> I've yet to try the new restore CD (I've been running 2.0 for a while
> now), but I agree, Alex should be applauded. Thanks!

Yes he does. I wanted to do it for the longest time, but that doesn't =
count!

Time is missing and wanted to do thing is just talks, doing is better! =
(:>

> As for the kernel, the point I was trying to get at is that since we
> have a bootloader on the linux partition, there's no need to be too
> worried about kernel size as far as GENERIC. It might be nice to
> mention that if one chooses to load the kernel directly via the
> firmware (ie the real kernel named vmlinux.gz rather than the
> bootloader named vmlinux.gz), then this won't work. And it would be
> nice to submit a known working kernel config to the source tree for
> this reason.
>=20
> But otherwise, GENERIC supports a great range of hardware in other
> archs, why not this one? The Qube has a PCI slot, might as well take
> advantage right? It's rather easier to trim the kernel down later
> after the hardware setup is solidified than trying to add it in not
> knowing exactly what to use, this is what adjustkernel is for.

Still the box is not super powerful. Not that it would make a big =
difference
anyway, but I guess it's always the choice of who make the kernel to =
start
with. I think Alex did a great job. I know if I would have done it, I =
would
have been even more restrictive has he did. I am more of the school, =
that
less is better. Don't run anything you don't need. Plus I do run GENERIC =
of
OpenBSD without never changing the kernel, but that's a different story =
all
together. On the Cobalt, I think supporting the box hardware as purchase =
is
the way to go and anything else should be add on. But hey, we are free =
to
see it differently for sure! (:>

In the end, who ever do the work have the luxury to select what he wants =
to
put in it...

Daniel