From: David Waite <mass@ufl.edu>
  To  : ggi-develop@eskimo.com
  Date: Wed, 9 Sep 1998 16:02:50 -0400

RE: kgi vs. fbcon

> > I think that was one of the major reasons why GGI wasn't accepted in the
> > standard kernel yet: replacing too many subsystems at once
> doesn't work with
> > Linus.
>
> Rightly so, think 24 million new lines of code (a la NT5.0, or 2.1.43 -
> coulda got in 2.1.43 easy ;)

Definately. We simply were biting off more than we could chew at one time.
This works out really well though. We have a really good driver code model
(KGI, fixing to get better with a new version, right?), with a wrapper onto
a interface that is currently there.

As long as things are broken up into small chunks (GGI Console, EvStacks
stuff, etc) I'm positive that we can get them all in on their own merit-
plus we already know that the code can work, there are no oversights (oh
gee.. don't we need a way to do mulitple imput?).

-David Waite