On Tue, Apr 13, 2004 at 03:56:18PM -0400, Shaun T. Erickson wrote:
|
| It would seem that perhaps I shouldn't use it, until it is a complete
| solution. It looks very promising, but I can't use something that breaks
| valid email. :-/
|
You're right. As with any transition there are some awkward spots.
Shaun T. Erickson has come in at the biggest awkward spot: we're at the
point where SPF is done and SRS is not. We need to really focus on SRS
before we can say our work is finished.
I am counting on two things to get us through the awkward spots:
- a certain amount of legitimate mail gets bounced by overzealous spam
filters today. During the SPF/SRS transition, a certain amount of
legitimate mail will get bounced as well. But I expect the overall
bounce rate due to SPF will be lower than the bounce rate due to
content filters.
- When legitimate mail gets bounced by an overzealous spam filter due to
content filtering, etc, the sender often *cannot* do anything about
it. They have no recourse. When legitimate mail gets bounced by SPF,
the sender (or forwarder) *can* do something about it. This makes a
critical difference.
Of course, the question is "what can forwarders do?"
Forwarders have to do SRS. Institutional forwarders like pobox.com and
gmx.net are doing it already.
But small ISPs and hobbyists may not have the resources to do SRS on
their own.
That's why the onus is on the SPF/SRS community to provide SRS patches
to MTAs, so forwarders can just upgrade.
So let's just do it! :)
And we have been doing it! We have been working on SRS patches over the
last few months.
We need to pause a moment to see where things stand. Can we review the
state of SRS support for postfix, sendmail, qmail, and the other MTAs?
The command-line interface makes it possible to just pipe to
srs-forward. I think that got written already ... ?
Shevek? James?
|
| It would seem that perhaps I shouldn't use it, until it is a complete
| solution. It looks very promising, but I can't use something that breaks
| valid email. :-/
|
You're right. As with any transition there are some awkward spots.
Shaun T. Erickson has come in at the biggest awkward spot: we're at the
point where SPF is done and SRS is not. We need to really focus on SRS
before we can say our work is finished.
I am counting on two things to get us through the awkward spots:
- a certain amount of legitimate mail gets bounced by overzealous spam
filters today. During the SPF/SRS transition, a certain amount of
legitimate mail will get bounced as well. But I expect the overall
bounce rate due to SPF will be lower than the bounce rate due to
content filters.
- When legitimate mail gets bounced by an overzealous spam filter due to
content filtering, etc, the sender often *cannot* do anything about
it. They have no recourse. When legitimate mail gets bounced by SPF,
the sender (or forwarder) *can* do something about it. This makes a
critical difference.
Of course, the question is "what can forwarders do?"
Forwarders have to do SRS. Institutional forwarders like pobox.com and
gmx.net are doing it already.
But small ISPs and hobbyists may not have the resources to do SRS on
their own.
That's why the onus is on the SPF/SRS community to provide SRS patches
to MTAs, so forwarders can just upgrade.
So let's just do it! :)
And we have been doing it! We have been working on SRS patches over the
last few months.
We need to pause a moment to see where things stand. Can we review the
state of SRS support for postfix, sendmail, qmail, and the other MTAs?
The command-line interface makes it possible to just pipe to
srs-forward. I think that got written already ... ?
Shevek? James?