> On Sat, 27 Dec 1997, Tuomas Heino wrote:
>
> > On Sat, 27 Dec 1997, Pavel Machek wrote:
> >
> > > Hi!
> > >
> > > > Maybe. It isn't as simple as changing the default to user defined map, but
> > > > if you say it would be better... Yes, it's more compatible. So could we
> > > > define such sequences NOW and implement them in the next kernel? It would
> > > > be really great!
> > > >
> > > > I can see two ways:
> > > >
> > > > 1. An escape would define the charset set by ^[c (maybe one of the four,
> > > > ISO-8859-1 / VT100 / null = CP-437 / user defined, or maybe only
> > > > ISO-8859-1 / user defined would be enough).
> > > >
> > > > 2. Escapes would tell if ^[c should set the G0 and G1 charsets at
> > > > all.
> > >
> > > This is *mess*. What about ^[c not reseting charsets at all, and ^[C,
> > > which would also reset pallette, 'saved' colors, charsets, everything?
> > >
> > Maybe that kinda thingy would be useful too... too bad resetting
> > _everything_ isn't possible without a good graphics abstraction layer -
> > which Linux is still missing :(
>
> No, it is not. GGI/EvStack is progressing quite nicely.
> Http://synergy.foo.net/~ggi.
>
Well I meant the mainstream kernel - yes ggi works for me partially but
that's just partially... anyway setting ^[c to not screw up with fonts
and alike with an escape or something is a Good Thing (tm) if (broken)
libs and progs use it to clear the screen...