> -----Original Message-----
> From: dane [mailto:[email protected]] On Behalf Of Lyndon Nerenberg
> Sent: Sunday, April 05, 2015 6:57 PM
> To: Paul Hoffman
> Cc: [email protected]
> Subject: Re: [dane] Proposed design for draft-ietf-dane-openpgpkey (and
> then for draft-ietf-dane-smime)
> 
> 
> On Apr 5, 2015, at 6:22 PM, Paul Hoffman <[email protected]> wrote:
> 
> > Greetings again. The discussion about exact-match and discovery in
draft-
> ietf-dane-openpgpkey has been useful for finding out what the use cases
are,
> and it's time to settle on a design that works for most people (we're
never
> going to make everyone happy).
> 
> How can we possibly do that without real experience in the field?
> 
> I know for a fact the case insensitive hashing rules will break Cyrus,
where
> foo+Bar@mumble will deliver to mailbox 'inbox.Bar', which is distinct from
> foo+bar@mumble, delivering to 'inbox.bar'.  Folder names are case
sensitive,
> so the two addresses are not equivalencies.

What I read of Paul's message was that the hashing would be on case
sensitive strings.  This means that the two strings above will produce
different results.  

> 
> Ultimately, I don't see a way out of this other than to make a standard
for sub-
> addressing, with a well defined cut-point delimiter.

If this is going to be done, then it will need to be back ported into the
S/MIME and PGP specifications as well.  The S/MIME specification says that
names SHOULD be matched case insensitive,   I was unable to find any rules
in the OpenPGP spec for how user IDs and email address are matched.
However, neither of these documents dealt with sub-addressing in any way and
would ignore the fact that one is trying to do this.  Thus foo+Bar and
foo+bar are going to be matched, but [email protected] and [email protected]
are not going to be matched by S/MIME implementations.

In S/MIME this decision was made to make most people happy.  However, I do
know that there are implementations where the names are matched case
sensitive rather than case insensitive.

Jim

> 
> But that still doesn't change how the LHS has been defined as
case-sensitive
> since the dawn of time.  If we are to change this, RFC5322 3.4.1 needs a
> change in its last paragraph.  As things stand, we cannot treat
<local-part> as
> case-insensitive in any interpretation of the grammar.
> 
> --lyndon


_______________________________________________
dane mailing list
[email protected]
https://www.ietf.org/mailman/listinfo/dane

Reply via email to