In <b6f3275205010301055f447e00(_at_)mail(_dot_)gmail(_dot_)com> Peter Karsai
A few months ago I wrote an SPF implementation which complied with SPF
Classic (http://spf.pobox.com/spf-draft-200406.txt, now expired) and
While the copy of this version has expired off the IETF website, it is
still a very useful version to reference. It is what a majority of
SPF implementations conform to.
http://spf.pobox.com/draft-ietf-marid-protocol-00.txt, then the
That wasn't an SPF spec, that was a SenderID spec. SenderID is not
SPF, nor is SPF SenderID. SenderID uses some of the SPF technology,
but it has many other problems. I can not recommend using SenderID in
Now I am going to update this library to the latest version of the
standard, however I did not follow the SPF discussion here recently so
I have no idea which version of the draft should be implemented. Can
anyone shed some light on this? Is there an IETF-approved SPF document
or is it still a draft? If SPF is still in draft status, when can
developers expect SPF to go "official"?
In my opinion, the "best" SPF spec is draft-schlitt-spf-classic-00.
It can be found in the following places:
While this spec is not yet ready to be made an RFC, it is pretty
close. The semantics and functionality of it should, in practice, be
very close to spf-draft-200406, while the text of the spec has been
largely rewritten to make it clearer. It represents the combined
efforts of dozens of hard core reviewers and many more people who have
Now would be an excellent time to try to use this newer draft to
update your implementation. If you find problems with the new drafts,
or discover important (practical) differences between spf-classic-00
and spf-draft-200406, there is still time to fix the spec.