Dnia 18.02.2026 o godz. 12:37:03 Bill Cole via Postfix-users pisze:
> When I reject messages from hosts on Digital Ocean or Linode or OVH,
> it is not to "punish" anyone. It is because those networks still
> have a >95% spam rate on the systems I operate. This is NOT intended
> to *punish* people for their poor choice of provider, it is a
> self-defense tactic. It harms NO ONE persistently.
[...]
> Rather than "collective punishment" that is "collateral damage,"
> i.e. unintended and unwanted damage that is unfortunate but is a
> natural consequence of legitimate self-protection. Unlike explosives
> from the sky, declined email only annoys senders rather than killing
> them and it is relatively easy to mitigate the harm.

It is easy to mitigate the harm, *if and only if* the recipient who is
causing this "collateral damage" is ready and willing to make exceptions for
legitimate senders - no matter how small perecentage this might be - in the
blocked networks. In other words: if you block some range at large, you
should always be ready to make specific exceptions in the block.

For me, the golden standard was set by the Russian "mail.ru" service. I
don't know if they still operate this way, but a few years ago, if you sent
mail to them and it was rejected (which happened to me), you got in the
rejection message a link to a very simple web form that you needed to fill
in, and in a few hours you were unblocked, without any hassle. It was my
most positive experience with any mitigation procedure for rejected emails
ever.

If everyone did like this, you could block whoever you want, no problem at
all.

> With Google and MS running so many people's email, that last
> condition is rather dubious. That is the biggest problem with email
> today.

As for this, I totally agree.
-- 
Regards,
   Jaroslaw Rafa
   [email protected]
--
"In a million years, when kids go to school, they're gonna know: once there
was a Hushpuppy, and she lived with her daddy in the Bathtub."
_______________________________________________
Postfix-users mailing list -- [email protected]
To unsubscribe send an email to [email protected]

Reply via email to