>I think the WG was fine with sticking to the currently defined >experimental now, and now delay another year before starting the >experiment. Such rewrite rules could be added later by those who >actually care and run this and write software, so they can come back >to us with a proposal. It sounds like you want to invent a way to canonicalize address local parts. If you want to do that, OK, although it's been tried and failed many times before. (SMTP has two commands that nobody implements, for example.) But it'd be an update to 5321 since I don't think anyone believes it'd be a good idea to have one set of rules for e-mail addresses in actual e-mail and a different set here. >> These systems have 10^8 e-mail addresses, not 10^8 DNS records. ... >This document is not Mandatory To Implement. If a certain big provider >cannot implement this in their DNS, they should not deploy this >document. If your point is that this document should never be published >because some large providers might not be able to deploy it, please state >so clearly. It seems to me that the point of creating a standard is so that systems can interoperate. If we know in advance that the systems handling the majority of the world's mail are vanishingly unlikely to implement something, what's the point? >>>> Large mail systems typically partition the users ... >The hash does not make it to the mail server. It is only used in DNS >to find the public key. The recipient name is NOT changed. Here, I think we agree. In a partitioned mail system (which is not limited to the largest ones), all of the partitions would have to export all the keys to one monolithic database. That's what I mean by scaling badly -- the techniques which work fine to scale up mail systems don't work here. R's, John