Re: emacs rendering comparisson between emacs23 and emacs26.3

[Eli Zaretskii]

> From: Richard Stallman <address@hidden>
> Cc: address@hidden, address@hidden, address@hidden
> Date: Wed, 01 Apr 2020 22:32:12 -0400
> 
>   > I think we do try: patch review frequently has these aspects
>   > discussed.  But it is a well known fact that humans are lousy
>   > predictors of code speed.
> 
> I'm not suggesting we try to _predict_ the effect on code speed,
> rather than we have a guideline to identify the cases where _if_
> the code gets slower it would be hard for a user to escape that.
> 
> What we would do, in those cases, is add a switch to turn off
> the changes.

Without some efficient method of identifying the cases where this is
probable, I fear we will have to add such switches for almost every
non-trivial change (and for some trivial ones as well).