Re: Speed of keyboard macro execution?

[Eli Zaretskii <eliz@gnu.org>]

> From: Joseph Mingrone <jrm@ftfl.ca>
> Date: Fri, 11 Dec 2015 22:14:28 -0400
> 
> Eli Zaretskii <eliz@gnu.org> writes:
> > It's a bit disingenuous on the part of that person to compare Emacs
> > with another editor _solely_ where there's a known inefficiency in
> > Emacs, and pretend that this is somehow representative of the general
> > differences in performance between these two editors.  After all, how
> > frequently do you need to edit files with lines in excess of 11K
> > characters?
> 
> The intention wasn't to be misleading or trash Emacs, but to highlight
> an area ripe for improvement.

That's not what the text there says.  It says that mg is simpler, but
faster, and gives this particular example of how it is faster.  The
impression that a naïve observer will end up after that is that Emacs
is simply unworkable, since it cannot even scroll through a file in
some reasonably short time.

A more honest comparison would be to show scrolling through some
source file (with "normal" line lengths), which is something any Emacs
user does every day many times.  I'm guessing that Emacs will be
slower there as well, but instead of showing how it takes Emacs an
infinite time to do something, such an example would provide a much
more realistic comparison of relative speed between the two.

> I also assumed it was common knowledge for anyone that would be
> interested in such an esoteric topic that mg is a shell of Emacs in
> terms of features, so the comparison is "contrived".

That's exactly my point: it's so contrived that it loses credibility
in the eyes of anyone who knows a little about Emacs.

> Below is a comment I added to the video shortly after it was posted.

No amount of commentary will make that video less lopsided.  A fair
comparison should simply compare other use cases, that's all.