Mr Barr did not "prove" that the truly fatal flaw that "caused" unintended acceleration in the affected Toyotas lay in the software that had been designed into the vehicles' computer systems. He postulated a diabolical combination of multiple failures that could possibly cause unintended acceleration. There is no data indicating that those failures have actually occurred in the real world.
@AeropsaceEngr, respectfully, I disagree. If What Mr. Barr did was merely "postulating a combination of multiple failures," Toyota hasn't actually proven that the sticky pedals or floor mat actually occrred in the real world, either.
But I digress.
You may want to read what B. Benjaminson wrote here in a separate thread:
@Aerospace, If I understand Barr's work correctly, he found a single-point failure mode, not multi-point failures as you claim.
Barr's conclusion that Toyota's software is defective is actually less important, in my view, than his conclusion that their engineering processes are defective. This validates the strong impression I and others have had in reviewing the 230 electronics-related internal Toyota engineering documents. The docs reveal a kind of seat-of-the-pants engineering, where engineering processes allowed designs in which Toyota engineers had to do detective work to find out why vehicles behaved unexpectedly, both during development (which may be acceptable for design processes of non-safety critical systems, but not for safety critical), and when receiving reports of vehicle behavior on the road, in the real world. I think you engineers can agree that a safety-critical machine must be absolutely predictable in its behavior, and the Toyota engineers recognized that the vehicles they were making were unpredictable. At times they used the word "ghost" to describe performance changes that they observed. Is this normal? Is it safe?
Barr's view of the substandard engineering practices that produced substandard designs has been supported in various ways by his colleague Dr. Koopman, as well as by engineers Gilbert, Leidecker, Barrance, Anderson, Rajkumar, Belt, Kirk, Hubing, Pecht, Armstrong, among others, both in the laboratory and in theory, the theory being primarily a comparison between proper safety-critical engineering practices and the practices that become evident in vehicle examinations and document review. The views of all of these respected engineers, taken together, substantiate Barr's conclusion about the engineering.
There is data that these failures occurred in the real world. The data exists but has not yet been published.
It is also plainly obvious that the hundreds of serious lawsuits against Toyota show that actual vehicle performance reveals engineering flaws. These vehicles are unpredictable on the road. That is a defect. As attorney Cole Portis pointed out in his interview on Israel's Channel 2, none of these lawsuits would have even been filed by law firms for plaintiffs if the facts could easily lead to blaming the driver for the death or catastrophic injury.
I will be very interested to learn why the DOJ ignored the electronics.
I think it's almost immaterial that Barr didn't prove beyond a shadow of a doubt that the mechanism he uncovered was the cause of a particular unintended acceleration event. The fact that this failure mechanism (apparently) exists, had gone unaddressed, and that these cars have a higher incidence of UA than any other, should be reason enough to address the problem. Matter of fact, the carpet problem may just be another contributing cause of UA. Certainly, the one doesn't rule out the other.
It could also be that what Barr demonstrated is impossible to occur in practice. If that's the case, it's incumbent upon Toyota to explain why. Speaking as an engineer and not a lawyer or a CEO, I would certainly take that approach, in Toyota's shoes. As hyper-relieved as Toyota must be, with this recent DOJ dodge, I'd still be investing time and effort to either fix the problem or to definitively show that it isn't a problem.
AerospaceEngr, you have a good point...Nothing is proven as far as what actually happened. I don't know if a legal burden of proof could have been met if the case had not settled and gone to trial.
I have experienced the floor mat UA problem on a few occasions (before my dealer trimmed the accelerator pedal). The floor mat explanation is credible, and judging from my anecdotal experience, apparently far more common than Barr's hypothesis. Fortunately I responded by hitting the brake which canceled the acceleration -- Toyota's hybrids have that feature, whereas it is my understanding that their non-hybrid cars do not. I can easily imagine a driver panicking in the same situation, or in the case of a non-hybrid car, not being able to cancel the acceleration by performing the typical reflex reaction of hitting the brakes.
Of course, not having the automatic release of the throttle when the brake is depressed, in the non-hybrid versions of their cars, is unexcusable, but I simply can't agree with those who buy the idea that Barr's explanation is the most likely one for any given case.
Junko: the burden of evidence is not on the defendant; your reply seems to indicate that you believe the opposite to be true.
One fine article, all the more so since it's based on David's firsthand experience. Government unjustifiably protecting corporations is a recurring theme in my blog, Beware of Toyota. Their next victim may be YOU..., and I referenced this article in today's post, "Mainstream media hints at bugs in Toyota's electronic throttle control" http://uc2.blogspot.com/2014/03/mainstream-media-hints-at-bugs-in.html
Toyota whistleblower Betsy Benjaminson had an experience similar to David's when she telephoned the DOJ, brought up Michael Barr's findings, and was asked to spell his name. Betsy discussed what happened in a comment (3/29/2014 3:21:47 AM) regarding Jessica Lipsky's article, "Toyota Pays $1.2B to Settle DOJ Probe" http://www.eetimes.com/document.asp?doc_id=1321525
EE Times is doing a superb job of exposing the facts and taking on the tough issues about Toyota's problems with sudden unintended acceleration. Despite the government's well-orchestrated efforts to keep the public unaware of compelling evidence of electronic issues - I don't for one split second believe the DOJ has been ignorant of Mr. Barr's findings - word has leaked out, and continues to spread. This is the focus of today's post on my blog.
Sad state of affairs when government agencies become mouthpieces for the businesses they are supposed to regulate.
The losers on this one are the informed, but sadly this is a win-win for both parties directly involved. The DOJ gets cash and the opportunity to toot its own horn in front of low-information voters about a sum of money (probably poll-tested) that seems like they've really socked it to the man. (Replay BP) Toyota gets to evade a dissaster that could have costs them much, much more.
I predict "case closed, no comment, nothing to see here, move along-move along, these are not the droids you are looking for" from both parties for a long-long time.
It's obvious to all concerned - including the DOJ - that Toyota's electronic throttle control won't stand the light of day. The Oklahoma case - much to Toyota's chagrin - proved that juries are quite capable of understanding the technicalities, especially when corroborated by 150 feet of skid marks from the plaintiff's tires.
What consumers are faced with is a government - Repukes and Demagogues alike - that unjustifiably protects corporations. It's the best system money can buy - I quit voting over a decade ago.
All this judgement illustrates is the technical illiteracy of the justice department and general technical illiteracy of the population in general.
In addition to the software issue is the NASA paper analyzing a failed accelerator sensor showing whiskers growing from the lead free solder shorting out the pot, indicating a full throttle command. Interestingly the link http://edn.com/article/520531-Toyota accelerations revisited hanging by a tin whisker.php?cid=NL EE+Times is now dead. Using a pot, subject to the type of noise we hear with a noisy volume control pot, is another technically questionable practice.
But floor-mats is something the average lawyer/justice employee/public can understand. Though one would expect that floor-mats would be a universal problem with vehicles. But then again, perhaps it was intended as an out for an appeal.
Politics is involved in every human decision. Like they say, "Justice is blind."
I was reminded, not long ago, of the ferocity and integrity of the prosecutors in the Southern District of New York, by a former federal prosecutor...just when I was also inclined to think that something improper may be happening there.
So if I pretend to be a lawyer, I can imagine that there may well be some perfectly proper legal reasons why the DOJ could not go for the electronics. Speculation alert!! Anyway, one reason could be that the government agency that got the ball rolling for this investigation (possibly DOT or possibly Congress, or who knows who else) did not have sufficient proof of the electronics defects to be able to ask DOJ to investigate the cover-up of those defects. Another reason could be that even if DOJ was tasked to investigate any and every cover-up, they did not amass evidence to a level of "beyond a reasonable doubt" that they need to have in order to prosecute for a crime related to an electronics defects cover-up. Come to think of it, they did make public that there were some issues with evidence being overseas, evidence insuffient for charging individuals, etc., if you recall. And the Bookout plaintiffs attorneys did comment during the trial that the standard of evidence that the law required for bringing punitive damages,--a "preponderance of the evidence," was far lower than the level required for a criminal case, "beyond a reasonable doubt."
So...that may have made the difference between what was effective evidence in the Bookout trial court and what was effective evidence in the DOJ.
David Benjamin, does this make sense to you?
But, even if these speculative reasons are true, they don't change the outcome, and I still feel very concerned about the DOJ settlement's implications for the American public and for auto safety.
If the DOJ lawyers simply cannot understand a defect due to its complexity, they may not have the confidence to prosecute a related crime.
There is an enormous problem that I experienced directly on Capitol Hill. Government lawyers simply do not seem to understand engineers. When I took experts into Congressional staffers, i.e. lawyers, and we tried to explain the documents that contained evidence of Toyota's substandard engineering and its engineers' concerns about many possible and likely electronics-related causes of SUA, we were faced with a serious knowledge/communication gap.
It is quite possible that the DOJ can hire experts, but nowhere near the extent they were required for the source code evaluation. I would not be at all surprised, though, if they shyed away from technical complexities and focused on simple problems.
Where can engineers learn how to talk to government lawyers? We need more who are able.
Lawyers for DOJ, NHTSA, members of Congress, etc. pretty much understand what they want to understand. Here again, the technicalities are not beyond the average person's ability to grasp. The Oklahoma jury demonstrated this fact most admirably. I noticed long ago that the government was trying to mislead the public into thinking folks had to be an embedded systems expert (or a rocket scientist with NASA) to understand the essentials.
We must remember that government officials, presstitute media, etc. all have a vested interest in "failing to understsnd" information unfavorable to Toyota.
A Book For All Reasons Bernard Cole1 Comment Robert Oshana's recent book "Software Engineering for Embedded Systems (Newnes/Elsevier)," written and edited with Mark Kraeling, is a 'book for all reasons.' At almost 1,200 pages, it ...