Meng Weng Wong wrote:
On Tue, Sep 14, 2004 at 08:17:51PM -0500, wayne wrote:
"Records that begin with v=spf1 MUST be interpreted as
spf2.0/mailfrom,helo"
I am in favor of this.
I agree with the sentiment behind this suggestion. It might
even be more useful if v=spf1 records were defined to mean
spf2.0/pra,mailfrom,helo,pra
Could you clarify this, please? I mean, currently, pra is not part of
v=spf1, right? I could see v=spf1 to mean this, though:
spf2.0/mailfrom,submitter,helo
I am not really sure, though, whether "helo" should be mentioned as a
separate entity. It feels to me that "helo" is an implicit part of the
mailfrom check (namely in the case of MAIL FROM: <>).
and if a sender is uncomfortable with those semantics then
they should publish proper spf2.0 records that override the
v1 entries.
As soon as this workgroup decides on its syntax, I will publish such new
records. :)
Well, having said that, in reality I will probably wind up publishing both
types of records, so as not to break downward compatibility.
- Mark