What to do with directionality and ruby in ITS 2.0

Hi i18n core colleagues,

I have an action item in the MLW-LT group to check what to do
with directionality and ruby in ITS 2.0. The current definitions
http://www.w3.org/International/multilingualweb/lt/drafts/its20/its20.html#directionality
http://www.w3.org/International/multilingualweb/lt/drafts/its20/its20.html#ruby-annotation
are out of date in terms of what is proposed in HTML5, see e.g.
https://www.w3.org/International/multilingualweb/lt/track/issues/50
Also, the so-called "global" rules for ruby, see
http://www.w3.org/International/multilingualweb/lt/drafts/its20/examples/xml/EX-ruby-legacy-1.xml
have never been used AFAIK.

What would be your advice? Here are some ideas:

1) Drop global rules for ruby
2) Drop all definitions in the two sections completely and refer to the
HTML5 models for ruby and directionality informatively.
3) Keep the models for directionality and ruby and align them with HTML5.

A combination of 1) and 3) might make sense too.

Another question is: do these "features" need to be provided normatively?
After all rendering of dir and ruby and the related testing is done in
HTML5 or HTML5 based formats like epub3. This is also reflected in the fact
that in MLW-LT (= no rendering related folks in the group), there is nearly
no commitment on testing, see

https://docs.google.com/spreadsheet/ccc?key=0AgIk0-aoSKOadG5HQmJDT2EybWVvVC1VbnF5alN2S3c#gid=0

Thanks a lot for your input in advance,

Felix

-- 
Felix Sasaki
DFKI / W3C Fellow