Slashdot Mirror
Open Source Car on the Horizon
PreacherTom writes "So here's a question: can open-source practices and approaches be applied to make hardware, to create tangible and physical objects, including complex ones? Markus Merz believes they can. The young German is the founder of the OScar project, whose goal is to develop and build a car according to open-source principles. Merz and his team aren't going for a super-accessorized SUV — they're aiming at designing a simple and functionally smart car. The OScar is not the only open-source hardware project out there: others include Zero Prestige, which designs kites and kite-powered vehicles, and Open Prosthetics, which offers free exchange of designs for prosthetic devices."
The young German is the founder of the OScar project, whose goal is to develop and build a car according to open-source principles.
Does that mean it will crash less than other cars?
The theory of relativity doesn't work right in Arkansas.
Open-source principles will be good for innovation.
But there will be a BIG problem with laws - especially mandated safety and emissions testing.
That's designed on the assumption that large numbers of essentially identical cars are produced by well-funded manufacturers, so the cost of a lot of crash and emission-control testing and design work can be spread out over many units and become affordable.
Even if you are building using zero-emission or well-tested stock power plants, good luck on getting the safety-testing requirements relaxed. A poorly-designed car endangers, not just those in it, but those in vehicles around it.
With cars the "blue screen of death" is literal.
Bantam Dominique roosters crow a four-note song. Once you've heard it as "Happy BIRTHday" you can't NOT hear it that way
Actually, depending on your variation, wouldn't it be...
:-/
"You will have to search out mechanics on your own, and in most cases if you find them they will laugh at you for being too stupid to use the car, and point you to libraries spread throughout the country. In each of those libraries there will be manuals that give small, different chunks that sort of relate to the problem you're having. Sometimes you will be lucky enough to find a mechanic who has seen your problem before, and actually gives you a straight answer and gets you back on the road. But good luck on the rest of the times."
I think the most important part of Open Source development of ANYTHING is standards. You need to have a standards body. The problem with the auto industry today is there are no real standards. Take for example custom wheels--a simple, non-moving piece of metal that basically holds the tire. It's main purpose is cosmetic after the basic functionality that all wheels share (round, has bolt holes in the middle, etc.). You'd think it would be simple to get a different wheel for your car, but if you ever try you'll find hundreds of different widths, bolt-patterns, diameters, etc.
This Open Source car would only be better if there were standards employed in these particular sections. Or have any connections be customizeable on both sides of the connection. So, if someone invents a better wheel pattern, it's easy to change the disc brake assembly to to fit it (dependency).
The problem is that just having the design isn't going to get you very far because of the specialized components involved. A car is very expensive to build but at million plus quantities it's very cheap. But try to one-off one gear for a transmission sometime (it'll be THOUSANDS to get the precision in a $900 off-the-shelf manual transmission like Mazda makes for Ford).
Instead, from the design stage, standardize everything. A standard ring or star topology for communications and power bussing throughout the car. Then each powered device has a microcontroller that turns it off or on. Then the microcontroller can report back it's status to a central computer. Most of the electricals are easily standardized. Where you run into problems is precision machined steel parts of an engine and transmission. Replacing also those with electrics is the way to go. Use electric motors, magnetic suspension, etc. Modular body panels can have their own microcontrollers also, so the car can reconfigure itself based on what you have mounted. You have the rear door in place, the rear door up/down button appears on the interface. The top is off, no sense showing the moonroof control. Etc etc.
RFC's and the like are what's really made stuff like linux possible. It's not just having the source but having the standards that really make everything easy to work with, and make sure that many different programmers can all work on different sections of the project without worrying about if their module can talk with the others.
Cool! Amazing Toys.
I'm with you, in that I drive older cars, mostly for this reason. I'm all for this "open source car" thing, though; at a certain point the future, virtually every car on the market will have a computer in it. Do we want to be able to service these things ourselves, or are we going to have to take them to a Certified Mechanic who needs an expensive proprietary interface to work on the car? First I guess it should be noted that you are taking all of this out of context. The quote in the article is referring to the fact that much of hardware design is done via mock-up in software packages, it is not insisting that running the car is mostly software.
Second: Older cars have the same problem. "What? They do not!" you say! Yes, yes they do. How much money does it cost for all the specialized tools needed in vehicle repair? Flare nut wrenches? No use other than brake jobs. Flywheel puller? Special presses?
You already need to use expensive, sometimes proprietary (Ford fuel line disconnect) tools to do the job, how is that different than needing to connect a car up to a computer interface?
BTW, you will find that those fancy computer interfaces can be had for under 200 bucks, which is less than many of your single-purpose tools needed for car work and supports a whole suite of diagnostic purposes.
Blue or I walk.
Sweet informative mod.