Should we make independent standards for MCE and OPC?

[MURATA Makoto]

Dear colleagues,

In Berlin, we discussed about the possibility of making Parts 2 and 3
independent standards rather than parts of 29500, although we have
made no decisions yet.

First, what is the procedure required for this move?  (Without committing
to such a move, we have to know the required procedure.)  New work item
proposals are probably needed, and the normal process for
standardization is used.  Withdrawing existing parts might be also
needed.  Parts 1 and 4 have to be changed so that they reference to
the new independent standards.  Kimura-san is going to contact ITTF
about the details.

Second, do independent standards for MCE and OPC make sense?  We have
to consider these two cases separately.

MCE provides generic mechanisms, which can be used from any host
markup language.  I am not aware of other specifications that provide
such generic mechanisms.  I am aware that some host languages (SVG and
EPUB) provide their own tiny mechanisms, which are similar to MCE.
These mechanisms are not separated from the host languages and are
specific to them.  A generic solution deserves to be clearly
documented and standardized, in my opinion.

Meanwhile, OPC is different.  OPC provides generic mechanisms, but
other specifications also provide generic mechanisms for packaging.
As far as I know, Part 3 of ODF1.2, OCF of EPUB3, and W3C Widget
Packaging are such specifications.  Long time ago, there was a W3C WG
(of which I was a member) for providing XML packaging using ZIP.
Unfortunately, no specifications were published.  We now have
proliferation of similar specifications, each of which has its own
advantages and disadvantages.  (Note: OPC lacks encryption)

If a new work item proposal for an independent standard for XML
packaging is made, it will probably open a can of warms: unification
of similar specifications for XML packaging.  I am now a bit
intimidated.  Wearing my IDPF hat, I hate delay of EPUB3.

Chris Francis wrote:
>
> I'm afraid I don't follow the logic.
>
> Either WG 4 as OOXML maintainers have specific requirements and require
> dedicated solutions OR it is a generic solution and need to be
> assessed/maintained/merged by the appropriate group who can put it onto a
> broader perspective and assure that conflicts and overlaps with existing
> and new proposals are handled.
>
> Otherwise I suggest there should be call out to liaisons and via SC34/JTC1
> to ask for their requirements of a generic standard or parts BEFORE
> wordsmithing the scope.  Otherwise by definition its still only based on
> the requirements of WG4.
>
> regards
>
> Chris
>
> Dr Chris Francis
> Technical Relations
> IBM
> 76 Upper Ground
> South Bank
> London SE1 9PZ

The JTC1 way to do such "call out" is a new work item proposal, which will
reach all liaisons,  and a ballot followed by it.



-- 

Praying for the victims of the Japan Tohoku earthquake

Makoto