Good point. OTOH, if we defined a special version of nameprep for
local-parts to allow mixed-case, then local-parts of internationalized
email addresses would be subject to case-sensitive comparison.
Are you referring to the encoded or unencoded (original) representation?
I'm assuming that the comparison will be done using the encoded version,
because many tools (including but not limited to SMTP servers) will not
be ACE-aware and will treat the addresses as if they were ordinary ASCII.
ACE encoded representations of variable-case local-parts will always fail
since the encoding sequence will differ for each input.
If case-folding is done before encoding as in nameprep, then the comparison
will succeed. If the case-folding is not done, then the comparison will
fail unless the two local-parts being compared are spelled with the same
casing - essentially forcing case-insensitivity.
OTOH, if the mail
system extracts the original unencoded sequence as part of delivery then
it can do whatever comparisons it wants.
Which essentially means that if the nameprep for local parts does not
do case-folding, then the burden for doing the case-sensitive comparison
moves to the SMTP server and other tools that must handle the mail - not
a desirable result IMHO.
The upside with mail is that comparison is purely a local issue (in
theory) as opposed to DNS where comparison occurs at every cache in
between the end-points.
It's more a local issue in theory than in practice. Lots of things
depend on the ability to compare email addresses - mailing lists
that check against a list of subscribers being one common example.
Keith