From: Jon M. Taylor <taylorj@ecs.csus.edu>
  To  : ggi-develop@eskimo.com
  Date: Fri, 3 Jul 1998 14:58:21 -0700 (PDT)

Re: job offers

On Fri, 3 Jul 1998, Matthias Grimrath wrote:

> Sorry for the late reply.  I've been quite busy this week...
> 
> On Sun, 28 Jun 1998, Jon M. Taylor wrote:
> 
> > On Sun, 28 Jun 1998, Matthias Grimrath wrote:
> > 
> > > No.  Not because it is a technical problem, but rather a political.  
> > 
> > 	It certainly is a political problem, and this is a political
> > solution.  Politics is the art of compromise, and up until now the GGI 
> > project has not been willing to compromise *at all*.
> 
> I wouldn't say that.  libggi has always been changed to make it compile on
> other platforms.

	I am referring to the issue of KGI and Linux specifically. 
Multiplatform KGI wasn't much of a compromise, since it was the obvious
thing to do and pretty much everyone wanted it that way.

> > > Hard, emotionial words.  But that's how I feel.
> > 
> > 	I am sorry you feel that way.  I do not share your feelings.  I am
> > still going to port KGI sans input devices and consoles to Linux/fbcon.  I
> > will be porting all the Dali KGI drivers sometime soon, so your Mystique
> > driver will be working in the Linux kernel should you want to work on it.
> 
> It's been some days and I have cooled down.  I won't leave GGI development
> unless you (the GGI team (whoever that is)) tell me to do so, 

	Of course we would never do such a thing!  Everyone is entitled to
their opinion.

> I am still
> willing to support the Mystique and port it to the new fbcon wrapper.  

	Great!  Actually I should be able to port it for you faster, since
I have made the changes that make "porting" necessary and thus I can very
quickly do the necessary (small) modifications to the driver. 

> So
> I'm compromising here, leaving GGI was never an issue for me anyway.

	I have gotten burned out on GGI and tried to leave several times
since I started working on it in fall 1995.  It has never worked |->.  I
always get sucked back in because I cannot stop reading this list and
eventually I see a good idea that needs coding and I just can't help
myself....
 
> It's just frustrating to see how some people reject our idea of graphics in
> the kernel, then fbcon appears that is inferior than our design. (only
> unaccelerated graphics).  

	Well, it isn't *quite* that bad.  fbcon has a few extra supported
features beyond dumb linear framebuffers: rectangular blits, superbitmaps
(panning), mouse cursor (hardware or software), and splitline.  Granted,
the blitting is the only real "acceleration", but keep in mind that fbcon
is supposed to implement a console system primarily.  blits are all the
acceleration you can really use meaningfully on a text console.  fbcon
supports the notion of framebuffers, and with that and the ability to
still send acceleration ioctls through to the KGI subsystem we really have
everything we need now. 

> Kernel folx object EvStack, then I have to read
> about a sub mailing list on kernel discussing the same subjects.  It
> wouldn't surprise me if someday a mailing list opens working on a
> DirectInput equivalent.

	Yeah, that annoyed me as well.  Andy is subscribed to the mailing
list that was started for that project, so hopefully he can help them see
the light. 
 
Jon	

---
'Cloning and the reprogramming of DNA is the first serious step in 
becoming one with God.'
	- Scientist G. Richard Seed