cars & computers

[Kragen]

Some guests (Baha'i travel teachers) were talking with me last night
about cars and computers.  Evidently, a number of mechanics they know
have stopped fixing cars outside the shop because, well, they can't.
They used to be able to go home and fix cars at home, or fix their own
car, but nowadays, you have to hook up to the car's computer, and it
tells you what's wrong.

Which is all well and good, but you can't hook up just any computer to
the car's computer.  It has to be an expensive proprietary system from
the car manufacturer.  So only the garages can afford these things,
and sometimes they're not even available to the general public!

That's the minus side.  On the plus side, all this computerization
apparently cuts way down on maintenance costs --- a lot of the diagnosis
of the problem is done automatically, by the computer, saving the
mechanic's time.  Which is good, because mechanics' time is getting a
lot more expensive, largely due to microcontrollers in the cars.

I'd like to see car computers get more open, myself.  I should be able
to get a standard RS-232 cable, plug it into my dashboard, plug the other
end into my computer's serial port, and use some sort of simple, open
protocol to find out what the car thinks is wrong.  Ideally, if the
car actually *knew* what was wrong with it, it could even give me part
numbers and detailed installation instructions.

This would cut *way* down on maintenance costs.  It would be a big win for
people who drive cars, especially after a few years when failures start to
become more frequent.  It might even be a big win for car manufacturers,
because it would cut down on maintenance costs during the warranty period.

How likely is this to happen?  It may be that the large automotive
companies will not want it to happen --- I'm not sure what their financial
incentive structure is with regard to the mechanics who work on their cars
(after the warranty).

-- 

But there's another current, pushing in the other direction.  Wire EDM,
abrasive waterjet cutting, more radical automatic-fabrication methods,
and advances in robotics are bringing the capital needed to build a car
way down.  In the near future, it will be possible to machine all the
parts for a car in a few days without any human intervention, on a CNC
(computer numerically controlled) mill costing about $1000 or $2000.
All you'd have to do is put the parts together, which should probably
take a week --- although some big assembly plants have it down to less
than ten man-hours per car for the final assembly process.

(More radical methods might make it possible to machine the whole car
in a matter of hours, with no subsequent assembly.)

Blueprints for the cars may be available over the 'net, like Linux.
I'm sure that the Free Blueprints Foundation will shortly follow the
availability of cheap automated fabrication plants.

I don't think that we're going to drive the big car companies out of
business in the next decade, but I think we'll see a lot of new companies
with less than 100 employees manufacturing custom cars, at a cost not
much more than the big car companies' products.

This could make it more attractive to open up the car's microcontrollers.

-- 
<kragen@pobox.com>       Kragen Sitaker     <http://www.pobox.com/~kragen/>
I don't do .INI, .BAT, .DLL or .SYS files. I don't assign apps to files. I
don't configure peripherals or networks before using them. I have a computer
to do all that. I have a Macintosh, not a hobby. -- Fritz Anderson