> Hmm, let me see...
> They want our drivers, to "establish growth for the UNIX community" (i.e.
> for CSO, Solaris, etc).
I have never seen anywhere where they say they want our drivers.
They want us to be able to use drivers written for a standard
interface, which would not preclude us from using native drivers.
> What's in for us? More work?
No. The message stated SCO were offering to help us develop the
ability to use UDI drivers. Maybe a bit of work to make sure UDI drivers
work well, but less work to support obscure hardware which nobody really
wants to write a native driver for since it is little-used.
> by the average Linux user. Or is there a major hole in the Linux driver
> availability
> chart I missed?
The major hole is the time to develop new drivers. As well as some other
items (eg: PCMCIA notebook floppy drives, future USB devices, ...).
> Will this initiative really make hardware vendors think twice, and force
> them to release
> hardware specs and driver sources which they withheld before, just because
> the Linux
> community is cheaper in writing drivers for them?
The idea is that the vendor writes a UDI driver. This is the official
driver. If the UDI driver is too slow, the Linux community may write
a native driver or an improved UDI driver.
> Having an UDI is a nice idea. But they (the commercials) should work our
> way, and not
> make us work their way. They want something from us, and have next to
> nothing in return.
They want to help us code something, or code it for us. They offer us
some hope for drivers released at the same time as new hardware in return.
This sounds mostly good to me. The main negative is that UDI may cause some
places to not release hardware specs, since we can use the UDI driver, and
the "I can't use it on Linux" argument wouldn't hold anymore. Then we may
be stuck with potentially slow, inefficient drivers. And of course the
threat of a proliferation of binary-only drivers, which is bad in the whole
GPL ideology and bad in that it's unknown code, possibly buggy, no chance
to review it, etc.
David.
-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@vger.rutgers.edu
Please read the FAQ at http://www.tux.org/lkml/