wayne wrote:
The one James referred to is spf-draft-200405.txt
Yes, it was only new for me that somebody (= James) considers
this text as the "real SPF". It explains match_subdomains=yes,
that's why I copied it.
Your list is far from complete, as far as old SPF drafts go.
I'm concentrating on relevant stuff with my own definition of
"relevant" ;-) From my POV something is "relevant" if it's
apparently implemented somewhere, or submitted to the IETF, or
it contains good ideas not covered elsewhere (like Kucherawy),
or simply because I like it.
May 16 2004 draft-mengwong-spf-01.txt
May 16 2004 spf-draft-200406.txt
Identical minus one minor point and the pagination.
May 10 2004 spf-draft-200405.txt
James' bible with match_subdomains=yes
Apr 25 2004 spf-draft-200404.txt
Is that "relevant" from my POV ? I'm not planning to create
a complete "SPF history", William already did this (for 2003).
I'll try and get an schlitt-spf-02 out with another stab at
the Received-SPF: parts soon.
Good idea, try to get rid of minor issues before the "council"
attacks any remaining major issues.
http://archives.listbox.com/spf-discuss(_at_)v2(_dot_)listbox(_dot_)com/200411/0929.html
That's something I didn't understand, therefore I ignored it,
mentioning only the obvious FWS and "empty comment" problems:
<http://article.gmane.org/gmane.mail.spam.spf.discuss/12618>
If you want _real_ comments, use CFWS instead of what you have,
but a _real_ comment is what the name says, its content has no
semantics, and its syntax is limited to reach the closing ")".
The "comment" in Received-SPF has its own structure => no CFWS.
You could pretend that it is a CFWS and hide its then obsolete
structure in an example (=> kids, don't try this at home... ;-)
I didn't replace SPF with TXT everywhere, just many places.
In all examples and whereever you don't need the SPF RR. For
the "namedroppers" I'd recommend exactly the opposite strategy.
Personally, I don't think that exclusively using a DNS RR in
the spec that won't exist on most people's name servers for
many months or years is very productive.
The IETF "namedroppers" won't care about our personal believes,
they have their own religion, and in this religion (ab)using
TXT is blasphemy. And because that's true to a certain degree
you just can't do it, unless you intend to commit net suicide.
BTW, I've mentioned two other issues in schlitt-01: 1*ALPHA
(=> copy LDH from 1035), and %{d} after "zone cut" (use always
the domain of the record for %{d} instead of one very obscure
exception), for details see the link to the archived article.
Bye, Frank