Well it turns out the issue of fair scheduling for a better desktop experience has already been integrated into the kernel. A possible reason why Con stopped kernel development is that Linux code maintainers decided to use the code from another developer.

Jukka Larja (roskakori2006@gmail.com) on 7/27/07 wrote:
---------------------------
>Foo_ (foo@nomail.com) on 7/26/07 wrote:
>---------------------------
>> The hardware may be the same, but the use cases are not
>> and thus the kind of "support" needed is not the same.
>> "Supporting" SMP in the sense of improving database
>> workload throughput is not the same as "supporting"
>> SMP in the sense of improving desktop latency in the
>> face of e.g. interactive applications running side-to-side
>> with CPU-heavy background tasks. They are different
>> problems with possibly different optimal solutions.
>
>Sure, but at least there is SMP support (or "support" whatever you mean by that)
>for desktop (I don't know if that is a very good example. Perhaps SMP support in
>general is so trivial thing to add that the fact that servers needed it first didn't
>have any effect on desktop). Perhaps there are some other similar technologies.
>
>-JLarja