Devel Digest, Vol 38, Issue 1

[Mitch Bradley]

Martin Langhoff wrote:
> On Thu, Apr 2, 2009 at 3:49 PM, Mitch Bradley <wmb at laptop.org> wrote:
>   
>> It's nice to say they should "see the light", but in my experience talking
>> to many such companies, the fact of the matter is that it is a hard nut for
>> them to swallow.
>>     
>
> I am under the impression that the relation between the OS vendor (and
> kernel/driver devs on the OS side) and the HW companies (with their hw
> and driver devs) is full of hard nuts to swallow. From reading horror
> stories around winhec, and ms bloggers making strained comments about
> driver developers, it doesn't sound like a bed of roses :-) But, for
> good or bad, management has gotten used to the dynamic w Redmond.
>
>   

That is absolutely correct.  Windows drivers are a total pain in the ass 
(speaking from my experience making Windows run on the XO).

But in compensation, once you get the driver right, or mostly so, it 
will work on hundreds of millions of machines for a few years.  By "it", 
I mean a specific binary that you can point to and say "that is the 
one".  When it stops working due to a new MS OS like Vista, you might be 
able to ignore the problem because the old product is several 
generations obsolete and you've already made your money on it.  "Sorry, 
you have to upgrade".

Having done the painful-but-necessary work to service the market that is 
big enough to return a profit, there is scant motivation to swallow 
additional nuts for a much smaller amorphous market that won't stay in 
focus.

One of Joel Spolsky's essays pointed out that to attract customers from 
an entrenched competitor, you must ruthlessly eliminate the many 
barriers that make it difficult to switch.  You must see those barriers 
from the customer's point of view; trying to impose your point of view 
on the customer just doesn't work.

I guess the main disconnect is that, for the FOSS community, the point 
of view is more important than the product.  The commercial world is 
just the opposite.

> I am interested in discussing
> the meatier parts of whether the commiditization that FOSS has applied
> to sw affects hw and how some time over a beer.

I would like that very much.  I sincerely hope that we have the 
opportunity to meet in person at some point.