[lkml]   [1998]   [Sep]   [19]   [last100]   RSS Feed
Views: [wrap][no wrap]   [headers]  [forward] 
Messages in this thread
SubjectRe: Linux, UDI and SCO.
On Sat, 19 Sep 1998, Stephen Williams wrote:

> said:
> > Unfortunatelly this will means that manufacturers could just write UDI
> > driver (buggy -- "Unix is mandatory; we do not have enough resources
> > to make deep testing of drivers for mandatory OS") instead of
> > publishing specs ("You does not need our specs: you have our UDI
> > driver").
> Linux users are an increasing purchasing block. A good binary UDI interface
> will expose more vendors and customers to the True Way. If a vendor makes
> a mess of it, we can either do it ourselves (no worse off then we are now)
> or use the increasingly viable options of spending our money elsewhere.
> A clueless organization will remain so in spite of our best efforts, so we
> might as well be helpful to those who wish to participate and not get all
> huffy about the others.

This is a very odd argument. If a vendor makes a mess of it, and
there are no specs, _who_ is going to do it themselves, and when?

You seem to be saying you don't care about the significant problems,
anything is okay if vendors provide Linux support for their devices.
You don't seem to care enough to be the one who will write a
replacement for a buggy driver.

Yes, we can spend out money elsewhere. Some devices will be lousy
with Linux because their close-source drivers are lousy; some will
be good because their closed-source drivers drivers are good, and
that is where we will spend. Thus, on the whole, the drivers will
be good.

They will be better than the drivers on competing closed-source OSs
simply because the vendor's programmers will know Linux better.
That is something, certainly, but is it the key issue?

Look 5-10 years down the track. None of the current devices will be
in use. None of Becker's network cards, and none of the other
devices for which other very able engineers in the Linux camp have
written drivers based on accessible specs.

There will be UDI, there will be NO accessible hardware specs, the
drivers will be 100% closed commercial. Only a few will remember
the time when you could read the driver source and spot for yourself
where the bugs were and resolve undocumented issues. You will wait
months for bug fixes, and then find, in the usual Microsoft style,
that the problem is only fixed in the driver for the next version of
the hardware.

Someone will have the idea of a mailing list where drivers can be
discussed and the writers of the driver software post patches, but
that will be only in your dreams. It doesn't happen in the full-on
commercial environment. It is a different world. A few
professional engineers might be admitted to such a privilege, but
that is all.

Then we will have the fraternal group of Sun, Microsoft, IBM and a
few others upping the membership fees of the "UDI Consortium" into
the multi-million-dollar bracket, and holding conferences among
themselves and the hardware vendors on the means of maximizing the
commercial payoff of UDI.

It would take a better strategic thinker than I am to advise the
important members of the Linux community on this, but please let us
take it seriously, rather than thinking only of the delightful
prospect of more hardware vendor support of Linux, and rather than
thinking of how nice it will be for me personally to buy and use the
x device I have been dreaming about on Linux.

IMHO it is enthusiatic fantasy to imagine that exposing the masses
to the "True Way", though suddenly finding that there is wide
hardware support for Linux, is a pre-emptive advantage of UDI.

Rgds, mtw

To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to
Please read the FAQ at

 \ /
  Last update: 2005-03-22 13:44    [W:0.132 / U:1.828 seconds]
©2003-2018 Jasper Spaans|hosted at Digital Ocean and TransIP|Read the blog|Advertise on this site