On Mon, Jan 12, 2026 at 02:17:48PM +0200, G. Miliotis via mailop wrote: > > On 11/1/26 12:32 μ.μ., Peter N. M. Hansteen via mailop wrote: > > > > when a domain has correctly configured SPF, DKIM, DMARC *that specifies mail > > claiming to be from that domain originating in the great elsewhere should be > > discarded*, and you count mail claiming to be from that domain that > > originates > > in the great elsewhere in the evaluation of that domain's reputation, > > **YOU'RE DOING IT WRONG** > > > > Well, to fix something like this one would need to know the nuts and bolts > of one's filtering system. Where would you put this 'short-circuit' if the > whole machinery has become too complicated for one person/team to tame?
These are really two separate issues. The article I pointed at earlier has an example of where these types of check would fit in. I would not want to comment on some codebase I have not seen other than if the code has become too complicated to maintain, that is a problem in itself and possibly a hint that at redesign and rewrite would be in order. > Especially now with people adding AI to everything and not understanding any > of its internal workings, I'm not very optimistic for the future of anything > requiring proper tuning. We're just training whole fields to trust a result > just because it's usually good - without caring about the why. People who put code into use without knowing how it works and its dependencies are in for a world of hurt soon anyway, unless of course the powers that be decide to not actually enforce the CRA and its eqvivalents. All the best, Peter -- Peter N. M. Hansteen, member of the first RFC 1149 implementation team https://nxdomain.no/~peter/blogposts https://nostarch.com/book-of-pf-4th-edition "Remember to set the evil bit on all malicious network traffic" delilah spamd[29949]: 85.152.224.147: disconnected after 42673 seconds. _______________________________________________ mailop mailing list [email protected] https://list.mailop.org/listinfo/mailop
