embedded devices and open source

[Jeffry Smith]

On 2/24/07, Ben Scott <dragonhawk at gmail.com> wrote:
> On 2/23/07, Jeffry Smith <jsmith at alum.mit.edu> wrote:
> >>  A common concern is support -- support that the manufactuer gives to
> >> people who buy their stuff.
> >
> > Of course, another concern is THEIR support.  Remember most embedded
> > folks don't write their own OS.
>
>  Right.  And eight years ago, that might have been a valid objection
> to Linux.  But these days, there are whole companies that do nothing
> but Linux support, and lots of it.  Plus there's support from others
> vendors.  Lots of people want to run Linux in the embedded space, so
> hardware vendors (chip, board, systems, etc.) offer varying levels of
> support, from fair to excellent, depending.  So I consider the "Can we
> ('we' being the integrated product manufacturer) get support?"
> question mostly answered these days.
>
> > However, from what I understand from embedded folks I've talked to, Wind River
> > & others don't give such great support.
>
>  s/Wind River & others/the whole damn IT industry/
>
> > And some don't understand what their product is.  ... Ask the
> > company how many drivers they sell.  No hardware, just the
> > drivers.  I'll lay you odds it's zero.
>
>  Of course, hardware without drivers tends to be zero, too.  Hardware
> and software need to go together.  Most companies are developing and
> selling the package.

But the key is what is the motivation?  People buy the HW, and get the
drivers because they're needed.  They want a video card/printer/modem,
the software is something that's needed to make it work.

>
>  Better questions are:
>
>  1. *Can* we sell the software independently?  Would it be worth the
> effort?  Are we likely to ever do so?
>  2. What knowledge is in the software?  What will giving that
> knowledge away cost us?
>  3. What potential benefit can we reap by accepting Free Software
> contributions?
>
>  #1 takes your question and looks at it properly.  Maybe a company
> might actually be able to sell their code to others looking to do the
> same thing.  If not, forget this objection.  Even if they *could* sell
> it, would the expected revenue from selling the code be worth the
> effort of turning it into a salable product?  Are they ever going to
> do so?

Right - fundamentally they are hardware companies.  the software is
only to support the HW - they need it to sell the HW, but it's a cost
of business, not a source of profit.

>
>  #2 looks at the potential dangers of reverse engineering.
> Sometimes, by reading source code, you can deduce quite a bit of about
> the hardware.  (Purportedly, modern 3D video cards and their drivers
> are extremely tightly integrated, to the point where the boundaries
> are blurred.  That might be a legitimate concern (or not).  But for a
> print server or NAT router?  Not so much.)  How likely is it that
> someone will go to that effort?  What' will it cost 'us' if someone
> does so?
>

1.  See previous answer - if they're doing the stuff in the driver,
they're in the software (WInModem) business.  If it's SW getting
loaded on the card (firmware) - that's what FPGAs & ROMs are fore

2.  Reverse Engineering - you're right, they need to be afraid of the
poor FOSS guys who can't afford to build cards.  The competitors would
never use debuggers, oscilliscopes, digital tracers, x-ray machines,
and all those other tools they themselves use to reverse engineer th
product :)

Especially as complex as modern video cards are, knowing the
input/output specs will not provide insight into the inner workings of
those complex GPUs.  How well does knowing the x86 pinouts and
assembly help someone know how the latest Intel processor work?

>  #3 looks at the other side of the cost/benefit question.  What's the
> ROI likely to be?  The LinkSys WRT54G shows that it can conceivably be
> quite high.  Third-party software has fixed bugs and added outstanding
> features.  If LinkSys had been smart, and released their own stuff
> under a license like the Mozilla MPL, they might well have been able
> to reap huge rewards from an Open Source approach, by getting others
> to (re)write their code for free.
>

That's a key question - and if it's open source you get free labor -
which is probably why IBM & HP have been supporting Linux.  You want
to spend money on your area of expertise and where you can make a
difference, not on the common area.

jeff