[no subject]

While doing that, they were also trying to add in all the "new" stuff--
Java, browsers, etc.  And trying to tackle MS on its own ground--the
desktop.  Unix ceded the desktop years ago--not because it's technically
inferior there, but because it never had the unified champion that could
push one, "true" desktop that all users could learn and recognize.  (Oh,
there are a lot of other reasons, but I feel that one made the biggest
difference.)

Unix/Linux evolved with communications.  Its paradigm leans toward lean,
specialized utilities and services that demonstrably do very well in
demanding applications.  But GUI was a late addition to the Unix model,
and one not necessarily seen as critical by the core developers for quite
some time.

Windows evolved with a different paradigm--the desktop was its venue,
and an integrated GUI was intended from the start.  Furthermore, MS
sees the world in terms of its own OS, with only grudging concession to
interoperability with other systems--it really, really believes that it
has the True Model and we'd all be better off if we dropped all these
annoying other systems.  (Don't agree?  Look at DDNS in W2000.)

NT--a late-comer, less than 6 years old in the wild--derives from that
same mindset, but tries to do duty as a server.  Badly, so far.

Will NT mature?  Certainly--enough money and PR can push anything
through.  (Hell, Windows 3.0 would never have come to light if M$
hadn't been pushing so long and so hard.)  But *it's* playing catch-up
in that arena.

Perhaps the issue here is diversity--the right tool for the right job.
The GUI paradigm is NOT the best for all environments; the MS model is
NOT the solution for all tasks.  And neither is the Unix/Linux model.

In that respect, something that's worried me for the last 15 years is
that I simply haven't seen anything new in operating systems.
Particularly with the dampening effect of MS pushing Windows-only to
the Corporate world, we've not seen any exciting or radical new OS come
from either academia or corporate research.  Plan 9 has languished at
the Labs for almost 20 years, for instance.  There's *got* to be a
better way to do some of the things we do today.  (For that matter, I
have to believe there's a more efficient way to push bytes than
Ethernet, itself over a quarter-century old...)

The web is a verifiable new paradigm, built on the old blocks.  Isn't
it time for new tools?  Linux is playing catchup on the desktop--and
leading in the server/comms world.  But more importantly, being open
source, it has the potential of leading in other areas--by being capable
of changing.  In the end, we may not recognize it; but that's not bad.

My $0.35 (more than $0.02, at this length...)
-- 
	Dave Ihnat
	ignatz dminet com