[rfc] improving 32bit user performance/experience...

Brian Fahrlander wheeldweller at gmail.com
Wed May 20 03:06:28 BST 2009

Christopher Chan wrote:

>> I think we have opportunity; Gentoo users tackled this problem in
>> their way 10 years ago.
> I wonder how many had actual Pentiums and how many months it took to 
> compile.
    Personally? I think it ROCKS to take an "x86" install, carry the 
hard drive to several other machines (including the AMDs that I really 
love!) and have no problems booting.  Windows just can't touch that. It 
comes from inspecting the evironment on every boot, not just taking the 

    I tried a literal 586 install; I think it was of SCO Unix. It was 
kinda depressing that all my files were on a hard drive I couldn't boot 
on a 486. And while it might be measurably faster to compile all the 
files like Gentoo, I don't find it enough to matter. And I *do* have a 
problem with an end-user even knowing he's doing the compile.

    Linux permits MUCH diversity. I know I never, ever want to go back 
to the days of Slackware 2.3 and load 40 diskettes, and then get almost 
no instructions on creating this new thing called an "X server" which 
I'd never seen before. I've been unzipping or untarring files since 
1978; I'm gettin' a little tired of the command line.

    And yet on the other hand, there's *no* better place to teach a 
young man about embedding or serious system-design than to sit him down 
with DSL or LFS or other 'from scratch' methods.  None. When it comes to 
education, seeing it all come to fruition in new (or even old) hardware, 
it's an opportunity you can't replace.  It's just not something I want 
to do anymore, ya know?  That's why I crave Ubuntu, even if I pronouce 
it wrong. 


 Brian Fahrländer                 Christian, Conservative, and Technomad
 Evansville, IN                                                         
 ICQ: 5119262                         AOL/Yahoo/GoogleTalk: WheelDweller

More information about the sounder mailing list