SASLprep: Stringprep Profile for User Names and Passwords
draft-ietf-sasl-saslprep-10

[Ballot discuss]
Holding a discuss on the NFKC stability per Russ's request.

Original message from simon:

FYI, as SASLprep is using NFKC per Unicode 3.2, you might be
interested in the NFKC stability problem discussed on
.  The page mention a
strategy to maintain stability across NFKC revisions, which might be
relevant for SASLPrep.  I'd be interested to hear people's opinions on
the problem.

[Ballot comment]
One thing I think this document is missing is an applicabilty
statement. I.e, what I assume the applicability is is something like
the following:

    This document provides a way for protocols to specify comparisons
    of user identifiers and other "string" objects in a
    character-encoding neutral way. This document by itself does not
    change any existing protocols. Any protocol that wishes to use the
    techniques described in this document needs to explicitely do so,
    through a published RFC, in which precise details are given as to
    what objects are to be compared in the manner described in this
    document.

Or something like that. Right?

[Ballot discuss]
I'm not sure if this should be a discuss or just a comment...

This document refers several times to "simple user names", but it doesn't seem to define what a simple user name is. How are "simple user names" different from a "simple user names which require different handling"?  Why wouldn't this work for domain names?  Would it work in cases where an e-mail address is used as a user name (which is now quite common)?

I apologize if this is just immediately obvious to all stringprep-aware people...

[Ballot comment]
One sigh:
The formality of stringprep profiles is VERY high.
I wish it could include even ONE example of "string before and after".
I'm pretty sure this profile is case-sensitive (a and A do not match), but that is specified by omission, not commission (nothing about case folding). An example would probably make that 100% clear.