Re: [Linux-fbdev-devel] Redesign of kernel graphics interface

[James S. <jsi...@in...>]

> > Another topic that I missed was, why did kgi fail and what can we do to avoid
> > repeating the same path this time around.
> 
> IIRC, the main reasons were:
>   - GGI wanted to do too much at once and was too intruisive (conclusion:
>     always advance in small steps, not big leaps):
>       o kernel graphics drivers (KGI)
>       o new input subsystem (similar to what we're heading to now)
>       o user space library (libggi, AFAIK the only part that's still alive)
>   - fbdev was better in multi-platform handling (m68k -> PPC -> ia32 -> SPARC
>     -> alpha -> ...)
> 
> Please correct me if I'm wrong ;-)

The KGI is way to complex. I attempted to write a driver using that system 
but I was totally lost. Fbdev, especially now is much easier to develope.
The input system we have is very similiar to what they had. I'm really 
glad we have a input system. As for the library. Well it has always been 
a dream to have a universal library by many people. It never happened and 
never will happen. When it comes it libraries you have issues like 
licenses and people fighting