[ietf-dkim] drop requirement to sign "From" or other "originator" headers?

[ned+dkim at mauve.mrochek.com]

> In listening to the responses and looking over the text, I see
> general agreement, except for Hector --- but only from a very small
> set of people.  Also, saying everything is optional creates an
> interesting glitch in the text: currently at least one header field
> must be included, and changing h= so it can be empty will require
> more thought.

> So, for -04 I change my proposal to go back to the wording that says
> that From MUST be signed (since, as Hector points out, From is
> required, we don't have to worry about what to do if it is missing).
> We also said in -03 that "any header field that describes the role of
> the signer" MUST be signed; I'm inclined to leave that out of -04.
> There will informative commentary about what a signer probably wants
> to do.

> Folks OK with that?

+1

				Ned