Well, to be truthful, what harms forwarding is mostly out of my control. To be frank, Google harms forwarding. For all of my users, 40-50% of outbound mail goes to Gmail. A good bit of that forwarded, and forwarding to Gmail is by far the most popular configuration among users. However, when doing so, you have to choose the lesser evil of the options in front of you:

1. Forward everything, let Google rate limit your IPs and begin viewing them as dirty because not only can you not filter 100% of inbound spam, but Google can be demonstrated as rejecting a forwarded email (with or without SRS) as spam that it would have otherwise accepted and delivered to a Gmail inbox if sent directly to their servers, increasing the counter on how many "spam" emails your IP(s) recently sent.

2. Heavily filter the mail. This is the what I do, and I think I have a good handle on it but it'll never be perfect. It'll never be a 1:1 of what Google considers spam, which will randomly infuriate a user. And even if you do manage a 1:1 of what Google considers spam, you have to refer back to the problem outlined in #1 which loops right back to the problem that infuriates the random user.

No matter what you do there are trade offs when trying to force one system to act as a front-end for another, where both systems are managed by someone else.

On 2021-10-12 14:04, Jaroslaw Rafa via mailop wrote:
Dnia 12.10.2021 o godz. 13:18:12 Jarland Donnell via mailop pisze:
Strong agreement here. Despite SRS I still think forwarding is one
of the major road blocks to progress with email systems.

No, it's the opposite. Things like SPF et al. are roadblocks to forwarding. Forwarding is a *very* useful technique, and what makes it more useful is that any MTA can just do it on its own, so to do forwarding you only need to
have a MTA installed. To "pull" mail from other mailboxes you need
additional tools, it's not a standard feature of any MTA, plus there is
inherent insecurity of storing your credentials to other mailboxes on the server that does the "pull". I've always viewed "pull" as a far inferior
solution to forwarding, a kind of a poor workaround provided by email
providers who can't or don't want to - for various reasons - provide proper
forwarding capabilities.

"Hardwiring" a particular sender domain to a particular sending IP, like SPF does, is a *step back* in email technology - not a progress. A receiving system should always assume that messages *can* be forwarded and *can* come
from different IPs than they were originally sent from.

Don't concentrate on eradicating forwarding. Concentrate on eliminating
things that *harm* forwarding.
_______________________________________________
mailop mailing list
mailop@mailop.org
https://list.mailop.org/listinfo/mailop

Reply via email to