Ssssorry /Gilly

February 8, 2011 4:00 PM

The U.S. Department of Transportation released results from the joint National Highway Traffic Safety Administration/NASA study on sudden vehicle acceleration in Toyota cars requested Spring 2010 by Congress. Short version: NASA engineers found no electronic flaws in Toyota vehicles capable of producing the large throttle openings required to create dangerous high-speed unintended acceleration incidents.

NHTSA's analysis indicated that these types of complaints generally do not appear to involve vehicle-based causes and that, where the complainant indicated that brakes were ineffective...the most likely cause of the acceleration was actually pedal misapplication (i.e., the driver's unintended application of the accelerator rather than, or in addition to, the brake).

The NHTSA has proposed a few new, non-Toyota specific actions as a result of this study:

* Propose rules, by the end of 2011, to require brake override systems, to standardize operation of keyless ignition systems, and to require the installation of event data recorders in all passenger vehicles;
* Begin broad research on the reliability and security of electronic control systems;
* Research the placement and design of accelerator and brake pedals, as well as driver usage of pedals, to determine whether design and placement can be improved to reduce pedal misapplication.

(The results and findings are available from the link as PDFs)

posted by 2bucksplus (69 comments total)

So pretty much the same result as the alleged sudden unintended acceleration in Audis in the 80s.
posted by jedicus at 4:04 PM on February 8, 2011 [8 favorites]

Doesn't surprise me at all that people who dialed 911 reporting their car wouldn't stop instead of turning off the engine or shifting into neutral would be inadvertently misapplying the accelerator.
posted by nathancaswell at 4:05 PM on February 8, 2011 [15 favorites]

Is this why our space program hasn't accelerated for decades?
posted by spasm at 4:07 PM on February 8, 2011 [3 favorites]

feet to meters, yes
posted by clavdivs at 4:09 PM on February 8, 2011 [1 favorite]

Seriously trying to comprehend "pedal misapplication". Are people trying to dance the tango down there or something?
posted by Jimbob at 4:15 PM on February 8, 2011

Turn It Off

Switching off the ignition is a sure way to silence an engine, but it’s probably the least desirable action because it will also make the car more difficult to maneuver. It causes a loss of power-steering assist, plus it will cut off vacuum boost for the brakes. The new wrinkle here: the keyless, push-button start-and-stop systems in many vehicles. Owners need to be aware that these systems require a long press of the button to shut off power when the car is moving (so that an inadvertent touch of the button by the driver doesn’t kill the engine). Here, too, the Toyota was slightly behind the curve; the Infiniti’s engine shut down after a 2.5-second press of the button versus 3.3 seconds for the Camry. In an emergency, that would probably feel like an eternity. For some perspective, if a V-6 Camry’s throttle became stuck at 60 mph, the car would accelerate to nearly 80 mph before the engine would surrender.
posted by Roger Dodger at 4:16 PM on February 8, 2011 [3 favorites]

Why did they send that job to NASA? That sounds like the kind of job that NIST should be doing.
posted by Chocolate Pickle at 4:24 PM on February 8, 2011 [1 favorite]

Switching off the ignition is a sure way to silence an engine, but it’s probably the least desirable action

I'm 100% sure calling 911 is a worse option than switching off the ignition.
posted by nathancaswell at 4:25 PM on February 8, 2011 [3 favorites]

Watched a lot of this on CNN today. Basically Toyota reps said, "It was human error." Without ever saying it.
posted by Splunge at 4:29 PM on February 8, 2011

Switching off the ignition is a sure way to silence an engine, but it’s probably the least desirable action

Turning an engine off is going to get even more difficult to figure out for the average driver, what with the move towards keyless ignition systems.
posted by linux at 4:29 PM on February 8, 2011 [1 favorite]

I'm 100% sure calling 911 is a worse option than switching off the ignition.

I'm 100% sure calling an injury attorney is a worse option than calling 911.
posted by Mister Fabulous at 4:30 PM on February 8, 2011

For some perspective, if a V-6 Camry’s throttle became stuck at 60 mph, the car would accelerate to nearly 80 mph before the engine would surrender.

Unless, of course, you had the presence of mind to actually switch to neutral as well.
posted by Nauip at 4:30 PM on February 8, 2011

Propose rules, by the end of 2011, to require brake override systems, to standardize operation of keyless ignition systems, and to require the installation of event data recorders in all passenger vehicles;

Event data recorders in all new vehicles? I wonder whether this is overkill that will cost more than the benefits, or if it will have a sudden and drastic effect on auto prosecution/litigation (fender benders, red light runners, DUI etc.). It could also provide solid statistical evidence in favor of automatic driver systems.
posted by anigbrowl at 4:32 PM on February 8, 2011 [1 favorite]

Event data recorders in all new vehicles?

For some reason (dollar dollar bills, y'all) I could see insurance companies wanting demanding the data, regardless of a crash or incident.
posted by Mister Fabulous at 4:43 PM on February 8, 2011 [3 favorites]

I always believed in you, Toyota
posted by Flashman at 4:54 PM on February 8, 2011 [4 favorites]

I read a story on Gizmodo regarding Steve Wozniak being able to demonstrate the problem repeatably in his Prius. I am more inclined to believe an engineer like Wozniak who is not under government pay, when he says that it's a software problem and not a mechanical problem. It doesn't hurt that I'm a very experienced software engineer, and one of the best in my field and I would be reluctant to trust my skills to properly code a drive by wire system and would be spending a ton of effort in creating fault-tolerant systems with rock-solid recovery.

Toyota will spend as much money as they can to create belief that it is pilot error or a mechanical problem. The reason being that *if* they concede that then every crash will be blamed on their software and that will cost them far more in lawyers and settlements. Color me unsurprised if Toyota is the first company to have the equivalent of a black box recorder.
posted by plinth at 4:55 PM on February 8, 2011 [2 favorites]

...insurance companies wanting demanding the data...

It'll start out innocuously enough: insurance rates will go up, but there'll be a discount if you agree to provide your black-box data. But once everybody is taking that discount, providing the data will become mandatory -- if you don't, then we won't write you a policy. Oh, and don't forget about the GPS logging that'll get added in along the way. New features, yo!
posted by spacewrench at 4:57 PM on February 8, 2011 [2 favorites]

My old Mazda Protege would get throttle stick from time to time (pretty rarely but it did happen). In every case the floor mat had slid up and caused the pedal to get stuck. I'd switch into Neutral, brake if needed, and reach down and pull the mat back. It wasn't a factory mat and I eventually made some modifications with a knife to prevent future incidents.
posted by juiceCake at 5:08 PM on February 8, 2011

plinth: "I read a story on Gizmodo regarding Steve Wozniak being able to demonstrate the problem repeatably in his Prius."

Intriguing. I went looking for details, and found this ABC report, where several commenters (I know, I know) said that he wasn't clear on the cruise control options. Did Woz ever make a more detailed report of exactly what he did during testing?

This other guy gives a detailed view of the Prius radar-driven cruise control, which tracks traffic ahead to adjust speed to conditions. Great if you're approaching a red light and cars ahead are slowing, but if you're on a road with turns, it can lose track of traffic ahead of you, assume that the way is clear, and speed up unexpectedly as it goes around that turn. But he also notes that if someone cuts in front of you suddenly, even on the straight, while cruise control was engaged, the car could get confused and speed up. You can manually override it, but this works against the mindset of cruise control, doesn't it?

I'm not a driver and not an engineer, but is this kind of adaptive cruise control a good idea in any car if people are going to use it in places where they shouldn't (twisty roads) and if it can't handle with relatively common highway conditions like someone cutting in ahead of you?
posted by maudlin at 5:15 PM on February 8, 2011 [1 favorite]

(that would be an event data recorder, no?)
posted by The Bridge on the River Kai Ryssdal at 5:15 PM on February 8, 2011

I am more inclined to believe an engineer like Wozniak who is not under government pay, when he says that it's a software problem and not a mechanical problem.

I appreciate this conspiracy theory, especially since the one I am so used to hearing is diametrically opposed. Once the government became the majority owner of GM and Chrysler, you see, it behooved them to wage "jihad" on Toyota, the major competitor to the American car industry. Some people would tell you that Congress needed Toyota as a scapegoat in order to recoup the taxpayers' investment.

Also, GM and Chrysler are largely unionized, unlike Toyota apparently, so the Democratic controlled Congress got the added benefit of boosting the Teamsters and the UAW; people would say. Also, don't cha know, Toyota has set up a bunch of new factories in red states recently.
posted by 2bucksplus at 5:16 PM on February 8, 2011

Color me unsurprised if Toyota is the first company to have the equivalent of a black box recorder.

They already do, and have since at least 2005. Our '05 Prius manual warns us that speed, acceleration, brake status, etc. are recorded and will be made available to a court if required by law.
posted by werkzeuger at 5:27 PM on February 8, 2011

I read a story on Gizmodo regarding Steve Wozniak being able to demonstrate the problem repeatably in his Prius. I am more inclined to believe an engineer like Wozniak who is not under government pay,

... than read the NASA report and properly critique it.
posted by benzenedream at 5:27 PM on February 8, 2011 [11 favorites]

Also, GM and Chrysler are largely unionized, unlike Toyota apparently...

A friend of mine works for Toyota in Cambridge Ontario. Every year (if memory serves) they have a vote whether or not to unionize with the CAW. Every year the vote is a majority of not to join. Toyota mirrors to some degree I am not familiar with, the deals that workers in the CAW get (and probably do so in the States as well). One of the arguments in favour of the CAW during the vote is that if you get in a fight on the floor you can't get fired for it.
posted by juiceCake at 5:33 PM on February 8, 2011

OK, here's my story. I was merging onto in Interstate with two-lane traffic on a notoriously hectic and 1950s-designed on-ramp (uphill/peak/downhill, curving to the right, followed by an immediate off-ramp - gotta love Pennsylvania) during fairly crowded traffic. I stomped the gas to the floor (my mistake) and it just stayed there, with no space between the pedal and the floor itself. While quickly checking all my mirrors to get a bearing on all the vehicles near me, I frantically stomped the pedal repeatedly to perhaps dislodge it from the floor, with no result. As my van accelerated, I put my left foot on the brake to stop surging ahead... this caused my brakes to labor mightily while the engine tried to rush me forward, but it kept my speed in check and I wrestled the van to the narrow shoulder on the right. I THEN slipped it out of gear into neutral, whereupon the engine howled full-bore, so I turned the off, which cut all my power - but I was on the shoulder by then and didn't need much steering or brake. All this took less than 10 seconds and 1/4 mile. When I looked under the hood, the accelerator cable seemed hung up in a "sleeve", and when I tapped on it, it popped back into position. I lubricated that damn thing then and there. It was pretty tense there for a bit, but I just reacted logically I guess. It's not the first time an automobile has tried to kill me.
posted by Ron Thanagar at 5:35 PM on February 8, 2011 [3 favorites]

Heh. I've done that a couple of times. It's really easy. Just have your mind stuck on autopilot, do a light-touch coast, forget your foot is on the gas not the brake, and push down to come to a complete stop.

Both times were in parking spaces, and both times I hit nothing. It was so immediately obvious what was happening (foot goes down, car speeds up = duh!) that I managed to release the accelerator and apply the brake in less time than it took to travel approximately six inches. If I'd been slower to react, like, say, [ageist slur], my driving record would probably be a whole lot less spotless, and my zones a whole lot more crumpled.

All that said, I'd be willing to bet everything I own that the majority of the Toyota cases were nothing like mine. I bet they involved cruise control. Cruise control is incredibly fucking dangerous and should have been Nadered out a long, long time ago.

How To Die in Ten Easy Steps:
1) Set the cruise.
2) Stop making contact with either pedal.
3) Stay like that awhile.
4) Get distracted.
5) Suddenly need to brake.
6) Panic.
7) Become unable to think.
8) Slam your foot down on a pedal.
9) Panic some more.
10) Become unable to revive.

Also: Hey, remember the thing about the wonky floor mats? That could be it, too.
posted by Sys Rq at 5:42 PM on February 8, 2011 [2 favorites]

For some reason (dollar dollar bills, y'all) I could see insurance companies wanting demanding the data, regardless of a crash or incident.

I do insurance defense litigation for a living. Believe you me, if this data was available, we'd be demanding that it be produced 100% of the time. As a significant percentage of the cases we see are utter crap, I have the impression that something like this would actually drive rates down, because insurance companies would wind up paying fewer claims they shouldn't be paying.

The plaintiff's bar would probably hate this, because it would undermine their ability to recover on marginal cases.
posted by valkyryn at 5:47 PM on February 8, 2011 [3 favorites]

...require the installation of event data recorders in all passenger vehicles;

Wait, these don't already exist? How do they debug them?

I really, really hate when lawyers and management do engineering's job and do it wrong.
posted by DU at 5:52 PM on February 8, 2011

Sys Rq wrote: "How To Die in Ten Easy Steps:"

Let's condense that to two easy steps:

1. Drive a car
2. Panic

The rest is pretty much irrelevant. Panicking behind the wheel will get you or someone else killed if you are at speed.

Actually, now that I think about it, only one step is necessary:

1. Drive a car

Cars are fucking dangerous. (in much the same way guns are..when operated by twits)
posted by wierdo at 5:59 PM on February 8, 2011 [8 favorites]

It wasn't a factory mat and I eventually made some modifications with a knife to prevent future incidents.

The sad thing is, in today's world this would probably constitute a violation of the DMCA. The knife is clearly a circumvention device. You even freely admit to modifying the hardware. You'd be so dead, man.
posted by Civil_Disobedient at 6:09 PM on February 8, 2011

Why did they send that job to NASA? That sounds like the kind of job that NIST should be doing.

The good news: no more NASA budget shortfalls!
posted by XMLicious at 6:22 PM on February 8, 2011

Yeah, weirdo, but if you panic without cruise control, you'll probably just react in a fairly automatic fashion, the first step of which is releasing the gas pedal, which provides a nice, safe deceleration and a little more time to think.

With cruise control, for one thing, your foot isn't on alert. If you panic (which is more likely if you're driving in an unusual manner), your immediate reaction is to get your foot into position as quickly as possible, which can mean stomping on a pedal. Gas or brake, that's a bad idea.

You know what else results in a released accelerator, provided the cruise control isn't on? Falling asleep. Granted, falling asleep while driving is pretty dang dangerous. But it's a lot more dangerous when the car doesn't slow down.
posted by Sys Rq at 6:23 PM on February 8, 2011 [1 favorite]

Seriously trying to comprehend "pedal misapplication"

Take a look at this New Yorker piece by Malcolm Gladwell from 2001.
posted by Dasein at 6:31 PM on February 8, 2011

Wow, there are a ton of misconceptions in this thread.

1) Toyota cars already have data recorders.

2) There was no electronic problem, but there was a problem with the pedals getting on the carpet in some models.

3) The case with the lexus out in the desert. It was a rental and it was a push to start model. So there was no way to stop the car by turning the key. In order to stop the engine he would have had to have held the power button down for 10 seconds, which isn't obvious. He might have been able to shift in to neutral or something like that, but apparently he didn't realize it. He had been pushing on the break, not the gas and in fact the breaks had been applied, and actually destroyed from being driven so hard.

4) The people who had gotten their cars to accelerate out of control electronically had apparently sent very unusual signals to the computer, basically hacking it. According to Toyota (and now NASA) they weren't thinks that would happen under normal conditions.

---

Anyway, it was obvious from the beginning that the rash of uncontrolled acceleration was caused by human error. The WSJ had a chart showing the number of recorded events per month and it basically SPIKED in one month when the accelerator problems were in the news, and then subsided right away. How could Toyota have an accelerator flaw that only happened one month and then went away?

Basically people were pressing the pedal and instead of thinking "Oops, I pushed the wrong pedal" they thought "OMG it's the accelerator problem"
posted by delmoi at 6:40 PM on February 8, 2011 [4 favorites]

With cruise control, for one thing, your foot isn't on alert

Well, you're not _supposed_ to use it that way, but I realize most people do (in anything resembling traffic I do keep my foot hovering over the brake during CC).

Newer cruise control is much safer though since it adjusts speed and even disconnects (releases accel, like you're talking about) based on proximity. This is pretty common in luxury cars now ("active cruise control" or w/e) and I'm sure will make it's way down the line.
posted by wildcrdj at 6:40 PM on February 8, 2011 [1 favorite]

Someone mentioned insurance. I think the future of insurance is actually per mile insurance, but not only that road specific per mile insurance. So you'll have a GPS unit that will tell the insurance companies what you're road you're driving on, how fast, etc and calculate your risk of accident in real time, and you'll get charged an appropriate amount.

Not only will this cut the insurance costs for people who don't drive much, it will probably give people real time feedback on the risks they are taking and cause them to take safer roads, etc.

And of course we'll have driver-less cars. It always blows my mind when people say that we can't have robot cars because they will always make mistakes, which totally ignores the fact that humans make mistakes to the tune of tens of thousands of deaths per year.
posted by delmoi at 6:47 PM on February 8, 2011

Roger Dodger writes "Switching off the ignition is a sure way to silence an engine, but it’s probably the least desirable action because it will also make the car more difficult to maneuver. It causes a loss of power-steering assist, plus it will cut off vacuum boost for the brakes."

Vacuum assist brakes easily have at least one good hard brake stored in their vacuum boosters. Many ABS systems have several hard brakes stored in their accumulators.

delmoi writes "The people who had gotten their cars to accelerate out of control electronically had apparently sent very unusual signals to the computer, basically hacking it. According to Toyota (and now NASA) they weren't thinks that would happen under normal conditions. "

Have these people never met a user? If it's even remotely possible to do it some barely competent driver out there is going to manage to do it accidentally. I had a girlfriend call me once in a panic because her check engine light was flashing (because she managed to put her car in diagnostic mode via the ignition key).
posted by Mitheral at 6:50 PM on February 8, 2011 [1 favorite]

Hell, I just wish every single local news channel would just learn that the issue was/is called "UNINTENDED acceleration" instead of "SUDDEN acceleration." Every auto manufacturer out there would LOVE for all their vehicles to display "sudden acceleration", and would indeed market it as a feature if they had it.
posted by rhythim at 6:51 PM on February 8, 2011

Have these people never met a user? If it's even remotely possible to do it some barely competent driver out there is going to manage to do it accidentally.

They are accidentally going to attach wires to their ECUs and send random electrical signals to it? I find that kind of hard to buy...
posted by delmoi at 6:52 PM on February 8, 2011 [3 favorites]

According to this what Woz found was the adaptive accelerator in cruse control mode, basically when push the ACC button down and hold it, you change the target speed in 5mph increments, so when you let it go it go the car keeps accelerating.

But, if you just tap the break once it will cancel the cruse control, or if you turn of cruse control, etc. I've had the same thing happen driving a 2006 Sequoia, it was a little surprising, but not really 'scary' in any sense. It's also unrelated to the main problem, which was acceleration that drivers could not stop.
posted by delmoi at 7:05 PM on February 8, 2011

The push-and-hold to turn the car off is a glaring usability problem for these kinds of situations.

I'm guessing that there's no obvious feedback when you press the ignition button to encourage you to hold it? No progress bar, no spinner, no 'bing', nothing?

3.3 seconds of panic is a long long time.
posted by anthill at 7:13 PM on February 8, 2011 [5 favorites]

I was actually kind of hoping this story would stick around long enough to drive down the prices on used Toyotas.
posted by box at 7:34 PM on February 8, 2011 [1 favorite]

I keep trying to imagine what I would do in an exciting Hollywood version of these unintended acceleration events, but the fantasy doesn't work because I have a manual, so the only thing connecting the engine to the drivetrain in the first place is my will that such a connection exist at that moment. It can be broken in multiple places.

But I'm still going to find out firsthand how long I have to hold down the start button to kill the engine while the car is in motion...
Just so I know.
Just in case...

:-)
posted by -harlequin- at 7:39 PM on February 8, 2011

I do insurance defense litigation for a living. Believe you me, if this data was available, we'd be demanding that it be produced 100% of the time. As a significant percentage of the cases we see are utter crap, I have the impression that something like this would actually drive rates down, because insurance companies would wind up paying fewer claims they shouldn't be paying.

I would love to believe that to be the case. I have a nearly spotless driving record. No accidents, one speeding ticket in 12+ years of driving, no other moving violations. My only insurance claims have been comprehensive. But I can see this scenario:

The day I finally do get in an accident, I'll have been going 7mph over the speed limit (as I often do on my daily I-5 commute). Some idiot cuts me off, we wreck. Cops find fault on the other driver for cutting me off. My insurance claims get denied anyhow because I was going 7 over. While it would be nice to eliminate the fluff: claims that are obvious bullshit and shouldn't exist, I cannot for any reason imagine one insurance company not abusing the shit out of this. Virtually zero accidents would be covered, because at any time a driver is or has recently done something illegal. The many lawyers will find a way to screw over every driver on the road.

And nothing will drive rates downward. Sure, some drivers would see their rates drop for awhile, but the average rates will always be rising, new technology or not.
posted by Mister Fabulous at 7:54 PM on February 8, 2011 [7 favorites]

The most pertinent fact to me:

In the same time period that 19 people died as a result of this issue, more than 22,000 people were killed travelling in Toyota vehicles as a result of driver error.

The way we think about this stuff is seriously fucked up.
posted by dry white toast at 8:09 PM on February 8, 2011 [6 favorites]

For anyone who missed it ~14 months ago, CARandDRIVER basically debunked this already by testing one of the recalled V-6 Toyota Camrys:

With the Camry’s throttle pinned while going 70 mph, the brakes easily overcame all 268 horsepower straining against them and stopped the car in 190 feet—that’s a foot shorter than the performance of a Ford Taurus without any gas-pedal problems and just 16 feet longer than with the Camry’s throttle closed.

posted by StarmanDXE at 8:21 PM on February 8, 2011 [11 favorites]

This was actually a pretty big story here in Japan (and time will tell if this half makes it here too). They did a fairly substantial recall campaign complete with prime-time TV ad spots.

What makes this more interesting, though, is that I've never once seen a car in Japan that had cruise control. There's a good chance that, especially given how twisty the highways here tend to be, the government doesn't allow them on the grounds that, y'know, it's probably bad to add a feature that encourages drivers to pay less attention to what they're doing.
posted by DoctorFedora at 8:22 PM on February 8, 2011 [1 favorite]

This sort of thing has cropped up before, and it has always been attributable to human error.
posted by hattifattener at 9:01 PM on February 8, 2011 [1 favorite]

Speech recognition to apply brakes upon a sharp cry of "OH FUCK, SHIT SHIT, WHY AM I STILL SPEEDING UP?"
posted by rubah at 9:32 PM on February 8, 2011

during fairly crowded traffic [...] I stomped the gas to the floor (my mistake)

And how. So you're in heavy traffic and you hit the accelerator? That's quite a mistake to make.

and it just stayed there, with no space between the pedal and the floor itself. While quickly checking all my mirrors to get a bearing on all the vehicles near me

Pretty sure that wouldn't be my first reaction in heavy traffic.

I frantically stomped the pedal repeatedly to perhaps dislodge it from the floor

Fantastic work, Einstein.

with no result. As my van accelerated, I put my left foot on the brake to stop surging ahead

Finally!

Love to know more about your car and this "sleeve" Model? Year? Cruise control?
posted by uncanny hengeman at 12:55 AM on February 9, 2011

Oh, joy! Here we go again. Yet more media-frenzy-induced governmental habit-tracking and incredibly-expensive, federally-mandated, vehicle mollification; because American drivers are dipsticks.

In the late Eighties there was a whole subset of mechanics who cruised around in Audi 5000 sedans instead of their usual pick-up trucks because they bought them dirt cheap from the easily-influenced dimwits who didn't grasp that the average driver is a complete pinhead.
posted by PareidoliaticBoy at 12:59 AM on February 9, 2011

Someone mentioned insurance. I think the future of insurance is actually per mile insurance, but not only that road specific per mile insurance. So you'll have a GPS unit that will tell the insurance companies what you're road you're driving on, how fast, etc and calculate your risk of accident in real time, and you'll get charged an appropriate amount.

Not only will this cut the insurance costs for people who don't drive much, it will probably give people real time feedback on the risks they are taking and cause them to take safer roads, etc.

When risk pools are shrunk down to tiny little risk puddles like that there is no value at all to insurance. It would simply be an accident prepayment system.
posted by srboisvert at 3:45 AM on February 9, 2011

My insurance claims get denied anyhow because I was going 7 over.

That's now how that works. You're thinking of contributory negligence, where if a plaintiff is even 1% at fault, there is a complete bar to recovery. That is how things worked historically, but in the twentieth century, most common law jurisdictions switched to comparative negligence, where the plaintiff's recovery is merely reduced by the extent of his negligence, not barred entirely.

So yeah, you may wind up getting a bit less than you would have, but unless the accident was entirely your responsibility, you will still recover something.

This is already true, by the way. There's plenty of accidents where the plaintiff is found to have been at least partly responsible for the accident, and courts and juries routinely assign fault and recovery along those lines. So while there are definite privacy and civil rights concerns here, this particular one isn't actually all that likely.
posted by valkyryn at 4:43 AM on February 9, 2011

Wow, there are a ton of misconceptions in this thread.
1) Toyota cars already have data recorders.

Do you ever read anything or just comment? From Toyota's own web site: "Toyota Clarifies the Facts About Event Data Recorders":

There are a variety of EDRs installed on Toyota and Lexus vehicles in the U.S. today. The data that is recorded and that can be read out and analyzed differs depending on the type of EDR that is installed in a particular vehicle and when the vehicle was manufactured.
posted by yerfatma at 5:28 AM on February 9, 2011

In my dream world, the code that runs all cars in 2020 is a single open-source operating system.
posted by yerfatma at 5:29 AM on February 9, 2011 [1 favorite]

Do you ever read anything or just comment?

yerfatma: I'm confused here. What about the link you posted contradicts what I said? I said Toyota's have data records. And the link confirms that. What are you talking about?

From the page:

"Toyota started phasing in EDRs into its vehicles in 2001 and has had EDRs in all vehicles from the 2007 model year forward."

Did you somehow think I wrote the opposite of what I actually wrote? Or what?
posted by delmoi at 5:34 AM on February 9, 2011

The future is here now:

Some drivers in Minnesota will soon have an added passenger in their cars: their insurance company. Progressive Insurance Company is using so-called “black box” technology to offer drivers discounts based on how much, how fast and when they drive.

The pilot program, which is strictly voluntary, involves 5,000 Minnesota direct Progressive customers. The drivers will install a data-logging device provided by Progressive into their cars and then upload the data to the insurance company for up to 25 percent in potential savings.

The device, called the TripSensor, is the size of a matchbook and plugs into the On-Board Diagnostic port found near the steering column of most cars made after 1996. The TripSensor records mileage, the time of engine start up and shut down, and the speed at which customers drive.

The interesting question is when will such devices become mandatory.
posted by mygoditsbob at 5:42 AM on February 9, 2011

Did you somehow think I wrote the opposite of what I actually wrote? Or what?

Yes, sorry, I thought that was one of the misconceptions-- not clear.
posted by yerfatma at 6:17 AM on February 9, 2011

Ah, I see.
posted by delmoi at 6:25 AM on February 9, 2011

valkyryn writes "That's now how that works. You're thinking of contributory negligence, where if a plaintiff is even 1% at fault, there is a complete bar to recovery. That is how things worked historically, but in the twentieth century, most common law jurisdictions switched to comparative negligence, where the plaintiff's recovery is merely reduced by the extent of his negligence, not barred entirely."

Here in BC if you are involved in an accident, even if not at fault, with a BAC over the legal limit your insurance can be voided. And the powers that be are rapidly adapting anti-drunk driving enforcement techniques to speed enforcement. They can already impound your vehicle for speeding so it is not much of a stretch to envision them cancelling your insurance.

mygoditsbob writes "The interesting question is when will such devices become mandatory."

Never. If only because their isn't any place to plug them in in many cars. And they can be easily spoofed I bet.
posted by Mitheral at 6:49 AM on February 9, 2011

box wrote: "I was actually kind of hoping this story would stick around long enough to drive down the prices on used Toyotas."

It did, which is why I now own a Toyota. :P

Mitheral wrote: "Never. If only because their isn't any place to plug them in in many cars. And they can be easily spoofed I be"

In the US, every new car (since about 1996), by law, has an OBD-II port, which can be used to access (and log) all sorts of data, including speed and throttle position. Certain insurance companies already give discounts for leaving an OBD-II reader plugged into the car and periodically uploading the data to them.
posted by wierdo at 8:25 AM on February 9, 2011

Here in BC if you are involved in an accident, even if not at fault, with a BAC over the legal limit your insurance can be voided.

Ah, but that's different, actually. Many US insurance carriers--and it doesn't surprise me that Canadian companies are following suit--have not driving drunk as a condition of coverage. They're permitted to do this because government regulators let them as a political decision to look tough on drunk driving.

That's different than just denying coverage simply because you happen to be partly at fault, as you've said, believe it or not. The "even if not at fault" part is what's significant there. There are similar conditions related to things like drag racing or using your car as a taxi: do one of those things and get in an accident and there's just no coverage. But all of those things are spelled out in the policy itself, and they all involve some kind of intentional act. The point of insurance is to protect people from the consequences of their unintentional negligence, so limiting coverage for things like speeding--which is a strict liability offense--would never make it past the regulators.
posted by valkyryn at 9:45 AM on February 9, 2011

You know what else results in a released accelerator, provided the cruise control isn't on? Falling asleep.

Not automatically true. In trains, they had the problem of an operator falling asleep or whatnot, missing a signal, and plowing through -- or going to pee and hitting something when they blew a signal. Or, heck, they died, then the train hit something.

So, the first generation dead-man pedals. You had to hold something down, if you let go for more than a couple of seconds, the train would emergency stop.

Fixed the whole falling asleep or dying thing, right?

Wrong. It became clear that you could fall asleep, or even die, and keep your foot on the pedal or your hand on the button.

So, the new generation systems require you to hold them down -- but at intervals, they require you to release them. If you don't after a certain amount of time passes -- typically 15-30 seconds -- then an alert sounds, and you have 10 seconds to release, then reapply the deadman. If you don't, emergency stop.

Indeed, the one time I briefly fell asleep at the wheel, I quickly awoke to me doing 90 and accelerating. I couldn't have been out more than a second or so, but my foot didn't fall off the pedal, it fell *onto* it.

I was lucky. Do *not* drive tired. Seriously. I could have died very easily in that second. I'm now religious about awareness and taking breaks.
posted by eriko at 10:08 AM on February 9, 2011

valkyryn wrote: "Ah, but that's different, actually. Many US insurance carriers--and it doesn't surprise me that Canadian companies are following suit--have not driving drunk as a condition of coverage. They're permitted to do this because government regulators let them as a political decision to look tough on drunk driving."

Thankfully, here in Oklahoma, the courts have interpreted the compulsory liability insurance law in such a way that it is essentially impossible to have your coverage entirely canceled after the fact, at least to the extent of the coverage required by law.

I have 100/300/100 coverage, and if I were to intentionally ram someone or drunkenly drive into someone, they would still be required to provide up to 25/50/25 (the present minimum here). Without that interpretation, rescission of coverage would only punish the victim, not the policyholder.
posted by wierdo at 11:17 AM on February 9, 2011

valkyryn writes "That's different than just denying coverage simply because you happen to be partly at fault, as you've said, believe it or not. The 'even if not at fault' part is what's significant there. There are similar conditions related to things like drag racing or using your car as a taxi: do one of those things and get in an accident and there's just no coverage. But all of those things are spelled out in the policy itself, and they all involve some kind of intentional act. The point of insurance is to protect people from the consequences of their unintentional negligence, so limiting coverage for things like speeding--which is a strict liability offense--would never make it past the regulators."

While at the moment no policies punish speeding it would be trivial to add it to policies. Weirdo's example means the technology has mostly already been rolled out; all you have to add is GPS and maps.
posted by Mitheral at 3:19 PM on February 9, 2011

I am more inclined to believe an engineer like Wozniak who is not under government pay...

I'm curious. How would the government get NASA engineers to lie about this and to what end? What would motivate a cover up? How does the ability of the engineers who did this at NASA compare to Wozniak's engineering ability? If NASA engineers can be readily made to falsify their findings for whatever reasons isn't possible that Wozniak could as well for whatever reasons?
posted by juiceCake at 3:33 PM on February 9, 2011

Indeed, the one time I briefly fell asleep at the wheel, I quickly awoke to me doing 90 and accelerating. I couldn't have been out more than a second or so, but my foot didn't fall off the pedal, it fell *onto* it.

I was lucky. Do *not* drive tired. Seriously. I could have died very easily in that second. I'm now religious about awareness and taking breaks.

Seconded. About 15 years ago I fell asleep while doing about 60 down a country road. I only woke up when my car slammed into a tree at what the police estimated was about 75mph -- and this was after I drove *through* a utility pole and severed it at ground level, doubtless slowing me down a bit.

I destroyed the front right of the car. Any passenger I had probably would have at least had his legs crushed, if not worse. It's only pure dumb luck that it wasn't me suffering that fate.

Since then, I am equally cautious about driving tired.
posted by jammer at 12:31 PM on February 10, 2011

And my husband doesn't understand why I hate cruise control and won't use it. If I use cruise control I am terrified that someone is going to pull out in front of me from nowhere, or I'm going to come over a hill and someone will have stalled in front of me, or...whatever. Which means I'm constantly having to move my foot over to cover the brake, just in case, which means it's much more of a pain to drive in cruise than to just keep my damn foot in one place and drive at a steady speed.

And dare I suggest it? Perhaps we need better systems for teaching people how to drive. I had a car-obsessed father and an old school cowboy for a driving instructer who both took the "setting challenges" approach to teaching. "See how fast you can drive around this curvy road without crossing any lines," was my dad's final exam.

But I'm always amazed by how little most people know about driving. Like that you should brake before a turn and then accelerate through the turn to give you more control.
posted by threeturtles at 10:31 PM on February 10, 2011

uncanny hengeman - Sorry, but I don't respond well to trollish comments from seemingly unpleasant name-callers with poor reading comprehension skills. If you're serious, try again, with common courtesy.

On second thought, never mind. Enough time wasted!
posted by Ron Thanagar at 6:58 AM on February 11, 2011