On Sun, 5 Apr 2015, Jim Schaad wrote:
That's not what I'm reading out of the draft, which seems to me to call
out for
explicit case folding before generating the hash.
Yes - that is what the note at the bottom of Paul's message was about
(PaulW: you threw in downcasing before hashing, which is definitely not what
people agreed to at the meeting.)
Which I did, based on the following related text in a draft by John
Levine that I thought was something shared amongst the app/email people:
https://tools.ietf.org/html/draft-levine-dns-mailbox-00#section-1
Mail systems usually handle variant forms of local-parts. The most
common variants are upper and lower case, which are now invariably
treated as equivalent.
This would greatly enhance the lookup mechanism to help in case of "auto
uppercasing" that is very common in webforms and mobile phone input
methods via virtual keyboards. That is, if I start composing an email to
myself, after I type "paul@" it will help me and correct it to "Paul@".
I believe this use case is MANY times more common than case sensitive
LHS. And as PaulH already said, we are not changing the case of the
actual email delivery.
Otherwise, everyone generating OPENPGPKEY records basically has to add
a CNAME for hash(Paul) pointing to the hash(paul) OPENPGPKEY record.
So it is the lesser of two evils,
Paul
_______________________________________________
dane mailing list
[email protected]
https://www.ietf.org/mailman/listinfo/dane
