Re: XML Blueberry (non-ASCII name characters in Japan)

[Joel Rees <rees@server.mediafusion.co.jp>]

David Brownell posited:

> > <half-joking>Actually, once the reference is available, I suppose
checking
> > on the availability of characters for markup would be the sort of job
> > managers love to assign to new hires.</half-joking>
>
> I don't know about that ... since it only takes a minute to run the
> markup through some "fully conformant" processor (that checks
> all the appendix B rules), getting any WFness errors, it's not a
> task that would keep anyone (new or otherwise) busy for long.

Oh, a couple of hours, maybe a day for really complicated stuff. After all,
the DTDs have to be checked sometime, and the checking process is just the
sort of thing to train or test new hires on. But unless UNICODE is
impossibly bug-free, we're still going to need the reference book that shows
which characters are supposed to be available for names.

Of course, I am speaking hypothetically. It's still more or less
wait-and-see on UNICODE, and there is a lot of hope that a Japanese solution
can be provided instead. We're mostly using JIS and relying on
round-tripping, and that may be the best solution, even for XML.

Joel