On 20Aug25, Wes Hardaker via mailop apparently wrote: > > their short (300) TTLs have no connection to their deliverability issues.
> My associates (and I) have produced a number of academic papers on the > subject that show the benefits of a longer TTL value: > > https://ant.isi.edu/%7ejohnh/PAPERS/Moura18b.pdf > https://ant.isi.edu/~johnh/PAPERS/Moura19b.pdf > https://datatracker.ietf.org/doc/rfc9199/ I only skimmed, but do these papers directly address the question of email deliveribilty and short TTLs? Since email is merely a client of the DNS infrastructure, I'm curious as to whether the OP is suggesting that there is some non-standard behaviour prevalent across email systems which causes them to treat DNS responses differently (and incorrectly) from say, web clients. I ask because, as we know, short TTLs of the order of 10s of seconds are pervasive in the web world for CDNs. Given that the vast majority of DNS queries are web-related, if short TTLs are a problem, one would think it would be a well-known issue which would be addressed fairly promptly. So my question is, on what basis is there a belief that email-related DNS queries differ from web-related DNS queries to such an extent that it affects deliverability? Mark. _______________________________________________ mailop mailing list mailop@mailop.org https://list.mailop.org/listinfo/mailop
