--- Michael Thomas <[EMAIL PROTECTED]> wrote: > That is not correct. The local part of the i= is intended to > provide a binding to the local part of outside origination > headers, not just the domain part. Which is why it is, > in fact, a primary goal. > > One only has to look at Yahoo's web mail interface to > understand what significance they place on that binding.
Eh? Which binding is that particularly? As I understand it, Yahoo's web mail interface identifies the verified domain with a message like: "Yahoo! DomainKeys has confirmed that this message was sent by *verified-domain*." The binding of interest, both in terms of rendering and input to internal reputation engines is on a domain basis (and bilaterally selector). The g= is an admittedly crude attempt to constrain the use of delegated keys and is not intended to be of particular interest to a verifier above and beyond ensuring the constraint is complied with as part of the verification rules. Similar to IDENT, localpart strikes me as being of most interest to the sending domain rather than the receiving domain. Certainly other receivers might find localpart interesting, but for the foreseeable future, it's not of much interest to me. Mark. _______________________________________________ ietf-dkim mailing list [email protected] http://mipassoc.org/mailman/listinfo/ietf-dkim
