The future of "Transitional": Japanese concern

[Francis Cave]

Murata-san

> (Again wearing my Japanese hat)
> 
> > The agreed UK position is that the purpose of the Transitional
> conformance class and
> >features, as specified in Part 4, is to provide faithful
> representation
> >of legacy documents.
> 
> It appears that your understanding of "legacy documents" appears to be
> different from ours.  In our understading, document that have been
> or will be created, modified, or viewed  by legacy software are
> legacy documents.  After all, isn't it very difficult to create strict
> documents using existing MS Office in the future?

Sorry for any misunderstanding. By "legacy documents" I do of course mean
documents produced by legacy systems, which includes pre-existing documents.
The UK fully supports the right of users to continue to use legacy systems
for as long as they wish!

> >If the old versions of Microsoft Office and other systems that were
> >used to produce these legacy documents were unable to format text in
> all
> >languages correctly (which is hardly surprising, by the way), Part 4
> >should enable us to represent those documents faithfully INCLUDING all
> >typographic lacunae. In other words, if a typographic feature was
> >missing and could not be expressed in a legacy document, it would be
> >wrong to try to rewrite history by enhancing Part 4 and thereby
> >"correcting" the presentation of old documents.
> 
> An important omission in MS Word is not  inability to format documents
> correctly but rather inability to recognize page layout specified
> in terms of KIHONHANMEN (see 2.2.4 http://www.w3.org/TR/2009/NOTE-
> jlreq-20090604/#en-subheading1_1_4
> of the requirement document) rather than margins.  I guess that
> it is possible to create OOXML extensions so that documents can
> contain specifications in terms of KIHONHAMEN *as well as*
> margins.  Legacy software is expected to handle such documents
> and produce good layout using margin information, while new
> applications can further provide better user interfaces in terms of
> KIHONHANMEN.

I would still argue that adding a new (in the sense of not previously
supported by legacy systems) way of specifying page layout is a new feature,
even if the layouts produced are no different from those that could be
produced (perhaps with difficulty, using margins) on existing legacy
systems. 

If the new feature is simply a more readily understandable way (to Japanese
users) of expressing what can already be expressed as margins, and there is
an algorithm (at least in theory, if not in practice) for converting from
one expression to the other, there might be a case for saying this is not so
much a new feature as a more correct representation (for some
users/applications) of an existing feature (cf ISO 8601 dates). 

However, even with this interpretation, it still feels uncomfortably close
to being a new feature.

Regards,

Francis