1 or #2 please make sure to post
> your comments.
>
> Cheers,
> P.S. The discussion on versionning is still going on:
> See this thread:
> _http://lists.w3.org/Archives/Public/public-i18n-its/2006AprJun/0015.html_
>
>
>
>