Re: MULE: incorrect utf-8 recoding

[Stefan Monnier]

> > From: "Stefan Monnier" <monnier+gnu/address@hidden>
> > Date: Wed, 12 Dec 2001 12:19:30 -0500
> > 
> > > Well, for one thing, it changes the way Emacs encodes buffers with
> > > mixed iso8859-x character sets, right?
> > 
> > Well, yes, that's the whole point: currently it can't do it
> > unless you use iso-2022 whereas with ucs-tables it now can.
> 
> Sure.  But that means that when this mode is turned on, Emacs will
> suggest (or maybe even silently do it automatically; I didn't
> actually try) to save such a buffer as UTF-8.  What if some users out
> there actually _want_ iso-2022 in those cases?

Then, they should select it explicitly, of course.

> Note that I'm just thinking aloud; I don't actually know any users who
> set things up that way.  But if there are, they might not like the
> new default.

I'm pretty sure they're dwarfed by the number of users who can't
save their latin-9 buffer because it contains a few latin-1 chars.


        Stefan