Hi people,

As my prior comments on this matter may suggest, I've long been in favor
of holding SASL hostage until we can get a full CAP 3.2 implementation
in place. However, I'm ready to bow to public pressure on this and
entertain a partial (hack) implementation, in part because I now
recognize that a valid use case may yet exist for wanting SASL alone
without proper capability negotiation (even after that's eventually
introduced [1]).

I therefore offer this compromise reluctantly and still fear that going
this route will only prolong our complacency and further delay the sort
of meaningful evolution our client desperately needs.

Thanks,
J.P.

P.S. Our ELPA recipe would need updating before these changes could
land.

[1] Speaking of which, various aspects of the attached bug set may
    benefit from additional context. And for that, I'd kindly direct you
    to bug#49860, whose WIP patches can be found here:

    https://emacs-erc.gitlab.io/bugs/49860/patches.tar.gz