Stuart D. Gathman wrote:
it also avoids the SPF problem with "traditional forwarding".
There are *no* SPF problems caused by "traditional forwarding".
We likely agree, but JFTR I really consider *accepting* SPF FAIL
as dangerous. Most of the time this will be spam with a forged
envelope sender address, users might put it in an (unread) spam
folder where it's automatically purged after say 30 days. They
could miss SPF FAIL "false positives" caused by their unilateral
forwarding arrangements. Neither the admin of the forwarder nor
the admin of the next hop at a 3rd party (in this example Gmail)
necessarily know what's going on. And unlike rejecting SPF FAIL
accepting SPF FAIL can go wrong (legit mail rots in spam folder
until purged).
more difficult for a large ESP, since it involves getting
accurate info on the forwarders from millions of users who
have likely forgotten that they ever requested forwarding.
Exactly...
This is why large ESPs like gmail tend to punt and not reject
outright on FAIL.
...with the worse outcome for clueless users, as explained above.
Unlike the senders, who are supposed to know what SPF FAIL is,
these receivers are IMO entitled to be clueless. Accepting FAIL
is the contrary of helpful in "clueless forwarding" scenarios.
There is simply no problem other than the usual problem of
getting non technical users to configure stuff correctly - and
that is not a specifically SPF problem.
ACK. In the case of Gmail "use POP3 polling if you can" is the
best solution. Did you ever read Google FAQs (outside of their
various "API" topics for developers) ? They always limit their
answers to a text fitting on a postcard using very simple words.
Frank
-------------------------------------------
Sender Policy Framework: http://www.openspf.org
Archives: http://v2.listbox.com/member/archive/735/=now
RSS Feed: http://v2.listbox.com/member/archive/rss/735/
Modify Your Subscription:
http://v2.listbox.com/member/?member_id=2183229&id_secret=80102526-1e0cb5
Powered by Listbox: http://www.listbox.com