Re: sqlite3 usage for multisession variable storage

[Eli Zaretskii]

> From: Lars Ingebrigtsen <larsi@gnus.org>
> Date: Wed, 15 Dec 2021 02:34:11 +0100
> Cc: emacs-devel@gnu.org
> 
> Robin Tarsiger <rtt@dasyatidae.com> writes:
> 
> > The difference is in crash recovery; if one instance crashes hard in the
> > middle of a transaction, it's "restore the file from backup" time.
> > Whether that's okay depends on your surrounding strategy, but it feels
> > similar to a subset of the lockfile strategies Emacs already considers
> > okay.
> 
> It's the "restore from backup" thing that's not acceptable, which makes
> MEMORY unworkable here, I think.

Maybe this aspect should be left to the application level (with some
reasonable default in place)?  It sounds like it's a judgment call
that can only be resolved with a specific application in mind.