Re: The Kernel

[Erik Walthinsen <omega@omegacs.net>]

> The more I think about it, the more I think that core sould NOT be
> debian packages because that enforces a certain filesystem layout and
> I think we might want to give the various distributions the freedom to
> put things where they will.
I agree that it doesn't have to be Debian packages, but I don't agree that 
the filesystem layout should be left up to the distrib.  The entire goal of 
building a Core is standardization.  If an application can't assume
libxyz.so is in /usr/lib (ok, that's ld.so's problem, but you know what I 
mean), then there is no point in creating a Core.  The FHS was written for 
a reason.

> That was my thinking.  I suppose we could put a lurk on the kernel-devel
> mailing list and if a patch is rejected for good reason (it breaks xyz) we
> do not incorporate it.  If it seems to be generally adopted into most of
> the kernel weenies' source trees and things go well, we can incorporate
> and test it ourselves.
Personally, I think that as long as we can pass decent QA runs with the 
kernel we put together, we should be willing to incorporate new features.  
In fact, this might be one of the ways SEUL pushes the envelope.  If we 
find that for instance a log-structured filesystem is better for end-user 
machines (reason: it survives ugly things like mounted power-off with 
almost *no* loss, *and* fsck's in fewer seconds than you have fingers even 
after that!), then we should seriously consider using it, if we can find 
one and "nurture" it.

As for the Core, a stock kernel is an absolute requirement.  We can't be 
requiring that other distribs support a kernel that isn't standard and we 
can't provide support for ourselves...

     Erik Walthinsen <omega@seul.org> - SEUL Project system architect
        __
       /  \                SEUL: Simple End-User Linux -
      |    | M E G A            Creating a Linux distribution
      _\  /_                         for the home or office user