David Kanter (dkanter@realworldtech.com) on 1/4/11 wrote:
---------------------------
>anon (anon@anon.com) on 1/4/11 wrote:
>---------------------------
>>Wes Felter (wesley@felter.org) on 1/3/11 wrote:
>>---------------------------
>>>David Kanter (dkanter@realworldtech.com) on 1/3/11 wrote:
>>>---------------------------
>>>>So I saw Charlie had a rant up on Sandy Bridge Linux support, and it seemed a little
>>>>odd to me. I'm not sure I see where the problem lies.
>>>
>>>Moronix has a more balanced take on it: http://www.phoronix.com/scan.php?page=news_item&px=ODk2OA
>>>
>>>>Based on reading what he wrote, it sounds like the issue is that driver release
>>>>schedules in the open-source community are a little less...co-ordinated...than in
>>>>the Windows world. Namely, that Intel cannot get ISVs (or FOSS projects) to incorporate
>>>>drivers before the hardware launch for press reviewers.
>>>
>>>Pretty close. The problem is that drivers must flow from hardware vendors through
>>>upstream and then through distros to reach users, and these gatekeepers have fixed
>>>release schedules and long latency. Thus if you release drivers in January, (if
>>>you're lucky) they reach users in April. If you release drivers in February they
>>>probably miss the .04 freeze and reach users in October. Thus to ensure that users
>>>have drivers when the hardware is released, you need to send the drivers to Linus
>>>6-9 months before the hardware ships. This is hard, and heaven forbid that you
>>want to keep anything secret pre-launch.
>>
>>Totally untrue. Distro kernels commonly and frequently add >driver updates on their
>>own schedule.
>
>Which are utterly out of Intel's hands.

I was debunking the first incorrect part of his claim.

>>But Intel wouldn't even need to have drivers >pushed through the distro:
>>the equivalent to Intel providing reviewers with Windows >drivers would be Intel
>>providing them with Linux distro drivers. HW vendors very >frequently publish Linux drivers on their own sites.
>
>I think you're missing the point. The drivers were available - in source form.

I'm not missing that, it was specifically mentioned and discarded as an option in the review. Whether or not you agree it is a reasonable option is another story. What I am saying is that a Linux distro has no technical disadvantage here regarding distributing drivers (unless you count logistical problem of smaller market share).