Greetings fellow Cobaltistas!
Chapter 4 - "A new "quieter" hard drive" 12GB --> 80GB (7200 RPM IBM 
Deskstar!!)

It's getting easier, no doubt about that :) It took 75 minutes to 
install NetBSD on the new large disk  though, and I attribute this to 
the fact that the disk had another Linux OS on it and needed to be wiped 
before NetBSD could be installed. The new drive *is* much quieter and 
also has a larger cache than the circa 1998 12GB Quantum Fireball that 
was in there, so reads and writes are significantly quicker than was 
formerly the case. I'm beginning to look at this as the "final 
installation" and therefore taking time to set things up in a more 
permanent way now, since I don't think I have any need to mess with the 
hardware in the near future. (I hope not anyway!!) Memory is maxed out 
to 256MB, and this drive is 'young and strong' so this is the 
configuration going forward.

Observation: I'm continually surprised at how long it takes to build and 
install stuff. I'd honestly expected slightly faster builds, but my 
guess is that the little MIPS processor just doesn't have the horsepower 
to compile stuff faster. As an example, bash took 2 hours to install, 
and various pkg_tools an hour or so apiece. This really was an 
eyeopener. I didn't expect this little guy to behave like a stacked P3 
or P4 of course, but I didn't think it would be this slow at compiling. 
Oh well.

One thing stands out in my mind, and this is the big one for me. I do 
realize I need to slim down the kernel and optimize it some, yet I've 
never done any kernel tweaking....ever, so I'm slightly intimidated by 
the prospect. My goal is *the* tiniest fastest kernel possible, 
consistent with the functions of the box, namely serving mail and an 
occasional web page. If anyone has the time I could *really* use some 
help in understanding what I can get rid of in the kernel and a bit of a 
n00b walk through of how to accomplish this. I've done some reading of 
cobalt/netbsd pages - as well as the kernel compiling chapter on 
NetBSD.org itself. Yet there is - of course - an underlying assumption 
that the reader may have done some kernel recompiling before, whereas 
yours truly has always studiously avoided such wizardry - until now when 
I realize it is absolutely inevitable no matter what.

Regards & TIA one 'n all,
-Colin