-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1
Frank Ellermann wrote:
Julian Mehnle wrote:
Who said anything about a "remove-scope=" modifier?
I did. Wayne did in one of the 1000 threads about scopes before the now
started 1001st thread, something along the line of "v=spf1 scope
modifiers are no problem as long as they don't remove the MAIL FROM
scope". [...]
This is all clear. I just wondered why you brought "remove-scope=" up now.
We all know such a thing doesn't work.
So what caveats are you talking about?
Nobody has implemented it. Nobody is forced to implement it.
Sure, but that doesn't matter. According to Wayne, the wording now is:
| Without explicit approval of the record owner, checking other identities
| against SPF version 1 records is NOT RECOMMENDED because there are cases
| that are known to give incorrect results. For example, most mailing
| lists rewrite the "MAIL FROM" identity (see <xref
| target="mailing-lists"/>), but some do not change any other identities in
| the message. The scenario described in <xref target="forwarding"/>.1.2
| is another example. Documents that define other identities should define
| the method for explicit approval.
That's good. But the benefit of also explicitly defining an "addl-scopes="
modifier ("addL" is a common abbreviation of "additional") would be that
the detractors of "NOT RECOMMENDED" could no longer say that we
_generally_ don't want to provide a way to make use of existing v=spf1
records for the checking of other identities. Then, if they still want to
re-use existing v=spf1 records that _don't_ have "addl-scopes=", they have
to admit openly that they deliberately want to change the semantics of
over 750,000 existing records.
It's not that I would want anybody to actually implement "addl-scopes=".
If somebody does it, that's fine. If not, that's fine, too. The point of
specifying such a modifier is not to encourage implementation, but to show
how things should be done _if_ people want to re-use existing v=spf1
records for other purposes, and that domain owners can't be _forced_ to
allow the semantics of their SPF records to be changed.
Obviously I failed miserably in making it comprehensible. Otherwise you
would not propose the same idea with a new modifier name limited to
add1-scope properties.
No. You just failed to explain why everything that is semantically
isomorphic should be syntactically transformed into "op=...".
Look, I'm not strongly against using "op=" for this case, it just didn't
occur to me when I read your draft or when I suggested "addl-scopes=".
Also, I see no practical net benefit in losing the intuitive nature of
"addl-scopes=" (or even "scopes+=") in order to save 5..9 characters.
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.2.5 (GNU/Linux)
iD8DBQFCgLxRwL7PKlBZWjsRAhBOAKDuWfqVd2ylH9wc5THJdtSIRRfITgCfb6wB
/MkbZNmVD+UDqoCtdxXJLV8=
=p3H2
-----END PGP SIGNATURE-----