Thanks for the elaboration. Doesn't that make 4314's use of identifiers qualify as Indefinite? (If so, Andrew's table should be updated.)
-Dave -----Original Message----- From: Alexey Melnikov [mailto:[email protected]] Sent: Sunday, July 17, 2011 11:33 AM To: Dave Thaler Cc: Andrew Sullivan; [email protected] Subject: Re: [precis] I-D Action: draft-ietf-precis-problem-statement-03.txt Hi Dave, Dave Thaler wrote: >Andrew Sullivan asked: > >>I'd be especially interested in whether people think that table is >>worth anything, before I complete it. I'm of two minds. >> >> > >I'd say yes. > >To its credit, it made me go look at one of the reviews (Alexey's >review of >RFC4314 at http://www.ietf.org/mail-archive/web/precis/current/msg00086.html) >for more info. I'm not that familiar with IMAP or SASLprep so I am confused >about >one thing and hope someone can explain it in simple terms for me :) > >The review says: > > >>Most likely case sensitive. Exact requirements on >>case-sensitivity/case-preservation >>depend on a specific implementation, e.g. an implementation might >>treat all user identifiers as case insensitive (or case insensitive for >>US-ASCII subset only). >> >> > >But RFC 4013 says: > > >>This profile is not intended for use in preparing identity strings >>that are not simple user names (e.g., email addresses, domain names, >>distinguished names), or where identity or password strings that are >>not character data, or require different handling (e.g., case >>folding). >> I don't think there is necessarily a contradiction. Case folding can be applied as an extra step after RFC 4013 processing. >The table Andrew added has "a,d" (but not "i"). That would be correct if all >implementations agreed on the case sensitivity rules. And RFC 4013 says it's >not appropriate for things that aren't case sensitive. > >So I'm confused about the part of the review saying: >"Exact requirements on case-sensitivity/case-preservation >depend on a specific implementation, e.g. an implementation might treat >all user identifiers as case insensitive (or case insensitive for US-ASCII >subset only)." > >Can someone elaborate on this? > > Ok, let me try to elaborate on my review and why I wrote what I wrote. I was talking specifically about IMAP/POP/SMTP usernames (passwords are another story). Many email servers support all of the three protocols and use the same identity format in all three. For a given email address the corresponding user identity is either the left hand side of the email address and/or the whole email address. (This isn't documented anywhere, but it is one of the things that all email implementors need to know.) The domain part (if included) is case-insensitive. SMTP (RFC 5321) says that the left hand side is case sensitive. In practice this restriction turned out to be quite problematic in real deployments (because humans just don't pay attention to case sometimes) and most of the systems I know of treat left hand sides as case sensitive. However until and unless an update to RFC 5321 declares that all left hand sides of email addresses are case insensitive, we can't say that left hand sides are always case insensitive, as some systems might rely on case sensitivity of left hand sides. Does this help? Best Regards, Alexey -- Internet Messaging Team Lead, <http://www.isode.com> JID: same as my email address twitter: aamelnikov _______________________________________________ precis mailing list [email protected] https://www.ietf.org/mailman/listinfo/precis
