Re: Emacs design and architecture. How about copy-on-write?

[Po Lu]

Eli Zaretskii <eliz@gnu.org> writes:

> Then try leaning on C-n or C-p _after_ everything is already
> fontified.  You will still see that Emacs sometimes cannot keep up,
> especially if lines are not too short.

Long lines are, at worst, infrequently encountered in source code,
aren't they?  And our troubles with long lines are an algorithmic
impediment, which cannot be ameliorated merely by redisplaying each
window on a different CPU.

> You will never succeed in convincing me that our redisplay is as fast
> as I'd like it to be.  I've seen too many proofs to the contrary.

>From my POV, we frequently encounter circumstances where GC, Font Lock,
process filters or post-command-hooks precipitate difficulties keeping
Emacs responsive to user input.  Seldom are such problems unequivocally
attributed to redisplay itself.

I guess we can only agree to disagree.