RE: access to MT editors version

> I would be particularly interested in comments on the treatment of
> datatypes with badly-formed lexical forms, in section 3.4.

That looks fine to me (more than fine really - a clever step to avoid
problems that we don't want).

I liked the word "recommends" in the list semantics section. (The other
words were OK too).
But I think I might suggest a significant strengthening of that using MAY,
MUST, MAY NOT style terminology - anyway I will make that suggestion in a
full review, after you give the good ahead for that phase.

Maybe we should just take a snapshot of where your at, and review that?

Jeremy