My resignation

Kevin P. Fleming kpfleming at
Thu Jul 15 09:25:39 PDT 2004

Bruce Dubbs wrote:

> I don't agree.  There still would be too much overlap.  What differences 
> do you really see between your educationalist and your functionalist 
> approaches?  I see the following possibilities:

I've tried to stay out of this discussion as much as possible, but I 
think a very important point is being overlooked here.

If, and only if, the LFS project is primarily "educational", then with 
the release of LFS-6.0 I would think the book has achieved nearly all 
the educational value it could ever have. By that I mean:

- there is a respected, reliable, repeatable build method
- the system can rebuild itself from itself
- it's using the latest stable kernel series and toolchain
- the build instructions are clear, understandable and don't generate 
tons of support when people bother to actually read them :-)

My point here is this: what is going to happen in the general Linux 
community after the release of LFS-6.0 that would require changes to the 
book, other than minor package upgrades? Sure, there will be a 2.8 
kernel at some point, but if the book is "educational", there is no need 
to even consider putting that kernel into the book until some of the 
major distros are using it (same as is happening with 2.6).

Personally, I think if the book is primary educational, then it has 
reached a plateau and won't really have anywhere to go from here except 
for pure maintenance tasks. I'm not saying I disagree with that, but I'd 
  be disappointed if continued improvements that members of the LFS 
community want to see couldn't be incorporated because they didn't 
directly increase the educational value of the book.

If by saying that I'm saying that I think the LFS product (book or 
otherwise) should continue moving in the direction of being a base 
"distro" with a well-written guide to learning how to build the distro 
and how it works, then so be it.

More information about the lfs-dev mailing list