wayne wrote:
Why do you assume that the examples are wrong and not the ABNF?
The ABNF is compatible with MIME and (2)822, the examples are not,
was that a trick question ?
Were the examples changed since draft-mengwong-spf-[01]?
No.
Was the ABNF?
Yes.
If one was changed, why?
We changed it because the old ABNF din't pass the MARID giggle test.
What do most SPF implementations generate?
Irrelevant, angle brackets in an unquoted value are just wrong.
Yet another bad side-effect of getting no proper IETF Last Call,
Bruce, Ned, and others would see such "obvious" nits immediately.
Why on earth did we pick <parameter> instead of <name-val-list>,
as it's used for the Received: header field ?
I don't agree that the solution is so obvious that such this
should be published as an errata without answering questions
such as above.
It is clearly an erratum. You can only pick <name-val-list> if
you don't like <parameter>, that's roughly "drop the '='".
it looks like most of the stuff on the errata page is bogus,
only the missing ABNF for the 'v' macro looks correct to me.
The "recommend an SMTP reply code for Permerror rejections"
should be s/Permerror/Fail/, maybe that's a typo (?)
The "permerror on invalid domains after macro expansion" is
about the thread on the devel list, if macro expansion results
in stuff like three adjacent dots that can't be queried, IIRC.
The "Danish" typo is MIA. An IESG typo is MIA (or rather it's
in an obscure "pending errata mbox" at the Rfc-editor). Very
vaguely I recall that somebody else submitted or tried to submit
errata (incl. the Danish typo), and this contained other nits.
There is no "uric" set in 3986, that's also MIA, Julian found
that nit some weeks ago.
Frank
-------
Sender Policy Framework: http://www.openspf.org/
Archives at http://archives.listbox.com/spf-discuss/current/
To unsubscribe, change your address, or temporarily deactivate your
subscription,
please go to http://v2.listbox.com/member/?list_id=735