[patch] remove wireless support

Kay Sievers kay.sievers at vrfy.org
Wed Jun 2 08:23:02 PDT 2004

On Wed, Jun 02, 2004 at 10:17:13AM -0400, Robert Love wrote:
> On Wed, 2004-06-02 at 10:32 +0200, Owen Fraser-Green wrote:
> > No, the Linux system needs iwlib but what about all the other operating
> > systems that HAL could (and hopefully one day, will) run on?
> This argument does not fly: the same thinking can be applied to iwlib.

Oh, I interpret Owen, that HAL should abstract the whole library and
then the argument is valid :) So on every possible OS HAL runs on, all
wireless stuff can be set by HAL, on linux with libhal, the other OSs with
something different.

So HAL would be a "hardware library abstraction". I never felt comfortable
with the abstracting all possible attributes thing, they came up last week,
so I'm really happy to hear that we are back to the more clear delimited picture.

I don't think, that we are able to manage, to define a generic model for
any hardware class we want to support. It may be possible for simple
well defined devices like sound cards or serial lines. But we will not manage
it for the complex stuff. I think it's simply not possible to create the
needed models - look at the kernel device drivers, there is a very long experience
with hardware abstraction and a lot of drivers are very far from perfect.

Yes, it sounds very attractive to have such a thing, but please let's avoid
the "virtual hardware layer" - we will at least be grandfathers before we can
release it :)


hal mailing list
hal at freedesktop.org

More information about the Hal mailing list