Since this is entirely out of scope for the WG this will be my last posting
on the subject.
> I'm not sure that IDN is the place to discuss it,
It isn't. See the WG charter.
> but
> there are several potentially hairy issues associated with using
> ACE for local-parts. including:
> ...
> now it's quite possibly the case that all of these can be worked out
> satisfactorily, and it Sure Would Be Nice if the same encoding
> routine that is used for IDNs could also be used for local-parts
> of an email address.
The constaints on local parts are a strict subset of those on parts of
domain names, so I don't see a problem here.
> but it's a bit of a leap to say that whatever
> ACE is chosen for IDNs will Just Work for local-parts.
Given the constraints on ACEs I disagree, but again, even if I'm wrong I think
letting additional requirements from other identifier spaces leak in here is a
really bad idea.
Ned
