-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1

In message <[email protected]
il.com>, Wei Chuang <[email protected]> writes

>I'm announcing the Internet-Draft for the "Domain Name specification for
>DKIM2".
>https://datatracker.ietf.org/doc/draft-chuang-dkim2-dns/02/
>
>DKIM2 intends to be compatible with the existing DKIM installed base of
>keys hence this part of the specification is essentially the same as
>RFC6376 with an update for more modern algorithms.  However to prevent
>ambiguity with the parent document, this document carves out a subset meant
>to be used by DKIM2 and puts it in a separate Internet-Draft.  It also
>provides a place to document future changes for DKIM2.  Comments welcome.

First I want to thank Wei for creating this document ... our discussions
around DKIM2 led us to conclude that we did not want to create new keys
for our new protocol but to re-use existing "DKIM1" keys. However, we
feel that in the fullness of time DKIM1 will fall out of use and having
the new protocol rely on some sections of 6376 (and all of 8463) would
be inconvenient for future developers.

Whilst stressing that the purpose of this document is not to actually
change anything really significant about the way that DKIM1 keys are
stored and used (in either DKIM1 or DKIM2) I think that we should take
the opportunity to assess what is actually being used in practice and to
deprecate parts of the design that have not turned out to be useful (or
used at all ?).

To that end I wonder if the people who collect DKIM keys at scale and
analyse what they see (they publish data about key lengths from time to
time for example) are able to inform us as to which tags are actually
being seen in the wild ???

Anyway:

RFC6376 lists the fields which are actually needed for things to work:

v=  version
h=  hash function
k=  signing algorithm
p=  public key material

but it also provides for

n=  notes for humans
s=  service type
t=  flags
    y   testing
    s   match to i= required

I rather suspect that

    n= is seldom encountered (sysadmins document what they are doing at
    complete different stack levels);

    s= was a Good Idea At The Time but other protocols want their own
    key definition schemes rather than piggybacking here; and

    t= is commonly seen but pointless...

    We don't need, IMO, to complicate verifiers by telling them that
    although there is a DKIM signature (t=y) it isn't one really because
    we are hoping they will help us in their testing (they won't, they
    will reject the mail !) and i= (I'll leave looking up that obscurity
    as an exercise for the reader) is seldom used

So I would suggest moving these 3 tags to a different section,
indicating that DKIM1 verifiers may take notice of s= and t= but that
DKIM2 verifiers will not.

- -- 
richard @ highwayman . com                       "Nothing seems the same
                          Still you never see the change from day to day
                                And no-one notices the customs slip away"

-----BEGIN PGP SIGNATURE-----
Version: PGPsdk version 1.7.1

iQA/AwUBaHc+6GHfC/FfW545EQJGxgCfdpnvegYmNz91v0lqqkPdclLU1LgAoOLu
RQXGbK/yHAm3X7abIdKXf7ty
=ghLN
-----END PGP SIGNATURE-----

_______________________________________________
Ietf-dkim mailing list -- [email protected]
To unsubscribe send an email to [email protected]

Reply via email to