On Fri 2016-Jun-10 12:32:20 -0600, Tim Starr <timstar...@gmail.com> wrote:
I am not saying this is a good idea, but it sounds to me like what would
fit the bill here would be a new folder for each user called "Bounced" in
which they would see all messages sent to their email address but which
were bounced by their mailbox provider. However, that would defeat the
purpose of preventing sufficiently malicious email from wasting mailbox
provider resources, and would seem to be largely redundant with the
purposes of having a quarantine folder. It would allow for users to report
undesired bounces to the mailbox provider, though.
That doesn't really seem helpful to me. If you've actually rejected the
message and communicated that to the sending MTA, it's no longer your
problem. The sender can take up the issue of deliverability as you've
given them sufficient information to do that. We're dealing with the
in-between zone where a message has not been rejected at SMTP time, but
it's still spammy. The Junk folder is the means to catch FPs and for the
user to report those to their mailbox provider. The problem is that
silently discarding things after issuing a 25x removes that avenue from the
user and the message vanishes into /dev/null.
Honestly I would love to hear how other large mail hosts handle this. The
reasoning for discards after 25x have boiled down to "we operate at a scale
you can only imagine; it doesn't work that way" plus some layer >=8 issues.
That said, in this admittedly small sample group, I've seen this complaint
leveled against Hotmail and the related services multiple times, but not
against other large mailbox houses.
Does Gmail & Google Apps do the same? What about the large filtering
services? I and many others on this list do not operate at nearly the
scale of the MS services, but some other orgs *do*. Are we just not
hearing about similar behaviour at those orgs?
I think everyone gets that the preferred behaviour is to reject at SMTP
time, that it gets difficult/impossible to do the more tests you try and
stuff into the filtering decision making, and that we don't want
backscatter. But what options are there for working within those
parameters while still honouring that a 25x response means that the user
will have *some* indication of the message arriving, be that in their
inbox, junk folder, or even just a report page/link of "these things were
so horrible we did not even bother putting them in your spam folder"?
Stuff the quick stuff in at SMTP time eval and reject the most egregious
ones, do additional processing post-25x, deliver to junk folder if found to
be spammy and feed back information from post-25x processing into the quick
tests if possible (e.g. IP blacklisting for heavy offenders etc.). Is such
a thing feasible at über scale at play?
Dropping it on the floor is Not Nice. We obviously don't live in an ideal
world, but it would seem unfortunate for us to give up the goal of actually
following through on our attestation that we will deliver the message to
the user (even if in their naughty folder) because of scaling issues, if at
all possible.
This is not any slight against Michael: You provide a valuable bridge to
the community and it is greatly appreciated that you brave the onslaught
and offer insight into the inner workings of the machine. You've indicated
that you have a similar distate for silent drop and I think we're all on
the same page with the objective. I'm just hopeful there are alternative
means that *do* scale and could be adopted to restore balance in the
universe...
Tim Starr
--
Hugo
On Thu, Jun 9, 2016 at 9:52 AM, Renaud Allard via mailop <mailop@mailop.org>
wrote:
On 09/06/16 17:26, Steve Atkins wrote:
Actually, what I do is that when a mail goes to the junk folder, the
server gives a 5XX error message to the sender at the end of DATA phase.
So the sender, if real, knows something happened to his mail and that it
might not be read.
So if you mis-classify mail - and the fact that you *do* misclassify mail
is implicit
in your having a junk folder - users get bounced off the mailing lists
they've
subscribed to, despite having seen the mail arrive.
I do not really mis-classify emails. If it appears in the junk folder,
there is an extremely high chance that it's junk. In fact, I should
probably not have delivered it, and that's what I was doing before I
configured the junk folder. It's just done that to avoid the very rare
false positive.
If I look at my personal junk folder right now (2 weeks retention time),
it's 100% spam.
_______________________________________________
mailop mailing list
mailop@mailop.org
https://chilli.nosignal.org/cgi-bin/mailman/listinfo/mailop
_______________________________________________
mailop mailing list
mailop@mailop.org
https://chilli.nosignal.org/cgi-bin/mailman/listinfo/mailop
_______________________________________________
mailop mailing list
mailop@mailop.org
https://chilli.nosignal.org/cgi-bin/mailman/listinfo/mailop
