Where did this information come from? Mozilla seems to mistakenly think this as well, but I've tested.
I checked some e-mails sent with recent versions of Outlook, and there is no User-Agent header. Outlook in the past did apparently send User-Agent for NNTP (back when it had support for that), but not for e-mail. Mozilla may have also been confused because older versions of Thunderbird would display X-Mailer from other clients as User-Agent in the header of e-mails by default, when you don't click on view source. So if you got that info from them, I think that's where the misinformation originated.
Interesting, this draft would actually have made both headers acceptable, because they are the most common. But it seems to have been expired/rejected for some reason? However, one piece of data in it is worth noting... your own draft shows that the people trying to write a standard noted that Outlook uses X-Mailer, not User-Agent.There is a draft that suggested it for mail and it is BCP.
https://datatracker.ietf.org/doc/html/d ... r-agent-00
Yeah, but if I can lower my SpamAssassin score by changing how my mail client identifies itself because it looks so close to a valid Mozilla string that they think I am "forging" it, then I don't mind changing it. I just don't understand the logic of that rule, all it does is require malicious actors to do a better job forging a Mozilla UA.In my opinion, spam filters must not check mails for "valid" or "common" User-Agent or X-Mailer headers, as they are not required.