[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

SEUL: Re: What's the diff to SEUL ?



> That's right and I see the need for it. I only want the prove if it's realy
> impossible to reach that goal with an existing distribution. You have
> to invest very much time to package the software to get the dependecies
> and so on, which is done in several distributions. It's best done in 
> Debian's distribution, because their dependecies possibilities of their
> package manager. The largest lack there is their horrible installation 
> program.  So I think you can save much time in developing an perfect 
> installation program. The new distributions of SEUL and e-Linux would consist
> of the complete debian package and two versions of the perfect installation
> program. And advertising is done by Debian.

I will be writing this up formally when I get the chance (argh, too much to 
do!!!), but here the basic idea, with e-Linux added in:

SEUL (w/e-Linux help maybe) will take the existing Debian distrib (bo) and 
strip it down to a "basic" distribution, i.e. kernel, sysvinit, *utils, XFree 
libs, etc.  Such a system will boot, but not do much interesting stuff at 
all.  This then would become the minimal "core" distribution.

Once this exists, SEUL and e-Linux would work on deciding on some fundamental 
packages, then start building up (using existing Debian packages or 
modifications thereof *whenever* *possible*) to a more complete distribution. 
At a certain point, e-Linux and SEUL will start going their own ways, given 
that e-Linux intends to use KDE (I still don't understand why).

Eventually we may see this "core" become the basis for other distributions, 
possibly even Debian itself.  Once this happens, what system a package is 
built on is irrelevant, because it will work on *all* machine with the same 
version of the core.

As far as installation and packager frontends, I think Debian, e-Linux, SEUL, 
and any others should work *very* closely together.  There is no point in 
developing many different installers in say 4mo when we can take what exists, 
combine our efforts, and develop the "end-all" of installers that works for 
*all* the distributions, and do it in say 3mo.

There are *many* pieces of all three distributions that can and *should* be 
done in collaboration.  Any other method would be ludicrous.  Put a few more 
people on the team, possibly do a partial redesign if something already 
exists, and you can develop a package configurable enough for everyone to use 
it.

I want cooperate with existing projects as much as possible, ideally with the 
effect of enhancing both.  That can't be done, though, unless all parties are 
willing to rethink their project some if necessary.  This is why SEUL will be 
based on Debian instead of RedHat, as was originally decided.  RedHat does 
not offer the ability to coordinate and enhance both parties' projects.  
Debian does.

     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