> On May 7, 2019, at 1:38 PM, Stephen Farrell <stephen.farrell@xxxxxxxxx> wrote: > > Question for ya on that Barry - do you think that MUA > and mail server implementers would actually bounce > messages as strictly as Martin's document might call > for? I'm not one of those implementers, so I don't know, > but I'd not be surprised to hear that in fact they'd > continue to prioritise mail delivery (for non spam) > over protocol purity. The alternative is what I see in "modern" specifications, like e.g. QUIC. Something like "MUST NOT" send extensions that have been previously negotiated and "MAY" terminate the connection with prejudice if the peer does something illegal. And yes, implementations follow the MAY path. -- Christian Huitema
