bug#48949: 28.0.50; Thread-Process Liberalization

[Eli Zaretskii]

> From: dick.r.chiang@gmail.com
> Cc: 48949@debbugs.gnu.org
> Date: Fri, 11 Jun 2021 06:26:28 -0400
> 
> Well, this response and that of #bug36609 effectively puts the kibosh on
> my dream of Gnus 2.0 (a working version at github.com/dickmao/gnus I
> run rather delightedly every day).  It's a more-than-5000 line change thus
> far, so given the vigorous pushback on what I considered to be
> non-controversial changes in #bug48949 and #bug36609, it never really had a
> chance.

I don't understand: why being able to wait for a process that is
locked to another thread is so important for what you wanted to do?
And if it _is_ important, then why not unlock the process, so that its
output could be processed by any thread?