Slashdot Mirror
Proposed Law To Open Code ... In Cars
SEWilco writes: "A Minneapolis Star Tribune story points out that small repair shops say they lose money because they don't have access to car computer codes. These 'codes' are the diagnostic messages used by onboard computers to report problems and perform tests. Older designs only required a jumper wire to make a test light flash code patterns, certain dashboard actions to display codes, or a cheap display terminal from a parts store. Now the interfaces and code meanings are more complex and undocumented, so only auto dealer repair shops can easily find causes of some problems. U.S. Senate S.2617 and House H. R. 2735
would force auto manufacturers to share the codes. Sen. Wellstone says that manufacturers are acting like a cartel, blocking independent shops and car owners. But GM is being helpful."
Finally an open source analogy your grandmother can understand
Step 1: Wait for this law to be voted and put into effect.
Step 2: Start pushing for a law that mandates opening code... everywhere. Say that it prevents small independant consultants from repairing Microsoft ware.
Its funny, because its sort of related to the Ask Slashdot question about car performance tuning software a week ago. All these computerized systems have left cars incredibly complex to tune, repair, or modify.
Thankfully for some vehicle types, the enthusiast market has reverse engineered a lot of these protocols and codes. I've driven two different Audi's for the last three years, and spent an ungodly large amount of time and money tweaking and otherwise customizing them. Our enthusiast community has software like VAG-COM which can provide a suprisingly large amount of capability for Volkswagen, Audi's and other VAG-group cars. But now Audi (and presumably Volkswagen) is changing their protocols yet again, keeping things proprietary and secret. Thankfully, I'm sure they'll be reverse engineered yet again.
Even with the capabilities the software has, we're still faced with having very good documentation for what most "sensor" blocks are, but essentially none for what the "settings" blocks are. I can read anything I want, but without insider VAG knowledge, I can't recode a damn thing.
Amazingly through trial and error, people have even figured out how to reprogram basic functionality on their cars, like how the automatic transmissions shift.
I would love to see this law passed, but it doesn't help things much if its just emissions codes that have to be released.
Think about what? What are they supposed to do - buy a car without and onboard computer? or maybe buy a car from all those car companies who willingly make their codes available for anybody? or they could just not get those parts of their car fixed. this is pretty much out of the consumers hands, other than writing their congressperson to get the bill passed.
Actually, some cars have this. Got a Malibu ... turn the key slightly differently on start up, and watch the dashboard lights, then cross reference that with the manual (Chilton's, IIRC, not the manufacturers). Yeah, I'd love to have it display "Ignition coil #3 is f$cked", rather than what amounts to a binary display, but, then again, it keeps casual tinkerers out from under the hood ... and, IMHO, there's a whole lot more stuff now (compared to a decade back) that is not user servicable.
The other sad thing is that computer diagnostics are replacing brains. Take the alternator ... 99% of the time, it's the diode pack. But garages love to go the long route ... which involves more parts, more time, and thus more money. One car I've looked at was quite clear on the error message (Dead battery and/or bad ground). Well, at least it had the right system (electrical), but it took a pair of 25 cent brushes (in the alternator) to fix it.
Keeping the error codes secret also hides another problem ... a good percentage of the time, the error message is just plain wrong!
They are talking about what amounts to ERROR CODES here. Not SOURCE CODE. Mechanics want to know what certain error messages mean, they don't want the fucking source code. How many mechanics are going to fix bugs in a RTOS car? And if they had access would you even want to drive your car knowing that "Big Joe" from down at Jiffy Lube reworked some of the brake logic?
Everyone saying this is some sort of victory for Open Source is an illiterate moron, or didn't read the article.
It's not a victory for Open Source per se, but it's a similar issue. It's the idea of openness versus closedness. Does the consumer benefit more if the protocols for talking to car computers are kept closed so that only the manufacturer can perform repairs? Or is the consumer better off if his car's computer has open protocols that allow any competent mechanic to diagnose and repair the car?
There's a very direct parallel to the open-source-vs-proprietary-software debate. If this issue gets a lot of publicity, it'll make more people think about then open-versus-closed issue. That will make it easier to explain the benefits of open source to non-technical people.
Wasn't it Bob Young who asked "Would you buy a car with the hood welded shut?"
TheFrood
If you say "I'll probably get modded down for this..." then I will mod you down.
Photocopier makers used to be notorious for this sort of thing (they still are, it's just that all new photocopiers are now laser printers :) ).
Any sufficiently advanced technology is indistinguishable from a rigged demo
--Andy Finkel (J. Klass?)
And if they had access would you even want to drive your car knowing that "Big Joe" from down at Jiffy Lube reworked some of the brake logic?
That's insulting to smart automotive technicians everywhere, and there are plenty.
It's like saying all programmers are dumb, just because there are some dumb programmers in the world.
"And like that
This is what makes too many laws. There is already a solution in place, the citizens and representatives just need to use them:
Not to excuse the other guy, but I expect the real issue is whether you want to drive a car run by source code that hasn't been *thoroughly* tested. I can't be sure that the auto manufacturers *thoroughly* test their code, either. However, the manufacturer has more time between writing the code and releasing it than the mechanic did.
On the other hand if auto manufacturers were run the same way software manufacturers seem to be, I'd be less confident that any serious testing had been done at all. =-)
-Paul komarek
"The old garage mechanic is gone," he said. "In fact, the term 'mechanic' is gone. They're called technicians now. These people have to be very intelligent people. They're working on computers, and it's a high-tech industry. . . . How does the guy that's in a little town of a population of 500 people get educated about the electronics of a $30,000 new automobile?
Could he be any more condescending?
Don't moderate flamebait as Troll. Know the difference or you will be Meta-moderated.
It is interesting how senators are willing to help the little auto mechanic, from being locked out by the large manufacturer, but not the little pc software developer from being locked out by the large software developers.
What is it? Is it because auto mechanics is an easier bussiness to understand? Are they better organized? Or is it because the small automechanic is an old institution.
This is NOT about cars, this is about software and about interchange formats!
I hope this law makes it: this will be the grounds for asking for a law that forces Microsoft et al. to open their document format.
Today, independent programmers cannot perform services for their customers because large software makers hide the specifications for document formats. This means that an independent programmer cannot properly access and service the customer's database/spreadsheet/file without specialized tools that he would be required to buy from the software maker. This, in fact, amounts to some degree of cartelization, in so far as only programmers "licensed" by the software maker have access to these tools, in what amounts to a cartel. By effectively preventing independent programmers from obtaining revenue from services, this situation limits choice for consumers and harms many independent programmers, who are sometimes the sole financial support for their family.
This about it. This may be the way to go.
free the mallocs!
Which brings up an interesting point, which is that aftermarket parts tend to be poorly manufactured and need to be replaced sooner than the factory equivalent.
This is a generalization, and like all of them, it is false. OEM parts do tend to be high quality, but that doesn't mean the aftermarket parts are not. Some aftermarket parts are better, some worse.
Many OEM parts are not made by the OEM, they are made by some other company, who slaps the OEM name on the first 1000 parts off the assembly line, and then their name on the rest they make that day. Often one assembly line will make the same part (which may or may not be the same quality) for several different manufactures. Manufatures outsource a lot of this because a company that focuses on just one part can often make that part cheaper and better. Sometimes the buyer will order parts with lower quality, othertimes it is exactly the same. As and example, 80percent of all spark plugs are made by Champion for example, they put whatever name the buyer wants on it.
I have used both OEM and non-OEM parts. When there isn't a cost difference OEM is a good way to be sure. When there is you need to check out quality, if there is a choice on quality or price stores will warn you and give you a choice.
http://www.obd-2.com/ is your answer. About $150 and you can interface with any of the 3 main OBD-II interfacing protocols. Downloadable updates. Error code sets for your manufacturer. Tons of information. And the author has a very impressive resume when it comes to automotive and computer diagnostic design!
:) Software has a hard time transforming into a torque wrench!
Mix this with a little bit of community-brainstorming on an automotive message board, and most car problems can be solved.Not having the tools to fix something is a whole other issue
.... um, i lost you after "0110100001101001".