bidi algorithm in charter (was: Re: sharp s (Eszett))

[Martin Duerst]

At 07:22 08/03/08, Paul Hoffman wrote:
>... our current charter text:
>
>In particular, IDNs continue to
>use the "xn--" prefix and the same ASCII-compatible encoding,
>and the bidirectional algorithm follows the same basic design.

I have been strongly involved in bidi for IDNA2003.
I have no idea what is meant by
"bidirectional algorithm follows the same basic design".

One can easily question whether the provisions regarding bidi in
IDNA2003 are an algorithm at all. They only describe conditions
that have to be met in order for a label to be legal, not how to
(algorithmically) check for these conditions.

If the point of that phrase was to be vague, I think the goal
has been achieved. If what was inteded was something like
"bidirectional restrictions will be based on character
properties and will not require the application/implemenation
of the Unicode bidi algorithm", then it should say so.

If anything else was intended, that should be written down,
but also made available for comments.

Regards,   Martin.


#-#-#  Martin J. Du"rst, Assoc. Professor, Aoyama Gakuin University
#-#-#  http://www.sw.it.aoyama.ac.jp       mailto:duerst at it.aoyama.ac.jp