On Wed, Jun 15, 2016 at 10:47:07AM +0300, Gil Bahat via mailop wrote: > Your users will pay a price and netease will pay a price.
There's always a price. The costs associated with both FP and FN are non-zero -- although they might be negligibly small -- for either sender or recipient or both. Attempting to make the costs zero, just like attempting to make FP and FN zero, isn't possible. (Except for the edge cases of allow-all and deny-all, both of which do so beautifully but are not viable in real world mail systems.) Thus the trick is to attempt to minimize both FP and FN simultaneously, and to do so not only for the sake of our own operations, but for others. (It's often lost on newcomers that the Internet is a cooperative exercise. Nobody is too big or too important to be part of that cooperation.) This is not a solved problem in mail system engineering, but if all concerned make best efforts, we can asymptotically approach something that looks like a solution for all of us. Thus the secondary trick is to attempt to make mistakes (and incur their associated costs) noisily, so that both senders and recipients have a decent chance of identifying them and reporting them to the appropriate people who will -- we hope -- pay attention and do something useful. This in turn requires working RFC 2142 contact addresses that forward traffic to clueful, diligent, responsible, professional eyeballs who understand that every message must be read, understood, analyzed, acted on (if necessary), and answered (if appropriate). Sometimes that'll help ourselves. Sometimes that'll help others. Sometimes it'll help both or maybe some third party who doesn't even know. It doesn't matter who it helps: it's necessary cooperation. ---rsk _______________________________________________ mailop mailing list mailop@mailop.org https://chilli.nosignal.org/cgi-bin/mailman/listinfo/mailop
