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

Re: Some unanswered questions about SEUL (plz ack!)

> No one ever debated me when I mentioned KDE/Qt.  I know Eric and Peter
> don't like the Qt license.  I don't love it, but I maintain my position
> that we should pick the BEST software for end users over a GPL
> "bigotry".
I don't know if I'd call it "GPL bigotry", and I don't rule out all non-GPL'd 
stuff, but from what you say in two paragraphs down the license as it stands 
may be too much of a hurdle.  If you have to spend $2000 to sell programs 
based on it, that can seriously limit small companies with good software 
starting up.  If SEUL uses Qt, and they don't have the startup capital to do 
it, they don't get the "SEUL stamp of approval" or whatever, and they can't 
do it.

> But if we're not concerned about commercial apps,
But we are.  SEUL will not gain any market share over what Linux already has 
unless a level of commercial programs is available to rival what's available 
for Windoze.  That's just how the market works.

> Someone mentioned here a while ago that we need to rethink our goals
> from the ground up.  Are we *sure* RedHat hasn't just done our work for
> us with 5.0?  The reviews make it sound pretty easy to install and use. 
I'll find out Monday, if I can get into redhat.com...  I plan to download 
as much as I can starting midnight if possible, and burn a CD at work around 
4:30pm.  I have a spare partition on my laptop, so I can play around with it 
(though I have to use PCMCIA and NFS because my CD-ROM drive doesn't read 
CD-R's... :(

> Perhaps we should forget about making a distribution alltogether!  Just
> work together to write programs and encourage others to do the same.  Fix
> bugs in other software (InfoWorld complained about a bug in their FDISK
> front end).  Write better configuration managers.  Write a help system and
> propose that it be the standard for all Linux apps.
> Once we do all that, we can hope that other Linux distribs incorporate
> our work.
Believe it or not, that's precisely what I'd prefer.  The vast majority of 
what we do will be exactly what you say above.  However, as I mentioned in 
the original message draft to seul-project, once we get around to doing 
everything that needs doing to make Linux a better place for end users, we 
will have touched everything there is to touch.  At that point, we can go one 
of two ways: 1) leave it at that, and let others take pieces of our work as 
they see fit, or 2) actually create a distribution.

The problem with 1 is what luka and I have discussed a lot, and that is the 
piecemeal aspect of it.  I will guarantee you that no distribution out there 
will want to take everything we have to offer.  If you think about it, if 
they did that, why would they have their own distribution?  They're going to 
take parts of it.  And when they do that, they don't take all 
interoperability we will have *designed into* all our work, thus they do more 
work to cap the loose ends, etc.  What we end up with then is one part of our 
work in distrib A, another in distrib B, part of a third in distrib D, and 
Linux fractures even more.

By the time we've done everything, and if we do it right (in CVS with RPM and 
autoconf/automake build structures), we will be able to form a distribution 
by compiling all the RPMs, putting our installer on a floppy, and shipping 
it.  Now, granted, we may not do that, but that's why I mentioned the
possibility of RedHat doing so for us...

> > The classic distrib for hackers, the SEUL distrib for the "masses".
> Why do they have to be seperate?
Well, they don't have to be.  If RedHat were to take us under their wings, 
however, they wouldn't want to replace RedHat itself blindly with our stuff.  
So the answer, of course, is for RedHat to 'become' SEUL, as we work they 
merge stuff into RedHat in an unobtrusive manner, then do a 6.0 release that 
is 100% SEUL.  But then either they lose control over their own product, or 
we lose control over our non-commercial plans.  Granted, it'll be 'free', but 
once RedHat Inc. has say over what we do, we lose much of the control we 
have now.  For instance, what if we want to include a really cool word 
processor in SEUL, but RedHat is in control?  RedHat sells Applixware, RedHat 
doesn't like shipping a free product themselves that competes with a 
commercial product they already sell.  Ugly.

So it might not even make sense for RedHat to take SEUL under its wings at 
all.  Lots of stuff to think about along those lines for the future, though.  
Of course, I have no objections to RedHat taking parts of SEUL and eventually 
being [almost] identical, and even doing most of the commercial parts of the 
process, but *only* if we retain full 'editorial' control over what we do.

> I do have big ideas on the application front though, and I want to see
> Linux excel in this area.  It will, in time.  I'm sure of that!  :-)
Good.  We're all on the same train in that respect, I hope... ;-)

One thing to mention, though, is that we (as leaders) have to be extremely 
careful about how we present our visions and goals to the 'general project 
populace', a term I should be shot for using as it implies an inferiority on 
'their' part...  But anyway..., I have found that in many cases relatively 
radical ideas (which I think most of ours will be, relative to the <I>status 
quo</I>) must be introduced slowly, even to the 'radicals' themselves
(seul-project members).  If coredumped in full force, they are quickly picked
apart and the whole vision disappears in the details.  I haven't quite yet
figured out how to best present stuff of this scale, and would welcome any
ideas.  Maybe we should start a new thread on it in here (and keep it in


      Erik Walthinsen - SEUL Project infrastructure/system architecture
       /  \           omega@sequent.com         Work: (503)578-5314
      |    | M E G A  omega@omegacs.net         Home: (503)281-4281
      _\  /_          psu12113@odin.cc.pdx.edu  Majoring in CS

       SEUL: Simple End-User Linux - creating a Linux distribution
     http://www.seul.org/            for the average home/office user

SEUL-Leaders list, seul-leaders-request@seul.org