On Wed, 1 Oct 2008 14:36:26 -0800
Henrik Hudson <[EMAIL PROTECTED]> wrote:

>On Wednesday 01 October 2008, Jerry <[EMAIL PROTECTED]> sent a missive
>stating: 
>> On Wed, 1 Oct 2008 23:25:19 +0200 (CEST)
>>
>> Wojciech Puchar <[EMAIL PROTECTED]> wrote:
>> >> In all likelihood, the probability of any vendor creating FBSD
>> >> specific drivers is directly proportionate to the expenditure of
>> >> funds to create and maintain the driver versus the expected
>> >> revenue from such an expenditure.
>> >
>> >giving out a specs will be the simplest way.
>>
>> Any entity, or corporation, has a right to expect a return on their
>> investment. To expect a corporation to simply give away something,
>> thereby depriving their shareholders, partners or whatever, of their
>> rightfully expected monetary reward is foolish. It certainly is not a
>> well thought out  business model.
>
>There is a difference between open sourcing the binary blob and
>possibly giving away optimizations, trade secrets, etc... and allowing
>easier access to either hardware register specs or specs to write a
>wrapper around a universal blob.
>
>Personally, I think they see it as a "quality" control issue, though
>the quality of their own code is sometimes circumspect. The card
>companies sell hardware and this is where their money is made and/or a
>better experience with the software drivers. Open the hardware spec,
>add a support clause that any "open source" drivers aren't officially
>supported and you're good to go.

Unfortunately, that is not a legally binding disclaimer in many
locales. It is equivalent to wearing a T-Shirt with "Touch me and I'll
kill you" embroidered on it. If someone touches you and you kill them
you cannot then claim that they were fore warned. That is from an
actual lecture I receive in a business class.

>Opening the hardware spec will do nothing except sell more hardware.

Unproven and if it were in factually correct, more businesses would
make use of it.

>a) the average joe will continue to buy systems with the supported
>hardware / drivers, most likely Windows, OS X or a major Linux distro.
>Probably wouldn't even know the open source ones exist.

A bit naive in my personal opinion. The overwhelming majority of users
that I am acquainted with and many of them are college students, buy
what ever works best for them with the least amount of user
intervention. The  majority of users that I am familiar with do not
want to spent valuable time attempting to configure a PC; they have
better things to do. How many discussions have we had on this forum
regarding the lack of a functional 'Flash' plug-in for web browsing,
etc. Until problems like that are solved, expanding the base for FBSD
is a touch uphill climb.

>b) the geeks of the world will start running the open source driver if
>it's better or not if it's worse for their applications. Either way,
>it will only sell more hardware.

Again, unproven.

>c) The FOSS only crowd will start using the hardware since it has a
>fully open source drivers.
>
>The open source driver doesn't need to be able to run Doom5 at
>incredible speeds, it just needs high quality 2d and the ability to
>handle some 3d compositing, etc... for desktop effects.

Sorry; however, I totally disagree. A half-ass driver is akin to being
slightly pregnant. You are either pregnant or you are not. A driver is
either fully functional, or it is broken. To make arbitrary limits on
what it should and should not be able to do is absurd. If I want to
play a game, and your half-ass driver will not run it, then that driver
is as useless as 'Tits on a bull'. It either works, or it doesn't. Too
many people today accept inferior products/performance under just the
guise you are employing.

>My .02$
>
>Henrik

Again, it appears that NVIDIA has attempted to work with the FBSD
community. Evidently, nothing has come of it. Unfortunate, to say the
least.


-- 
Jerry
[EMAIL PROTECTED]

Not Hercules could have knock'd out his brains, for he had none.

        Shakespeare

Attachment: signature.asc
Description: PGP signature

Reply via email to