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

Re: My doubts about Debian choice



> X-Priority: 3 (Normal)
> Content-Type: text/plain; charset=iso-8859-1
> X-URL: http://thunder.prohosting.com/~oehansen/
> Date: Mon, 09 Feb 1998 21:58:19 +0100 (CET)
> Reply-To: oe.hansen@isholf.is
> Organization: Private Home
> Sender: oe.hansen@oehansen.pp.se
> From: "Orn E. Hansen" <oe.hansen@oehansen.pp.se>
> Cc: seul-leaders@seul.org, body.unsubscribe.seul-leaders@oehansen.pp.se,
>         mail.majordomo@seul.org, SolariS@cicei.ulpgc.es
> X-UIDL: 1a281f6740d840ce4955d7b4b83b66f1
> 
> 
> Þann 08-Feb-98 skrifar jfm2@club-internet.fr:
> > 
> > Whan I played competition chess it happened me not a few times than I
> > had a superb strategic plan only than I was crushed a lot before
> > applying it thaks to brilliant tactic moves of my opponnent.  That is
> > one of the first lessons you learn: it is the coupling of strategy and
> > tactics who differentiates a grand master from a normal man.
> > 
>   It has very little to do with strategy... that's usually the arrogant
> explanation, you'll hear from a self-important University Professor who feels he
> would have given a much better relativity theory, than Einstein.
> 

Chess teachs an essential thing: the value of time.  A world champion
is the guy who is the fastest at deploying his pieces and the better at
making his opponent pieces reach their position in two moves instead
of one.  And a perfect long term plan is useless if the short term
opponent threats crush you at move 18.



An now I will explain why I want SEUL _fast_.  The rest (my resistance
to Debian) is only a consequence than I believe Debian is not the
shortest path.


Much things can happen: Linux can lose its momentum due to Linuxers
becoming old, losing illusions and so on.  More importantl at present
times Linux is too weak to survive if the commercial Unixes are killed
by NT.  Without commercial Unixes Linux would not get any software
because the market would be seen as too small if developping only for
Linux.  Also that would mean no job propositions for Unix so no Unix
teaching in universities, few if any people programming for
Unix-Linux: the new free software developped would be for NT.  I let
you imagine what would be the quality of software once Microsoft would
have destroyed all competition.

Is that a possibility?  Yes.  The Unix label is in the hands of SCO
not very sytrong hands.  Digital has been sold to Compaq.  IBM relies
on mainframes and AS400 for its income: AIX is not a central product
for them.  Only SUN is adamant in supporting UNIX.

I want Linux strong enough to survive alone, the faster the best.  One
of the factors limiting its expansion is the technical barreer: people
without technical knowdlege cannot use it and end up using Microsoft
products (and reinforcing it).


Now I pose an hypothesis: at presnt times Debian is less user friendly
than RedHat.  Please do not start telling me than Debian is better or
more integrated, I concede on this, I am speaking about user
friendliness.  Accept my hypothesis and follow me.


Now suppose we start from Debian.  What heppens?  First we cannot
release SEUL even in Alpha until SEUL has caught up with RedHat in
user friendliness.  It would make _very_ bad propaganda if we sticked
a SIMPLE End User Linux in a distribution where the user has only the
XFree tools for configuring X or where he has to recompile the kernel
specially when the users of the most widely used distribution get
kernels where kernel compiling is more a proof of hackerhood than
aything else: the original RH kernel is both lean, able to cope in any
networking role and with sound support out of the box.  Please don't
tell than we can solve it.  This is true but we will lose _time_ doing
it and until we have solved all these problems we cannot realease.
And what happens in the interim?  Until then SEUL is only vaporware.
We will not get help from people using SEUL, we will not be very
credible so Debian will send few help, the only developpers we will
have is the people in this mail list but I suspect than there are much
more onlookers than there will be active participants and than not all
participants will be programmers (note this point: if you choose a
road where you have to design tools then you need many C or perl
programmers)

Now about the Debain help: thay are 200 but look at what Aldo (a
Debain partisan) said: they are 20 who make all of the work.  I will
rtell you what are the 180 others: suppose a Debian mathematician, he
uses Pari (our mathematician is a number theorist).  Like all people
who are used to packaging systems he hates having unpackaged software
on his box, so he builds a DPKG and sends it to Debian.  The 180
people are people packaging the software they need for their work.
That is the reason Debain has so few end user software: Debian people
have no use for it.  Then there the twenty than Aldo speaked about.
These people want the spreading of Linux.  They will come to help but
thay have already their hands full of trouble with Debian.  Some of
them are already sleeping 5 hour nights due to their work on Debian.
They are unable to work much more.  In addition note than Bruce Prens
did not commit himself: "we will decide if you desrve be helped and
perhaps we could a better job by ourselves".  Not the exact words but
that was the meaning.  As I have said we will not get help until we
are credible, to be credible we need to have something released and we
can't release until our product is easier than Redhat.



Now suppose we start from RedHat.  We can remove some of those idiotic
and arrogant brain damages in RedHat than I pointed yesterday in my
exam.  Some of them are pretty easy and do not need much programming
skill.  We can add some packages RedHat is missing (and we can get
prebuilt ones in the Caldera or Suse distribution than will need only
minimal tweakings ).  At this point (one or two months work) we will be
already ahead of RedHat in the matter of user friendliness.  (Notice
than we get at this point pratically without writing a single line of
C).  We are probably still behind Caldera or SUSE but we can anoounce
a pre-alpha version indicating some future developments.  And then we
begin to have SEUL _users_ the sincerest species of collaborators
because if we do it well they will begin wanting to improve their SEUL
(they will have an interest in it) abit like happenned with
Linux-kernel.  And this how we get our two hundred people working in
SEUL.  In addition these people will be people who will have some
understaing and empathy about the poor Joe User fighting his distrib.

The starting hypothesis was when I said RH is easier to use than
Debian.  If you disagree write a review: look at your distribution
with disapassionate eyes, Joe User's eyes not hacker's eyes.  Ask
yourself what would be an obstacle or completely ininteresting to your
sister.  Then send your conclusions.  Last June I began reviewing
every distribution I had.  I only had an old Debian so I asked than a
Debianer did the proper.  No answers.  They kept hyping about the 200
people and about Debian being better than sliced bread but never told
why and specially not why Debian would have been good for their
sisters.

The other point is than I want SEUL and Linux world domination fast!

-- 
			Jean Francois Martinez

==================== The Linux.  Use the Linux, Luke! =======================

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