Richard L. Hamilton wrote:
For some more cold water (not that it should stop anyone that
believes in the idea enough to do the work!), remember UDI?
Doesn't seem to have amounted to much - I haven't heard a thing
about it in 3 or 4 years, nor do I see evidence that it resulted
in a bunch of compliant drivers being produced.
(UDI = Uniform Driver Interface, still in the kernel though)
There's a couple of sites, but they seem to be rather inactive:
projectudi.sourceforge.net
www.project.udi.org
I suspect that with the climate created by all the SCO FUD, the
utter lack of interest by the Linux crowd (who probably thought
that if they adopted it, they'd end up giving more than they got),
probably killed it.
Here's an old thread which points to possible reasons why UDI didn't get
much traction:
http://www.ussg.iu.edu/hypermail/linux/kernel/9809.2/1045.html
Solaris and Linux are the only real survivors from the proof of concept
OSs targeted. Linus seemed interested in the concept and Intel was an
early advocate. But even in 1998 FUD surrounding GPL pollution
prevented what was technically a very good idea from getting a
foothold. The fact that SCO was an early backer probably hurt in the
long run. I love this quote from this 1998 linux thread:
There are basically only two ways this UDI scenario can work.
1. All the commercial backers of it, switch to using Linux for their OS
and they just build hardware, or in SCO's case additional software
add-ons.
2. SCO releases all UNIX source code under the GNU GPL,
HP releases all sources code of HP-UX,
/Sun releases all source code for SunOS and Solaris.
You might as well ask Apple to release all the source code to
NextStep/OpenStep/Rhapsody./
That will never happen right? ;-) So, now that Sun have released the
source code for Solaris, what is holding us back? I don't think we have
to worry about HP-UX and if SCO is a problem for Linux, it's a problem
regardless of whether UDI becomes a common standard.
I think a collection of notes by people who have ported e.g. *BSD
drivers to Solaris would probably do as much good as anything,
although perhaps if enough Linux, *BSD, Solaris, and other
open-source OS users/coders/supporters made a point of choosing
hardware for which full specs (sufficient to write drivers) were
available, that might result in an improvement in the common problem.
It's obviously a good idea, but I don't think it will be sufficient
until OpenSolaris exceeds Linux's market share and we start getting HW
manufacturers interested in whether they are OpenSolaris ready.
Perhaps a new and neutral site, that listed devices and chipsets that
had (or did not have) sufficiently open specs, with an underlying
database of product vs chipset (because it has to be easy for a
potential buyer to determine what to avoid) might apply a little
pressure...
I like this idea. Such a database would take some work, but probably
not much more work than a hardware compatibility database. In fact,
just add a column to the existing HCI database. Linux offers
incentives to get HW manufacturers to open up. For instance, they are
currently offering to create device drivers for free, if HW
manufacturers open their specs:
http://www.linux-watch.com/news/NS8802144045.html.
If we had a similar program targetting UDI or a similar virtualized
hardware interface spec which would work within OpenSolaris, it could
give linux (and even OSX?) advocates an incentive to play nice again.
If we could get OSX, Linux and OpenSolaris to share the same drivers, we
stand a much better chance against the company which offers cash and 90%
market share as incentive to HW manufacturers.
_______________________________________________
opensolaris-discuss mailing list
opensolaris-discuss@opensolaris.org
