Rewriting the scope of Part 3

[MURATA Makoto]

Chris,

Florian proposed that the MCE processing model should introduce an
initialization step,
which creates a set of understood namespaces and another set of
(namespace, local name)
paris.  The second set is used for addressing the processing inside
application-defined
extension elements.

Nobody argued against this initialization step.  The only difference
is that should we
introduce a markup language for describing the two sets.  Alex thinks
that it is convenient
for MCE testing while others did not look convinced.  Although no
final decisions are made,
I do not think that this issue hampers our attempt to make MCE generic.

As of now, MCE is not recognized as generic and the text is horrible.  WG4 is
certainly responsible.  I would have sold MCE to the EPUB WG if MCE was
cleaned up.

Regards,
Makoto

2011/7/1 Chris Rae <Chris.Rae at microsoft.com>:
> Hi all - if I remember correctly, we do have a slight problem with making MCE generic with regard to the suspension of MCE processing inside extLst collections. I know we talked in Berlin about some possible ways to get around that, but I don't think we settled on anything - I'd be retiscent to head down the road of generalising MCE before that problem is solved.
>
> Chris
>
> -----Original Message-----
> From: MURATA Makoto (FAMILY Given) [mailto:eb2m-mrt at asahi-net.or.jp]
> Sent: 29 June 2011 08:05
> To: SC34
> Subject: Re: Rewriting the scope of Part 3
>
> Doug,
>
> Thanks for your rewrite.
>
>> Doug then wrote:
>>
>> >     Good point, that's a potential are of confusion for those new to MCE.
>> >
>> >     Perhaps the best solution is to not attempt to define what MCE's
>> >     alternate content blocks might contain, but rather address the
>> >     purpose/application of ACBs in general. That would eliminate the need
>> >     to talk about "elements and attributes that do not exist ..." and then
>> >     we could describe the processed document in terms of what it contains,
>> >     rather than what it does not contain. In other words, instead of this:
>> >
>> >     "This Part of ISO/IEC 29500 describes a set of conventions to clearly
>> >     mark elements and attributes that do not exist in an XML-based markup
>> >     language but are introduced by extensions of the markup language,
>> >     while providing a method by which consumers can obtain an XML document
>> >     that are free from extensions for interoperability."
>>>
>> >     Something more like this:
>> >
>> >     "This Part of ISO/IEC 29500 describes a set of conventions to identify
>> >     alternative representations of XML markup and character data that a
>> >     markup consumer may choose from based on the namespaces understood by
>> >     that consumer, while providing a method by which consumers can obtain
>> >     an XML document that only contains elements and attributes that exist
>> >     within the namespaces of a single markup language, for
>> >     interoperability purposes."
>
> I think that this scope eliminates ignorable elements/attributes.
>
>> >     "A single markup language" ... "a markup language within which MCE is
>> >     being used?" They both sound awkward to me, although that's the core
>> >     concept involved. Thoughts?
>
> Host markup language?
>
>> >     Do we need to state that the namespaces may or may not be part of the
>> >     markup language within which MCE is used? Simply not constraining
>> >     "namespace" in any way seems sufficient to me.
>
> I do not think it is required to be stated.
>
> Cheers,
> Makoto
>
>



-- 

Praying for the victims of the Japan Tohoku earthquake

Makoto