Re: Indentation and gc

[Eli Zaretskii <eliz@gnu.org>]

> From: Ihor Radchenko <yantar92@posteo.net>
> Cc: arne_bab@web.de, spacibba@aol.com, emacs-devel@gnu.org
> Date: Sat, 08 Apr 2023 08:15:32 +0000
> 
> Eli Zaretskii <eliz@gnu.org> writes:
> 
> >> I am attaching some stats I recorded by varying gc-cons-threshold and
> >> using Emacs.
> >
> > Thanks.  This is just for 14 to 25 hours of uptime.  I think we need
> > statistics for longer periods of time.
> 
> Do you have a particular number of hours in mind?
> For me, Emacs uptime is usually several days and sometimes less when I
> am doing Elisp development.

Well, several days would be a good time frame, I think.

> There is a possible argument to be made that we should only look into
> the real uptime people use rather than trying to measure some
> unreasonable long uptime. It will also be more natural to collect
> statistics then.

I agree.  However, my production sessions normally run for several
weeks, and sometimes for several months.  I think at least some part
of the Emacs users, particularly those who don't track development
branches, have long-lasting sessions.  We should have statistics from
these long sessions as well.

> Maybe you have thoughts about what else can be deduced from the data I
> provided. Or about what else should be recorded. (For some context, I
> realized that it is necessary to subtract idle time when recording Emacs
> uptime. My overnight graphs show flat lines when Emacs is idle).

I think your data is sufficient for now, we just need it from more
users and for longer times.  Maybe later we will have insights that
will help us identify additional data to be collected.

Thanks.