Re: [Imap-protocol] DKIM signatures on this list

Bron Gondwana <[email protected]> Mon, 16 Mar 2015 07:40:37 +1100
Newsgroups gmane.mail.imap.general
Message-ID <1426452037.1937115.240685285.7A0571FD@webmail.messagingengine.com>
On Mon, Mar 16, 2015, at 02:05 AM, Joshua Cranmer wrote:
> On 3/14/2015 11:26 PM, Jonathan Guthrie wrote:
> > Beyond that, and beyond the procedures necessary to set it up, I
> > have no knowledge of DKIM.  I have no idea why Mr Levine assumes
> > that because I have a passing familiarity with RFC 3501 and RFC 2822
> > that I will necessarily have detailed knowledge of the contents of
> > RFC 6376. Upon reading his emails and reflecting on how such things
> > work, I do understand that a lack of signature is semantically
> > equivalent to an invalid signature, but I don't understand why there
> > is an assumption that this knowledge should be intuitively obvious
> > to the casual observer, even on a list such as this.
>
> If you're going to complain about people doing the wrong thing and
> breaking the specification, common sense says that you should
> understand what the specification actually says you're supposed to
> be doing.

Indeed, I had some sympathy for the "if you're modifying the message
then you should be stripping out sections that you are making invalid"
theory until I read that self-righteous whine about not wanting to
bother understanding anything that is required to inter-operate with
the rest of the world these days.

I still think it's polite to strip DKIM headers which have zero meaning
if you modify any of the fields which they sign, in the same way that
it's polite to trim the parts of a message you aren't replying to, and
post responses after what you are responding to (thanks Joshua for
doing both of those).

Bron.

-- 
  Bron Gondwana
  [email protected]
_______________________________________________
Imap-protocol mailing list
[email protected]
http://mailman13.u.washington.edu/mailman/listinfo/imap-protocol