Slashdot Mirror
How Did Volkswagen Cheat Emissions Tests, and Who Authorized It?
Lucas123 writes: The method by which Volkswagen diesel cars were able to thwart emissions tests and spew up to 40X the nitrogen oxide levels set by the Environmental Protection Agency was relatively simple. It was more likely no more than a single line of code used to detect when an emissions test was being performed and place the emissions system in an alternate mode — something as simple as a software "on/off" switch. Volkswagen AG CEO Martin Winterkorn, who stepping down as the result of his company's scandal, has said he had no knowledge of the emissions cheat, but software dev/test audit trails are almost certain to pinpoint who embedded the code and who authorized it. You can actually see who asked the developer to write that code," said Nikhil Kaul, a product manager at test/dev software maker SmartBear Software. "Then if you go upstream you can see who that person's boss was...and see if testing happened...and, if testing didn't happen. So you can go from the bottom up to nail everyone."
Correction: "You can nail everyone that's in the official audit trail."
The people at the top that authorized it (or at least didn't condemn it) probably never actually sent a traceable e-mail to anyone. Nor did they touch any code. Nor do they appear in any meeting minutes. These sorts of discussions tend to happen over a drink in a bar somewhere, and for good reason.
Clearcase, Git, Sourceforge, or CVS?
It's cute how he thinks no one thought about this and sanitized the audit trail. I'm sure he also thinks his 4096-bit disk encryption thwarts even the most determined ne'er-do-wells.
#DeleteChrome
I *highly* doubt it was a single line of code. To toggle the car's "EPA Cheat" mode, maybe, but by all accounts, the system used a variety of inputs to detect artificial driving conditions (including, apparently, barometer data), as well as needing code to define what engine parameters to change once the mode was entered.
"However, not all companies follow detailed auditing processes. The primary reason, Kaul said, is the speed at which software is being released to the marketplace. It necessitates an "agile approach," resulting in millions of lines of code being worked on and checked into production every minute."
love is just extroverted narcissism
The code was written many years ago, and who knows if the archives still exist. And even if the original programmer can be determined (and the code might have been installed and removed several times), memories of who said what and when are likely to be vague, at best.
when you are out driving around the emission system just gets outside the ideal curve and produces more emissions then when running smoothing during testing.
Same thing happens with gas MPG testing. Real world results don't work the same as standardized testing.
I wonder whether someone actually gave the order to implement a 'test defeat device' or they just started to optimize the engine to comply with regulation and to pass the test and then they went too far.
Then again, something similar might make a nice contest topic.
VW must have thought they were a Japanese company or something, to be allowed to get away with such a trick on the American market. I'm astonished they thought they could get away with this.
Damn_registrars has no butt-hole. Damn_registrars has no use for a butt-hole.
Someone should have leaked this a looong time ago. Perhaps some dev, why not.
Hell it would have saved VW a lot of money! Think about recalling 1mln cars instead of 11mln!
Did VW really think it could get away with this indefinitely?
Fucking corporate morons...
If they get fined $18,000,000,000 who gets that money and what will they do with it?
Most rolling roads don't spin the non powered wheels, so if the powered wheels are spinning and the others are stationary for any length of time its a good bet its having some sort of test. Obviously this isn't going to work with 4WD however.
No flame here, just wondering. In my travels to Europe I haven't found them to be any worse pollution wise than American cities. Are these cars really that bad physically or are we talking goofy government crap?
Just asking.
Peace is easy to achieve, just surrender. Liberty is much harder get/keep.
I have been teaching UML, modeling and systems architecting at several companies that directly supply to the German car industry ( especially to Volkswagen and BMW ). It is the car makers themselves that impose rigid rules and constraints upon software traceability and configuration management. So the idea of
"software dev/test audit trails are almost certain to pinpoint who embedded the code and who authorized it"
is not that outlandish, and following such audit trails may well lead to (at least some of) the culprits.
Religous speak to God. Insane are spoken to by God. When all shut up, one can finally hear Shostakovich in peace
If I were doing it I would have placed "// FIXME DEBUG" on that line of code. Like it was an internal testing mode which wasn't switched off, by accident of course.
I applaud these engineers for their hard work and creativity.
There's not a single actual bit of information about how the cheat was implemented in the article. Can we stop the hype (at least on tech oriented websites) until someone with inside information can actually tell us more about the real details? Not that they matter, but the rest is totally uninteresting.
EPA Cheat Code: Up, Up, Down, Down, Left, Right, Left, Right, B, A
"Never give up, for that is just the time and place when the tide will change." -Harriet Beecher Stowe ^_^
I would really like to see VW's source code, to see if they took the care of plausible deniability. There was recently a nice contest over here.
Like the chemist in the French Connection.
The paper shredders must be working overtime at VW right now.
According to a NY Times article. A billion dolalr fine for Harvesters truck engines. I suspect many cheats not caught.
I envision a scenario where as soon as the shit hit the fan on this, somebody really high up in the company who was in the know on this went into the necessary files and altered the documentation so that they were not incriminated.
File under 'M' for 'Manic ranting'
If the code is all in perl it will be close to impossible to find out which part of it does what, a masterstroke of covering tracks.
Nullius in verba
That was a different issue - auto air conditioners and home refrigerators. But as CFC emissions decline, so doe the size of the annual south pole ozone hole. Still needs decades to heal.
Yup. This is why you shouldn't be afraid to be a bastard with your supervisors over matters of ethics (eg: time reporting, license compliance, government regs, etc.). It's most likely your neck on the line, not theirs.
Fiddling with graphic card drivers so they perform better on benchmarks is a way of life in the computer industry. No sane person would believe that only Nvidia ever does it. One would suspect that all the manufacturers do it (when they think they can get away with it)
I really would be amazed to find that only VW was using special code to pass the emissions tests. I have zero evidence to back up this wild accusation, but I do note that almost nobody ever gets the same mileage as the EPA tests.
None of them can see the clouds; The polished wings don't care.
The purging seems to be happening from the top down with three more execs losing their jobs today. Perhaps they already know who authorized what?
TDIClub has a good summary and a list of relevant articles for those wishing to know more about the details of the technology and emissions violation.
I am not interested in articles about life extension advancements.
It's been a while since I watched my car being tested but do they hook up the car to a computer terminal of some sort? Could those be used to trigger test mode?
The test mode was triggered by monitoring which wheels were turning, position of the steering wheel, etc.
Basically they wanted to avoid the cost of installing a urea injection system so they cheated instead. Honda engineers were reported to be perplexed about how they managed to do this miraculous feat of engineering.
Here's a good article about what is known so far:
http://www.msn.com/en-us/autos...
Maybe there is a line to switch but there is a lot of data to add also. ...) and the output (ex precise injection timing (this is Diesel) ).
The data is used to correlate some input parameters (ex air pressure, air temp
This data is acting like a N-1 surface, and there are a lot of points between which the system interpolates.
So it might be a line to switch the the data pointer from "Regular run area" to "Emission cheat area" but the data segment will be double.
Is not something one rouge programmer is doing over night , with no red flags (used to 12 K now is 24 overnight ?)
What would be a big deal however is the code that detects whether the car is in test bench or on the road. Apparently it uses steering input and other such details. So that code block is the interesting part. Proper audit of the code changes, pull request authorizations would nail the engineer who actually did the dirty deed. But would there be code review meeting/minutes, comments fingering higher management?
This scandal will have some salutary effect in engineers who manage code, they would refuse to merge or pull such cheating code changes because it would leave their fingerprints indelibly for ever. They might even add comments in code covering their tails fingering the actual perp in the higher management.
sed -e 's/Chuck Norris/Rajnikant/g' joke > fact
"Then if you go upstream you can see who that person's boss was...and see if testing happened...and, if testing didn't happen. So you can go from the bottom up to nail everyone."
Oh yeah, like that's gonna happen. Admittedly it would be easier than chasing down decisions in (1) crappy bank and investment fund decisions, (2) not sending aid to the US Consulate in Libya, (3) the Enron California power scam .. and any number of other catastrophes. Actual people hardly ever get punished in these things, and that's too bad.
At least that peanut butter CEO might be getting actual jail time; that's nice.
My strong suspicion is that VW is not the only auto maker up these shenanigans. I'd like to see source code placed in the hands of the appropriate regulators, along with the tools necessary to build it. Then the EPA, or their European analogs, could audit the code, build it, and compare it with the object code in randomly selected test vehicles. Obviously there would be some technical glitches to overcome to get a system like this working, but it's definitely doable and most certainly worth the effort.
I find it hard to believe that any engineer would be that thick headed as to actually record in the audit trail the fact that they designed the system to cheat. That's pretty out there. It's not impossible, but highly improbable.
Make sure to get it in an email form?
How are you going to do that?
What happens is you get called into an office, you raise your objections, your boss says do it and if it's a problem, then maybe this isn't the job for you.
When you leave, PHB then starts documenting your sub-standard work and lack of ability to work on a team.
Your next review shows that you are not a team player, attitude needs improvement, and your work does not meat expectations.
Got it? You are a peon and if you buy into the STEM shortage and think your job is secure, well, good luck with that.
You can't just change the working file because the change would be stored in the version control system. You'd have to dig down into the underlying version control system files and change it there, then remove all evidence that you had done so (which would require root access), then retroactively change the equivalent files on the backup drives, the offsite backups, any private copies of the source code repo, etc.
I wonder whether someone actually gave the order to implement a 'test defeat device' or they just started to optimize the engine to comply with regulation and to pass the test and then they went too far.
Someone in management made the decision to not install a urea injection system which is necessary to keep emissions to legal levels but costs a lot of money. Reportedly something like $400/vehicle. So it seems pretty clear that their "solution" to the problem was simply to cheat. This wasn't a case of optimization gone awry. They flat out knew what they were doing and went ahead with it anyway. As soon as they made the decision to not install urea injection, they effectively decided to cheat at that time because they were asking for the technologically impossible. There is no way they didn't know that their decision to leave off such a key piece of equipment would not result in unacceptable emissions. The engineers at VW aren't dumb. The decision was made for financial reasons (not surprising) but was aided and abetted by a bunch of engineers that should have known better.
The only real question seems to be who made the decision and who was responsible for executing it and covering it up.
I work for a company that has paid out some 30 Billion Dollars worth of fines to the US government. Where does that money go? I think it goes directly into the pockets of well-placed individuals, because we never hear about where that money goes.
When you pay a parking ticket, where do you think that money goes?
If telephones are outlawed, then only outlaws will have telephones.
What is the point of this article, just to speculate on a bunch of stuff that people don't really know anything about?
These are the exact questions the courts will need to answer, and nobody has enough information (and certainly not more information than the courts will get) to have any sort of insight into this. If people just want a chat-room to discuss rumours, IRC is still available..
What gain did they achieve by doing this? None of the vehicles in question are "performance" vehicles, so it's not like they were doing it so that the vehicles had "great performance" when not being tested, yet still passed.
Another non-functioning site was "uncertainty.microsoft.com."
The purpose of that site was not known.
However, not all companies follow detailed auditing processes. The primary reason, Kaul said, is the speed at which software is being released to the marketplace. It necessitates an "agile approach," resulting in millions of lines of code being worked on and checked into production every minute. "And, right now, everyone is saying, we did not do it," Kaul said in an interview with Computerworld.
Kaul is definitely wrong here. Established mechanical engineering giants, especially from Germany, leave extensive audit trails. Come on, guys these people set up tables and recorded the names, addresses and parents of people being sent to concentration camps. Any vendor who sells software to Volkswagen knows how much testing and auditing they do before upgrading the software. Further auto companies operate on very long product cycle. Not like the web operators. It is alleged that Facebook did not even have a source code control system till after it went public!
They are an auto company. Strike one. They are German. Strike two. Even agile development has strong change management paper trials. Strike three. We can nail the engineer who checked in/merged/pulled the change.
sed -e 's/Chuck Norris/Rajnikant/g' joke > fact
It seems nonsensical to me that this is a small change. First, both software developers and the engine engineers have to agree on how the polluting mode is supposed to work. Then the software engineers have to detect, with the help of the people who understand all the sensors, what an EPA test's signature is. Then they have to add that code to the car's software system. It sounds like many groups working together, each of whose management must have agreed to cooperate. And all of those managers, I'm sure, required sign off from their common manager. I bet this goes up pretty high.
I remember when I had a 78 Cougar, that was a bad emitter of emissions, and guys who ran Emissions/Inspection shops and would tell me, "I can get your car to pass". I didn't know what they were doing, but it worked. Or I could just register my car in a county where that wasn't required... Which also worked.
Amazing that we are in 2015, and a huge company like VW would do the same with their "clean diesel" technology.
I always wondered how they were able to do it.
Now we know!
We play the game with the bravery of being out of range
It sounds like Volkswagen DOES have the technology to be clean. So why wouldn't they enable this clean consumption mode by default???
This behavior was known from a test some year s ago. VW did a voluntary callback for a software update. Now this cheat still happens. This is nothing that suddenly happened, this is that state regulator that has draw a line in the sand and said "till here and not futher". Do you think that if some engineer explained that there was a software malfunction if it was strapped in testing? And that the people who took that call understood it was a "defeat device"
Did you see anyone trying to calculate how many deaths this has cost because of small particle dust?
In Europe we still use decades old tests to determine fuel usage (= co2 exhaust). It is known that cars behave different in such test than in real life. why? rules....
I'm pretty sure it takes more than one line of code to (sneakily) check for an emissions test, and a whole lot more lines to implement said cheaty-mode.
The reports so far seem to indicate that they looked for an OBD connection, atmospheric pressure and temperature at a certain level (specified in the test protocol), the steering wheel not moving (because the car's on rollers), the engine being run at certain speeds (to keep somebody from spotting it during a dyno test at a tuner shop.)
That would be one heck of a line...
You get requirements that you aren't sure are a good idea or the right thing to do, and you question them - but PHB assures you that it's all been approved and cleared by the people on higher floors, and you may even contact some of them and hear their agreement.
Then congratulations you are a rube who got used and have now committed a crime. But frankly I have a VERY hard time believing that the engineers involved did not know that what they were doing would violate the law. They wrote code specifically to determine when an emissions test was being performed and to adjust the emissions substantially. The ONLY reason to do that is to cheat a test which they damn well ought to know is illegal. So was it negligence or fraud? Either way they basically engaged in the functional equivalent of dumping toxic waste and should be accountable.
If the vehicles computer is properly designed to reconfigure itself based upon the performance needs of a vehicle, then a scenario where the rear wheels are moving with negligible resistance, the tuned result may be valid.
Write a section of code that purports to do some routine bit of engine parameter control while actually nefariously watching for specific sequence of a parameter (say engine speed) and actuating the EGR valve. Plausible deniability essential.
Why is everyone assuming that there is actually anything malicious or obvious looking in the code? Its a simple problem of industrial controls. You have an emissions target to meet at a specific barometric pressure/temperature/output condition, and you adjust the fuel/air/RPM curves to meet it. If the software engineer was wise, he has an archived email thread between his boss and the department that tests performance and drivability that shows what was likely constant complaining of stumbling and low power output and demanding that something be done. The poor guy simply complied knowing that the official test was one specific point in the fuel/air maps.
"You can actually see who asked the developer to write that code," said Nikhil Kaul, a product manager at test/dev software maker SmartBear Software. "Then if you go upstream you can see who that person's boss was...and see if testing happened...and, if testing didn't happen. So you can go from the bottom up to nail everyone."
I sense a feeding frenzy in law firms all over the country, with lawyers asking themselves, "How can we get a piece of this pie?"
The audit trail might not include everybody, but it'll include enough people so that the ones not included will likely be given up by the ones that are. I doubt that anyone who was involved is going to escape unscathed. And the audit trail at this point appears to include a devastatingly detailed trail of managers and executives who knowingly signed off on this. Some will deny that they knew what this was all about, but some simply won't be able to claim that.
There's no way to make this into an "oopsie" moment or claim it was a simple mistake. This was planned and approved at the highest levels.
Heads will roll, and I expect a river of blood and money will be flowing from Volkswagen for years over this...and that is as it should be. They did this in the name of profits, and now it's time to make those fuckers pay.
Just cruising through this digital world at 33 1/3 rpm...
I did get an email verifying that I had questioned it, but then I found out that all our emails are automatically deleted after 6 months or something like that.
Nothing prevents you from printing emails of instructions to implement dubious decisions. I've done this from time to time just to protect myself when I worked at a large company.
You get fired now, or you implement something dubious - what do you choose?
If it is clearly illegal or will be very likely to cause major problems then you should seriously consider walking. If it isn't so clear then you get them to document their instructions to you and you keep a copy (print if you have to) for your records to cover your ass should it be a problem down the line. Make sure you document your objections and make it clear that you have taken every reasonable effort to ensure that what you are doing is legal. If the decision is merely dumb but legal, same thing but don't worry so much about ensuring legality.
I'm trying to understand WTF happened because currently nothing is making sense. Currently the only fact I am sure of is cars were observed to emit more pollution than allowed.
What additional facts exist that provide evidence of intentional cheating?
Around here emissions testing consists of plugging in an OBD-II cable waiting a few seconds and being charged $30 for the whole ordeal.
The articles say things like "For years, millions of Volkswagenâ(TM)s deisel cars contained software that turned their pollution controls on only when the cars were being tested by regulators" ...
Even TFA has no clue how the triggering works it just assumes it must be triggered maliciously in response to detecting an emissions test and dreaming up "single line of code" because it sounded good to the author. What is the basis for these beliefs aside from imagination of the authors? Has a specific reproducible emissions testing trigger been found?
What if there was just a glitch in a fuel trim or something that presents after driving around for awhile? What actual facts exist that provide evidence this was an intentional malicious act?
Make sure to get it in an email form? How are you going to do that?
Easy. You start asking questions in email. Act innocent or ignorant if necessary. Unless the person is ordering you to do something obviously illegal (in which case you should quit or get legal representation) they aren't likely to act all cloak and dagger. If they do, do you really want to work for that person anyway?
Maybe their code is in Perl.
Slow down, cowboy! It has been 4 hours since you last posted. You must wait another few hours.
Once again, the German propensity for record keeping is going to be the death of them. The last time was, I believe at the Nuremberg Trials.
Probably because if the "testing mode" were enabled all of the time, the driveability performance would be lousy. The car would probably have no acceleration, and would have other issues. It would probably burn so lean that it would burn up pistons, for example.
I worked at Chrysler in the 1970s when they were first having to work on "pollution controls." Congress, which is almost all lawyers who are ignorant of the laws of physics and chemistry, mandated that certain types of emissions be sharply curtailed. Specifically, they wanted to cut back on carbon monoxide, unburned hydrocarbons, and nitrogen dioxide emissions. When the emission controls were working, the tailpipe was emitting LESS carbon monoxide than was in the ambient air! Wow! It looks like it was cleaning the air of carbon monoxide! However, as some folks with a scientific bent know, you cannot create or destroy matter. If you have 78% nitrogen and 21% oxygen going into the carburetor (or fuel injection system in modern times) (plus other things, like sulphur), you MUST have the same percentage of those elements leaving the tailpipe. So, while the cars would reduce the amounts of the mandated stuff, they would put out other compounds that hadn't been mandated against. In particular, if your car wasn't in good tune, you could emit phosgene (a poisonous gas used against enemy soldiers in World War I). Theoretically, catalytic converters were supposed to convert engine emissions to carbon dioxide (CO2) and water (H2O)--but what about the 78% N2 (nitrogen)? The nitrogen still gets combined with other stuff and gets emitted.
So, seriously, if you are going to burn hydrocarbon fuels, you are going to get something out of the tailpipe, and not all of it will be agreeable to you.
BTW, since I worked at Chrysler, I learned how to defeat the emission controls. They had an Exhaust-Gas Recirculation (EGR) valve which returned some engine exhaust back to the intake manifold to attempt to burn any gasoline that made it through the engine without getting fully burned the first time. What it really did was cut your gas mileage. You could put a BB in the rubber vacuum hose to that valve, and cause it to stop working. The other, more nefarious device was the "Orifice Spark Advance Control" (OSAC) valve which was inline with the vacuum hose between the intake manifold and the distributor. The purpose of the OSAC valve was to stop you from being able to stomp on the accelerator which opened the throttle wide open, also quickly lowered the vaccum in the intake manifold, and advanced the spark coming from the distributor. With that OSAC valve working, if you were on a two-lane road and needed to pass slower traffic, you couldn't do it. Rather than have your car immediately kick down into 2nd gear (auto trans) and immediately speed up, instead it would slowly gain speed. Totally unsatisfactory. So, you could rearrange the rubber vacuum hoses. Instead of a hose going from the intake manifold to the OSAC, then another hose from the OSAC to the distributor, you would change the hoses so one hose went directly from the intake manifold to the distributor, then rout the other hose out of the OSAC back into the OSAC. And since both hoses were taped together near the OSAC valve, one couldn't tell by looking that the hoses had been rearranged! After that, you could safely pass again.
Rusty
In practice, if you've got mouths to feed and not enough savings, that's a difficult proposition. In addition, you might get blacklisted and will then be unable to find another job.
In reality, if it's not a matter of life and death, it seems like you should do the best you can do within the confinements that are placed on you.
No matter what the chain of reporting is on the bottom, the CEO is on the top and thus responsible.
It's a tradeoff between NOx production and performance, lower CO and CO2 production, lower particulates, lower maintenance and better mileage. Volkswagen probably figured that since worldwide vehicle NOx production is several orders of magnitude below that produced by lightning, nobody would care as long as all the other numbers went in the right direction.
Have gnu, will travel.
In business, if you have to ask "why?", the answer is "money."
The article mentions something to the effect of looking at VW's testing audit trail but I doubt this will be effective. There may be a test case which reads something like, "verify that emissions test mode produces expected values", which obviously would have passed. I doubt there is a test case to cover the contra-positive case, since there likely isn't any such requirement that reads, "when vehicle is not in emissions test mode, do not reduce performance to meet emissions regulations". It's also laughable that they're trying to pseudo-blame this on agile programming, as if agile methodology itself prescribes not referring to work items when checking in code. I love it when a non-technical (non SW) author tries to comment on such things, makes for a very amusing read.
Hey!! That sounds remarkably similar to the pattern I use when I am flicking my tongue on your mom's clit.
Have you been watching me?
The real world is never that black and white.
I've been in the real world a long time. Frequently the real world it is actually that black and white. Not always but quite often. This VW thing seems to be pretty clearly in the black and white camp. It is REALLY hard to imagine a credible scenario where the engineers who implemented this were not fully aware (or should have been) that what they were doing was against the law. They (almost certainly) knew it and they went ahead anyway.
They use the python emissioncheat library.
Professional Engineers have the power to say no and they have Ethics rules to fall back on.
Who knowing coded it?
Who authorized it?
Who made it a requirement?
Who knew and said nothing?
Did VW charge consumers for that technology in the price of the car?
Does it mean now that millions of VWs are not legally drivable?
All those registrations based on successfully passing emissions are now null and void?
Somebody knew, somebody high up knew, but I rather doubt that everyone on the engineering bench knew, and that means that they had to be fed plausible stories along the way.
Spare me. The engineers were the ones that eventually spilled the beans on this. They weren't fooled by some clever management strategy. They knew exactly what they were doing and they knew or should have known that it was illegal. While maybe not every engineer involved knew, more than a few certainly did without question. The engineers at VW aren't dumb. I know a few personally. Please stop with the attempts to find clever ways to not have to blame the engineers who were guilty of helping to commit fraud. Management may have ordered the crime but the engineers were the ones that carried it out.
This is bigger than "the guy who wrote the code" because an ECU ROM image is comprised of two parts, really... the program code (which obviously had to be altered to detect the conditions for switching emissions controls in) and the calibrations - a symbolically mapped area of constants used to tune the controller for each individual platform and for regions that platform is sold in.
Calibrations are a checksummed block of the ROM image, mapped and edited through a calibration editors. I've actually architected one such calibration editor in the late 90s for GM (Cal-Link) and I've seen the competition (oddly enough, German-written "ETAS" software).
What has to be the case, is that there are at least two sets of calibrations - one for "emissions cheat mode" and one for "performance mode" - and calibrations involve a team of engineers, testing on a dynamometer, on the proving grounds tracks, or on the bench; at each phase, they are checking against emissions testing and performance parameters, tweaking and tuning until they get an acceptable, marketable product.
So to make this cheat happen, across not only individual product lines, but badges as well (Volkswagen AND Audi), you are talking a large number of teams working on multiple sets of calibrations - whose purpose must be obvious the first time they put it on a dynamometer. They are working with an ECU whose code was written by another party not even in their food chain (one controller, lots of products... and every car platform is tuned differently).
I have no idea how high this goes, but it has to be somebody with oversight of not only multiple product lines, but both badges, too.
One team wrote code to detect an emissions test.
There is no need to write special code to detect an emissions test. The cars emissions should meet legal standards at all times. The moment someone asked for such code alarm bells should have been going off right then and there. Especially given the not-a-secret decision to not install a urea injection system. These aren't engineer who write code without any knowledge of the effects of their code and they don't work by themselves and never talk to others. Plus there appears to be evidence that the engineers were WELL aware of the problem because when it first came up they engaged in all sorts of delaying tactics.
No, there were engineers at VW who knew what was happening and it wasn't just one guy with a beard.
I can imagine multiple groups each working on a set of engine parameters: one for cold weather, hot weather, baseline, performance, clean, towing, etc. Each would be creating, essentially, a subroutine. Each independent from each other as it takes time to fin tune engine parameters and the ship date is getting closer! Then one person would put in one line that said " if NoSteeringMovement and FrontWheelsNotMoving and BaramPressure=x and RPMSteady then load Clean. The other groups would never know what happened.
The submitter really does not understand what is going on at all and how these things work.
First, you need to actually have all the sensor readings and detect the test situation. You may use wheel revolutions, GPS, acceleration sensors, etc. For that alone you will have a few hundred lines of code. Then you need to very reliably detect the test situation, giving you at least another hundred code lines. And then you need to make very sure things are not obvious like a strongly different engine sound, vibration, or accidental activation or deactivation under test, etc. For that you need extensive tests with each car this goes into and more optimization and even more code, some of it specific to the car model and engine and even the gasoline quality expected in a country.
The problem is this: If you have even only 1 in 1000 cars behave differently during test, or if you have too many activations of this mechanism in non-test situations, the presence of the mechanism will become obvious relatively fast. That _must_ _not_ happen.
This is not "inserting a single line" at all. This is more like initial development and test of the mechanism with something like 10 experts working on it for two years and then continued maintenance and adjustment to different cars keeping said 10 experts busy permanently. Say 10 Million for initial development and then another 5 Million per year.
This was not a small project. As this cannot be done in the usual hierarchy, the team responsible likely had one dedicated lead and that lead must have reported directly to top management, i.e. those guys now claiming they knew nothing.
Most ACs are not even worth the keystrokes to insult them. Be generically insulted by this and ignored otherwise.
I find it odd to hear how programmers seem so abused by PHBs. Maybe it's an American thing, but in the UK, I've always found that employers want to keep hold of skilled people like programmers, because new ones are hard to find and take a while to get up to speed. This means that saying no is always possible.
(Nothing to to with official engineer status and ethics. There's no general requirement for engineer certifications for programmers here.)
I worked for a Small software house that made SAP type ERP software before SAP ate the majority of the market. This was 1998 or so...
We had a customer come to us and ask for certain modifications. Then a few more. Then a few more.
Not unusual, we made a lot of money from change orders. So the first few were done. All were acceptable in the Generally Accepted Accounting Practices guidelines.
Somewhere along the line the GAAP accountant realized that this last modification set would, taken in combination with all the other mods, make a check disappear from the system and become untraceable.
We refused to do it, and the customer dropped the product, saying we were too hard to deal with. A million+ of revenue were lost, no small amount for the company.
That customer? MCI Worldcom.
They clearly had picked apart the source code and found the edge case that triggered the behavior. I had left the company before MCI blew up, but my understanding is that they were called to give testimony/evidence in the trial.
This could be the same thing, a series of unrelated changes that trigger a diagnostic mode when hooked up to the test equipment.
If so, it would be very hard to trace who made the ultimate decision to do this, as it might be spread across many teams working independently.
Never answer an anonymous letter. - Yogi Berra
Intel does it. Apple does it. Nvidia does it. Microsoft does it. And now VW is about to be punished for it- despite the fact that a German company is merely copying SOP of every significant American tech company.
BENCHMARK MODE means a purposely created path designed to make a gizmo look better than it is in practise when tested by an 'authority'. Every Intel chip, for instance, runs vastly faster for a minute or so BEFORE thermal throttling kicks in. Intel pays the companies that write benchmarking software to craft tests that complete BEFORE the Intel chip goes into thermal throttling. Intel pays tech sites to use these benchmarks. Conceptually, this is IDENTICAL to the stunt pulled by VW.
BENCHMARK MODE is so ingrained in American engineering, it isn't even seen as cheating any more. PR companies, paid hundreds of millions by the big players, have successfully legitimised BENCHMARK MODE shenanigans via decades of propaganda.
And BTW, since VW represents the BEST of diesel engineering, does anyone here think any of its competitors make 'cleaner' engines? This 'green' shit is all one gigantic JOKE. Once the emissions bandwagon grew large enough (and created so many insanely well paying enforcement jobs), it was bound to move beyond true usefulness (getting rid of lead and carbon monoxide) and move to faux concerns like the oxides of nitrogen. These neo-liberal organisations of 'social concern' always need new targets to justify their existence and growing power. Like the witch burners of old, they always need new witches to accuse and burn.
"With one line of code you can break down how it happened," Kaul said. He described an "'if' statement with two clauses: If you do this, then do that. If something doesn't happen, do this."
Shit, I already knew that, and I only know some basic programming. Wake me when someone explains some technical details of how the engine ran in test mode and in real world driving.
as there is a perfectly legal reason to detect an emissions test -- that the traction control and stability control doesn't go crazy.
The caveat is that the car cannot distinguish running indoors vs EPA check. Thus it might be a (very good) safety feature: if computer detects that the car is running indoors or in any other situation where ventilation might be a problem (e.g. your father's garage, bad tunnel traffic, cart in a warehouse) engine is switched into cleanest running mode possible to avoid sending hapless mechanics, who screwed up their ventilation setup, to a hospital.
And that's it. This theory does not require neither malice nor stupidity. It also has an explanatory power - it explains why there were no whistleblowers.
Or just:
if DoorAjar EPAon fi
It's fascinating to see how many posters here automatically assume that it must be the PHBs who pressured the engineers into this. Very few assume that the engineers saw an opportunity for a bonus or for the PHB to owe them one, and added the cheat function voluntarily. I've not seen any posts so far that suggest an engineer thought of the cheat and suggested it to a PHB.
A reminder that we tend to think of our peers as being much more ethical than "them" and look for reasons to think of them as victims of force or circumstances, and assume that "they" are only motivated by sheer callous greed. Whoever the "them" is.
I read the "single line of code" editorial as a distraction away from what matters: accountability and prevention.
Accountability can come in the form of lawsuits from affected car owners and those who can show the subsequent environmental harm caused a problem for them. Letting VW negotiate its own fate is ridiculous and, if the government's action with GM is any guide, unlikely to result in more than a slap of the wrist.
Prevention must also be dealt with, and strongly copylefted free software licenses will help here. Whether this was the result of a mistake (VW's years-long negligence) or planning (VW's years-long fraudulence) is a detail as far as prevention goes because either way VW should be freeing the complete source code to the cars and providing complete specifications for any code it cannot provide so as to allow the easiest possible reverse engineering. Any cost of purchasing code for freeing should be borne by VW.
VW is not in a position to dicker here. I don't buy the excuse of uncooperative upstream providers VW depends on for their code and the public shouldn't either. The stakes (our health) are too high to settle for less than complete corresponding source code under a strong copylefted license so that any published improvements are also free. Keep in mind, this is code car owners should have had from day 1 under a free license so they can fully own their own cars, taking code to experts they trust just like many take their car mechanisms to garages they trust to get fixed. Trusting the market got us where we are now, the market apparently will not grant us the freedom to let us help ourselves and our air-breathing neighbors by fixing the defective VW cars already out there since 2009 (over 480,000 of them). Not buying VW reaches the same conclusion. Not recognizing software freedom for its own sake and the preservation of that protection in copyleft will increasingly become a matter of life and death as we entrust more of our daily functions to software.
Digital Citizen
Mortal Kombat 3 secret combo move actions. How did you....
by TheSpoom (715771) Uncaring Linux user here. I have nothing to add to this but please continue. *munches popcorn*
>> up to 40X the nitrogen oxide levels .... something as simple as a software "on/off" switch
Like this one :
https://www.holley.com/product...
aaaaaaa
Germany won't allow it. The automotive industry is Germany's biggest export; they'll want this handled swiftly and satisfactorily and get things back and running before it causes too much damage to the economy. Sacrifice the CEO, knock down a few low level scapegoats, a new management comes in to make it better, and in a few years VW is back selling cars. The German and frankly the European economy cannot afford to have this drag out for long, so there will be no top to bottom audit trail on this.
In the '70's, it was usually possible to tune a car to pass emissions even if it had something wrong with it. It might drive like crap, but it would pass. Then, after it passed, you could tweak the tuning slightly to make it drivable.
B, A... talk about missing the G-spot.
Did you ever look into these US emission standards and tests?
Look at what they do to make the cars comply....the make then less efficient!!!
I remember in US light trucks it used to be a fantastic Cummins 6.0 liter diesel engine. Now they changed it to 6.7 liter, with less power output that burns more fuel. Well freaking DUH! If I get 15% less power out of that fuel I burn and I need to then burn 15% more of it, please explain to me HOW IN THE HELL this is net positive for emissions? Sure the engine produces less emissions PER GALLON, but you end up burning MORE GALLONS! !!!! GENIUS !!!!
Typical government retards at work. All this crap does is protect US big oil and US automakers.
PS - I love driving my VW diesel....here in Europe.
Und piss in da exhaust.
And I find it hilarious that everyone here states that programmers and engineers work without requirements or documentation. I've worked places where the verbal meeting would have the engineer agreeing with everything, then when it's not written in the requirements document and signed off by 10+ people, it doesn't get built. Seems like all the programmers on Slashdot have never worked in a company larger than 10 people.
Learn to love Alaska
Thats for noobs, IDKFA FTW!
I like how it was discovered. We're in West Virginia, and need to evaluate tailpipe emissions, so we get some nice cars and spend lots of time driving up and down the coast of California. Sounds like a good team to be on, but then again, maybe driving around in California wasn't so pleasant. At least the project produced good results.
Beware of the Redittor who loans you a Sharpie.
This kind of knowledge is why I read /. even after all these years. Thanks for the insight.
I think we can all agree that this whole VW fiasco clearly shows us that test-first development is a bad idea.
"There is more worth loving than we have strength to love." - Brian Jay Stanley
Because it is always the bosses who pull this kind of shit? We've all seen it and been a part of it many times. Bonuses to engineers? Those are paid to management, the workers get scraps, if any.
Shutting down free speech with violence isn't fighting fascism. It IS fascism!
Nah, I've done it half a dozen times. It worked the same way every single time.
Boss: I need you to do this thing.
Me: I don't think that's legal. But I'm not a lawyer, so you can just give me a signed piece of paper telling me to do it, and I'll do it.
Boss: I'll get back to you on that. (leaves my doorway, proceeds to next doorway)
Boss to other coder: I need you to do this thing.
Other coder: Uh, OK.
EVERY. SINGLE. TIME.
Google's huge problem with their software collecting locations of routers (legal) also collecting data transmitted by these routers (illegal) was apparently due to an engineer who thought it was a good idea. Cost them many millions plus a huge amount of reputation.
I've seen it happen quite a bit on Federal contracts believe it or not. Things get in a crunch, can't disappoint the customer (feds) that keep asking for things verbally, scope creep..and well, things just have to get *done*.
Light travels faster than sound. This is why some people appear bright until you hear them speak.........
I find it odd to hear how programmers seem so abused by PHBs. Maybe it's an American thing, but in the UK, I've always found that employers want to keep hold of skilled people like programmers, because new ones are hard to find and take a while to get up to speed. This means that saying no is always possible.
(Nothing to to with official engineer status and ethics. There's no general requirement for engineer certifications for programmers here.)
Pure exaggeration. Developers who don't want to make a change, feature, etc always blame their manager for making the decision. Product not going the direction you want? It must be the PHB's fault.
Volkswagen has a retention policy that automatically deletes any documents older than about 3 years old. We lost a lot of system documentation when it was implemented sometime around 2007. The genesis of the "defeat device" has to be from more than 3 years ago.
The Underhanded C Contest would be a better place for this. And actually; I would enjoy seeing that contest.
Lawyers have higher ethical standards -as a point of law and professional certification- than just about any other profession.
Just sayin'...
"Ethical standards" are often considered low-grade toilet paper by folks wanting to make money/impress their boss/get promoted/not get fired/etc.It really doesn't matter whether or not there's paper that says you should be held to higher ethical standards. I'm aware of quite a number of PEs that can't get work. Not sure why, but they get cool license plates on the Corrolas they drive around to interviews.
In America, we don't believe in using American talent. That's one reason for out-of-work PEs. H1-B workers are usually little better than serfs.
Nor anywhere.
Programming evolves too fast for any kind of certification not even mentioning degrees.
Here in the US, programmers are viewed as fungible, because there are 10-100 qualified candidates for any dev job that pops up. Combine this with offshoring, consultant companies like Infosys or Tata, and H-1Bs, and pretty much programmers are viewed as a disposable asset where guys are rotated out after 90 days.
Of course, code quality is ass... but if you look how code quality is across the board in the computing industry, the only good, bug-free code being produced is by the malware organizations.
When payment is involved, I've only ever seen it happen with change requests, well documented. But then, I've managed to avoid government work.
Learn to love Alaska
Professional Engineers have the power to say no and they have Ethics rules to fall back on.
But do they have the power to say no and keep their job, and keep their job without management making their work life miserable?
It completely ignores the fact how it knows to enter into this alternate mode is complete speculation. Worse, other than that it's an alternate mode, nothing else is disclosed about this mode. The line of code in question is a simple If statement, along with the associated flag that they do not know how it was flagged.
How they used to get "your car to pass": put another car on the testing stand that actually does pass.
How is the Riemann zeta function like Trump rallies? Both have an endless number of trivial zeros.
The premise of the original article is wrong. The whole point is, that the emission tests are meant to be cheated. You pass the test, no matter how, and the car is ok to be sold. The rules are written in a way that you follow them to the letter and everything is ok. It doesn't matter that you don't follow the rules how the voter thinks the rules are meant.
And volkswagen did a professional mistake in failing to adhere the rules to the letter. That's all.
All car manufacturers are "cheating" here. They are just lucky that they managed to adhere to the rules to the letter. Maybe we'll find others failing too in the next weeks.
This is all bad and immoral, but that's the way it is at the moment.
The rules are a result of political actionism. The politicians look good for in fact doing nothing. Additionally there is corruption that the local car manufacturer gets things done cheaply while making it hard for foreign competitors at the same time.
It's an American thing.
Employers here don't care about holding onto skilled programmers or other skilled people, because PHBs think they can just hire replacements on a whim.
Yes, in reality new ones are hard to find and take a while to get up to speed. The PHBs will even acknowledge this when they're trying to hire.
But once they have one employed, they don't care about keeping him happy, because they think they're al interchangeable cogs.
If you're seeing a giant disconnect here, yes, there is. This is how American corporations think; it makes no sense at all. I can't explain it. It's the same phenomenon where corporations will give a big salary offer to a new engineer, but once he's employed there, they'll just freeze his salary or give him paltry CoL raises, while giving new hires even bigger salaries, causing employees to switch jobs every 2-4 years (in Silicon Valley, it's 12-18 months).
Seems like all the programmers on Slashdot have never worked in a company larger than 10 people.
You'd be surprised; a lot of engineers and programmers are just like this, and have only worked in very small companies where there's no documentation of requirements and a completely ad-hoc development model. I've even worked in workgroups in very large companies that were like this; the company might have tens of thousands of employees, but the workgroup might only have 5. There's no formal development going on in a group like that.
Most people who get to write actual production code for cars have absolutely no clue how the cars need to behave. This is decided by people who have absolutely no clue of software. Therefore, it is highly unlikely that someone writing software had the idea to intentionally implement some kind of cycle detection to cheat something like emissions, which are completely foreign topics to the implementer. If there is any code explicitly dedicated to this, there is some documentation existing, necessarily, otherwise it wouldn't have made it into the software. That's the unlikely scenario though.
Software itself is only on part. Cars are all different, change every day during development, and people without much knowledge of software do want to contribute to define the behavior. To facilitate this, control units dedicate significant memory to store data these people can modify. Traditionally this has been mostly data to describe the transfer from inputs to outputs in a flexible way (when the inputs are the same, the output will also be the same), which doesn't allow much cheating. Complexity is rising, and control units are no longer stateless. Of course, also that behavior is tunable. What may have been intended/described as some innocent function, may turn into a defeat device with clever choice of data. Then, of course, documentation of the software wouldn't yield any evidence. Documentation of data is typically much more limited.
While it is likely this was a sin of commission it remains plausible that no one did this at all. My thinking is that if instead of being programmed explicity the computer program was allowed to train itself for it's emission and performance tuning that a very natural outcome would be for it to learn to minimize emissions during emission type testing. Then on the test track it would learn performance and handling. etc... and so you end up with something that cheats but no one told it to nor was anyone even trying to make it cheat. It's just the result of getting what you optimize for.
One reason that I like that theory is that if you consider the opposite, that it was a conspiracy, then this is not the sort of thing you can keep secret easily. You might succeed but that's pretty hard especially considering the time span and the inevitable entry of new personnel and suppliers into the supply chain. So I don't think this was intentional. The exception might be if if it's a conspiracy of one. that for some reason there was just one guy who could pull off everything. THen you would have a shot of keeping this secret.
Some drink at the fountain of knowledge. Others just gargle.
While it is possible that it is widespread and I have just had the luck to never experience it, I tend to doubt it. Of all the issues I have taken to management I have only been told to "sit down and shut-up" twice and both of those cases were terminated shortly after (who knows if my input was heeded or not).
In my opinion, for the last couple generations we have been breeding a culture of victimization so you tend to see these things talked about way more than they are experienced. I have seen junior engineers burst into tears when mildly reprimanded and senior engineers fly into full on persecution complexes when they think their credibility is on the line.
I have yet to see a question of legality brushed aside without consideration. I have yet to see a real ethical question (forget the "cause heads" I am talking about fraud) not at least punted around a few times.
This is Germany, right? Back in the olden days the culture was if someone on high told you to do something (turn on the gas) you did it no questions asked. Maybe that culture still exists (following directions without asking questions).
France TV did a program about it few months ago, they tested random cars and NONE passed limits when tested properly.
https://youtu.be/5JFprj6v37Q
Who logs in to gdm? Not I, said the duck.
Japanese companies build a boatload of cars in the US with non-union labor, and VW actually encouraged their employees to join the UAW .
Never let a lack of data get in the way of a good rant.
I don't know what kind of places you've worked but programmers DEFINITELY work without specs all the time. Often I have been tasked to write my own specs from a brief 30 minute meeting. So I'm expected to write the "Functional Business Requirements" then the "Technical Specifications" and then have to code it and then write test plans for QA - well if there is any QA beyond my own testing.
Requirements and Documentation have mostly been seen as a pain in the ass that no one wants to do. It takes time, has no obvious immediate return on investment and besides we can always fix it later, later meaning when a customer is screaming at you that the system is down because of some stupid reason and it doesn't matter that I didn't put the bug in there. It's suddenly my problem and apparently my fault at 3 AM on a Sunday.
I actually try hard to document my shit because I know what it's like looking at something I coded 2 years earlier and asking myself "WTF was I thinking at the time? I'm sure I thought it was very clever but now I can't make sense of it and since I'm looking at it I may have a bug in it." /* You should probably comment this function but ain't nobody got time for that */
It gets even worse when you're looking at code that a dozen different people have worked on over a number of years and the only comment in the whole fucking program is a huge banner-type comment at the top stating the name of the file, the original coder's name and if you're lucky an incomplete sentence vaguely stating what the purpose of the program is. The variables are all named x,y foo and bar.and there are goto statements cluttering up this vile serving of pasta-like-code.
Someone thought it was humorous to label their error-handling HELL so they could write shit like:
if (errno) goto HELL;
I should be grateful there was even error handling - a lot of people don't even bother.
I actually had a boss look over my shoulder once and saw I was checking for error codes after SQL statements and I shit you not he said "I hope that's just for testing purposes".
He was an accountant turned programmer and his philosophy was to hide errors. Don't bother the users with that. It was a company that managed people's 401(k) accounts. Think about it - he would rather hide errors with people's investments than admit that his code might have a bug.
Maybe I've just worked at shitty places but I find it hilarious that you think programmers actually work with requirements and documentation. Well, not hilarious actually - more like delusional.
The problem is that layoffs have not just become not just a business need to cut budgets, but it becomes a mark of honor for a manager to fire people. Being the company axeman helps come their performance and raise time, and keeps that manager from getting the pink slip themselves.
Because of the extreme low threshold of what it takes to get fired, it isn't that people are wimps... but that if there is even a hint of trouble, they go into full panic mode, re-up their LinkedIn subscription and start pounding the pavement... because it is pretty likely that they are going to be ousted... not because they are bad employees, but because it just looks good on the books.
Working as an IT manager, firing people (especially for cause so they can't get unemployment) is like popping undead in a MMO... exp points, pure and simple.
It isn't a "victim culture". It is a culture where employees are treated like garbage, and they have to survive, so even a "word to the wise" meeting translates to "we are firing your ass, as soon as the PO to Infosys gets cut, so better start looking for your ties again." Sad thing, that is usually the case.
I've never worked in a company as small as 10 people but I have worked places where I was the entire IT department.
I've also worked for one very large company.
Documentation in general is largely ignored in my experience. One big difference may be whether you work for a software company or a business that only sees the IT department as an expense that doesn't directly add to the bottom line.
I've seen many half-hearted attempts at documentation - policies where someone generated a form or a change management system which were good ideas to do but very poorly executed in practice.
I've only worked for one software company where IT was their whole business and they did the best at requirements and documentation of any place I've worked but even they were usually too busy to do it well and when times got tough their one technical writer was let go along with others of course....then they actually realized they needed a technical writer and she was rehired a couple of weeks later.
Did they really expect developers, salespeople and managers to write that stuff? I'll write a technical spec if need be. I'm actually glad to do it but if I'm not required to I'll probably just scribble down some notes and think about it for a while. You can make me put it in a formal document if you insist but even though it may make perfect sense to you and me because we've been discussing this project for a while in 5 years someone may look at it and it will all seem like gibberish.
Dammit Jim, I'm a programmer, not a writer!
Seriously - I have an engineering degree.....I think I only took one literature class in college and absolutely no writing classes.
Best explanation I've seen yet. Although I'd add it's probably a mix of both.
That the emissions team only trained it for the emisions test environment. Not bothering to train "real world" emissions.
This is reinforced by the "up to 40x" I've seen elsewhere. I wonder if they brought in new testing standards recently...
I've been in the business since the 1970s, and have never been asked to do anything like this. The worst I've had has been Marketing attempting to rush through QA because the product was a year late. They did not succeed.
Google's huge problem with their software collecting locations of routers (legal) also collecting data transmitted by these routers (illegal) was allegedly due to an engineer who thought it was a good idea.
Fixed it for ya!
Who?
Who?
This wouldn't be a plug, would it?
Confucius say, "Find worm in apple - bad. Find half a worm - worse."
+1. American Thing. One can request a vacation day, but then I might get all the vacation I want (fired). One can do the right thing, but then I'm sure to suffer the consequences at the next performance review, and so does my role and salary. Demotion is very real, for very non-real reasons. I don't care what laws exist, or Human Resources policy. With a cost of living /far/ exceeding any reasonable salary, and months if not years before one can reasonably find gainful employment before a foreign national at a fraction of the fully loaded head cost, that job is key. There is no opportunity loss or retrain line-item on the bottom line of the P&L, no one cares.
make them flip that switch to allways on? fine them but the cars are legal again. Customers may have a class action suit they dont get the performance they were sold still.
Global warming scientists have been fudging and manipulating data for years.
I'm surprised more people aren't making the point of the implications of this for german us relations.
Definitely this - it stems from hiring PHBs that come from the "business world" to manage engineers rather then hiring managers from the ranks of engineers. Because obviously you need to have an MBA to be a good manager.
Argh. The laws of science be a harsh mistress.
No article has told me what would have happened to performance/fuel economy/WHATEVER if VW had simply let the clean emissions mode always run?
This seems like a fundamental aspect of the story to me.
tone
tone
I find it odd to hear how programmers seem so abused by PHBs. Maybe it's an American thing, but in the UK, I've always found that employers want to keep hold of skilled people like programmers, because new ones are hard to find and take a while to get up to speed. This means that saying no is always possible.
(Nothing to to with official engineer status and ethics. There's no general requirement for engineer certifications for programmers here.)
In the UK, non-competes are illegal and good talent is not willing to put up with abuse.
Calling someone a "hater" only means you can not rationally rebut their argument.
I don't work for VW, but I have worked many years developing software for major automotive OEMs. It's funny how often I read here in Slashdot that this detection was probably coded in one line, or in a few lines, etc. Slashdoters need to realize that most automotive powertrain software nowadays is developed using model-based software, such as Simulink. The mechanical engineers developing powertrain software have literally no clue about source code, operating system, real time systems, interprocess communication, bus latency, code generation. And they do not really need to know either.
So the proper statement should be something along the lines "this detection was probably a single switch block or a single if block in your model."
Though some of us have lucked out in retrospect, e.g. lots of "fruit company" stock.
> In the UK, non-competes are illegal
Blackballing usually is not, because it can be very difficult to prove. Getting a new position, especially one that requires a visa, for an employee whose previous or current employer will not give positive recommendations can be very difficult.
It _is_ a good idea. It's invaluable data for tracking individuals and identifying valuable marketing and personal information.
> In the UK, non-competes are illegal
Blackballing usually is not, because it can be very difficult to prove. Getting a new position, especially one that requires a visa, for an employee whose previous or current employer will not give positive recommendations can be very difficult.
True, but blackballing is very difficult to do with professionals. All it takes is one company not to go along with it and the whole thing falls apart.
This is why blackballing is reserved for the worst of the worst. A good coder, engineer or scientist who is a little bit difficult to work with is not worth trying to blackball.
Also companies can be sued for deformation, so deliberately trying to blackball an ex-employee is dangerous. The worst thing a company can safely say is "I would not employ this person again".
Getting a new position, especially one that requires a visa
Any position that requires a visa or sponsorship is going to put the employee over a barrel regardless. Its a crappy situation for anyone and generally is a last resort when you cant get a visa yourself. For Australians it's pretty easy for us to get working visas in the UK (Youth Mobility visa, Ancestry visa, I know the latter can lead to an ILR), for a non commonwealth country it is a lot harder.
Calling someone a "hater" only means you can not rationally rebut their argument.
It's possible, but the EPA laws specifically ask you to sign off on what sensors alter smog conditions, so if it was some rogue engineer, then there was certainly a lack of due diligence by whoever signed off to the EPA.
Depends. The EPA laws say you can't use sensors to alter smog output AT ALL unless you report it first to the EPA with justifications and get sign off on it. So yes you could have a learning system, no problem. But if that system gets input from say traction control off (which it is in testing) or steering wheel position, bonnet/hood up etc, then you are in violation.
I may have been some poor coder, just on a contract job, no job security, do what is asked for in the requirements or don't come in next week.
And I work in a multi-national company that should know better, and yet we are doing exactly that. No requirements or documentation, at the behest of the PHB. When we refuse to build stuff, we are beaten.
I'm kind of surprised that nobody's brought this up. This case better than any other highlights why we need to get software completely out of vehicles (not just cars). The requirement for ECM software to be tailored to meet the legal emission requirements of specific countries means you've essentially created DVD/Bluray Regions for vehicles. It's now illegal to buy a vehicle in one country and drive it in another country because it cannot be guaranteed to be compliant with the local emissions standards there.
The EPA has a budget of $8B and 15,000 employees.
They cannot take 10 employees and randomly rent cars, bring them to the lab and verify that cars in the field are performing as expected? I mean, air quality is a top charter, and cars are the #2 contributor to air quality. And being sure that cars are performing as expected in the field is thus a critical task. Don't tell me that with 15,000 people you cannot afford to field audits.
When people want to know why I don't want to pay more in taxes, it's precisely because of this. $8B and 15,000 and they cannot even be sure cars are performing as promised. It took a university with a paltry $2M budget to find this.
If the gov were competent, I'd be happy to give them more money.
Start?
can this be a result of VW doing more business in Russia? http://www.bloomberg.com/news/articles/2015-09-04/vw-ford-open-engine-plants-in-russia-amid-car-market-decline
> True, but blackballing is very difficult to do with professionals
Oh, my. I'm tempted to say "you youngsters!", but blackballing is alive and well throughout the worlds. I've certainly had references that an interviewee cited try to blackball them. It caused me to dig further: in some cases, the blackballing was for good reasons, for employees were horrible and, for whatever reason, didn't reason their own personal references would blackball them. Those were people who were so foolish they didn't admit how awful they were.
I've also seen it where employees were let go for age or medical reasons, or for having kids, and their former company wanted to hide why they were _really_ discarded, and so invented other reasons. I can think of several job applicants I've interviewed who ran into this, and I was very fortunate indeed to look deeper and find the _real_ reasons. Sadly, I'm afraid that I may have been fooled a few times and missed out on some good people: my time to dig past the surface and poisoned references for really good engineers is limited.
Industry wide blackballing for illegal reasons is harder than it used to be as the web and online communities have grown. But it certainly still exists. It's also tied to the "glass ceiling" women and minorities encounter in many professions. Networking and word of mouth can sometimes work _against_ good people who are merely so honest or so responsible their current employers can't keep them, especially when a current unscrupulous manager or colleague is willing to lie about them.
But the engine did pass NOx tests. It just runs better when it doesn't. Maybe it gets 140hp@4000 rpm, but with the emissions system working it only gets 90hp@2000rpm instead of, say, 120hp@2000rpm. (Numbers are completely made up).
So where does the workgroup get its requirements from?
It's the same problem pushed up a level.
Confucius say, "Find worm in apple - bad. Find half a worm - worse."
[ ] Customer never needs to refill DEF/AdBlue/whatever outside regular maintenance.
[ ] Customer gets really good fuel mileage, instead of just average.
[ ] Customer gets lots of engine power in a wide operating range, instead of getting lots of engine power at one operating point and not so much power at all other points.
Oh, and guess which of these points actually sell cars.
I had a hunch you were going to say that.
Either Latin do "yes" or Latin do "no."
You Latin do "guess so".
Confucius say, "Find worm in apple - bad. Find half a worm - worse."
Professional Engineers have the power to say no and they have Ethics rules to fall back on.
Programmers are not necessarily professional engineers.
To have a right to do a thing is not at all the same as to be right in doing it
Having letters after our name and belonging to a professional institute gives you added leverage when it comes to declining unethical orders from your bosses, although in reality whistleblowing will just make it harder to get a job in future (unless it is very definitely an individual suggesting the unethical behaviour, and not a company policy).
To have a right to do a thing is not at all the same as to be right in doing it
But this doesn't explain why there wasn't a test case to catch this particular obvious bug - especially where the system is training itself without a degree in law or a sense of ethics. E.g. kicking off passengers and luggage by opening seat belts and opening doors is the next step to increase performance and fuel efficiency.
So the lack of test case happened with at least the QA engineer / QA PHB's knowledge.
Bingo Dictionary - Pragmatist, n. A myopic idealist.
It was a compiler optimizing bug followed by the linker pulling an old code blob...
Exactly... you then Start the car. ;-)
"Never give up, for that is just the time and place when the tide will change." -Harriet Beecher Stowe ^_^
Dunno why you'd avoid it....LOTS of money to be made doing it.
Light travels faster than sound. This is why some people appear bright until you hear them speak.........
The engineers wrote legitimate code (engine mapping) that tested the cars under various scenarios. Nothing wrong with that, and there were probably hundreds of mappings and testing routines in various builds. The final build that was flashed into the ECM could easily have been completed by a low level manager. No one would know unless they did real world testing. The "EPA subroutine" checked for non-drive wheel movement from the ABS sensors (and other sensors to a lesser extent) to determine if the carewas on a stationary dynamometer. If is was, then the EPA engine map was used, and the car put out few emissions. When the car was moving on the road, it spewed like a chimney.
I have to wonder if the other auto manufacturers knew anything about this. They all tear each other's cars down for design intelligence. Mercedes and BMW are still using urea injection, and here VW comes out with an engine that meets emissions standards. Hmmmm?
So you go to ebay and you put in your car's model and there are mod chips for all the cars out there now. They increase performance and allow you to do exactly what VW has done with the OEM chips. So the way I see it is VW has given the public what they wanted, great fuel mileage mixed with performance and a mode that fools the man into thinking you are not polluting the atmosphere. So now the government knows that none of the VW diesels will past an emissions test. VW tried twice to send out a rom upgrade to no avail. The cars performed like crap and lost mileage. I'm kinda scared what the DOT will do now. Will they tell people they can't drive these wonderful cars? Will they force VW into a buy back? Will VW bite the bullet and retrofit a chemical scrubber system into the exhaust manifold? If I had one of these cars I would not want to give it up for nothing but what good is it now if they won't let anyone drive it? I would like to get in touch with the Rat's who let this cat out of the bag and kick their asses for ruining for the rest of us.....
Paul E. Bahre
The code might also have been put in for testing:
What is the optimal performance?
What is the optimal emission?
What is the best combination?
So I have a 2006 Jetta. VW has had several engineering issues with the car. Including a major flaw with the camshaft. I have complained to and spoken with several VW executives and they are so cocky and won't admit to anything. Basically the conversations led to "So what. Do what you want because VW will not do anything about it". They even refused to fix the speedometer being off by 5 MPH which I know affect the total driven miles. They told me "VW engineering made it so. We will not fix it." There are other issues that have resulted in lawsuits like the door and frame cables braking but VW settles and won't admit wrongdoing or engineering fault. Which by the way since my vehicle was out of warranty, it could not get fixed. I said "but there is a freaking lawsuit and settlement for it!". They said "It's out of warranty. That will cost you $360." I just walked out. My cables in the door were all broken and fixed them myself. Had the camshaft changed by a local mechanic, etc. etc. Screw VW and their culture of bigots.
As time goes on, we discover that every diesel powered car manufacturer was cheating and undermining emissions testing.
Why....
Diesel was 30% less expensive for the car owner and that meant that (in Europe), diesel powered cars were 50% of all car sales.
Leslie Satenstein Montreal Quebec Canada
Occam's razor (or maybe just cynicism) makes me doubt whether you're right in this case, but I think you've predicted a class of scenarios that is almost bound to occur in the not-to-distant future as machine learning takes hold. The ways ML finds to "cheat" will always be ahead of the safeguards on cheating.
"The wisdom of the Patriarchs was that they *knew* they were fools." --Master Foo
Hopefully VW will grow a pair and perform discovery on California's EPA and demon spawn, CARB. It's the only state level EPA, which due to the legal timeline, is allowed by the federal clean air act.
When Obama took office, CARB dropped the NOx limit to 1/2 that of the EU. Note that California hates diesels -- they have a consumption tax on gasoline making it $1 higher, per gallon, than the rest of the nation, and diesel fuel is exempt from at least some of these taxes. Every diesel sold in California is a huge loss of consumption tax revenue for the deep blue, progressive democrat controlled state.
Apparently VW's emissions team went to management and told them that the only way to meet the new standard was to install the expensive AdBlue / Urea injection system on the little 4 cylinder, E189 diesel engine. VW refused to authorize the addition of AdBlue technology, claiming that it would price the low cost diesels out of the market.
VW's emissions engineers, stuck between the arbitrary California demands and their employment with VW, appear to have flipped California the bird. The real question is -- given the same choice -- serving a foreign government, a state that is biased against your product, and your paycheck and employment, what is the decision? Quit or hope you don't get caught? Apparently they didn't quit.
At some point the California EPA and CARB are going to need to be dissolved. It's crazy to have so many standards bodies, especially when the governor simply appoints his pals and lets them run rampant over the global economy and what is, in reality, a pretty good automaker.
Unless there is discovery on CARB, we can't know what processes were used, and why they chose to suddenly lower the NOx requirements so drastically below the levels of the EU. If it was to enhance tax revenue, CARB and CAL-EPA should be dissolved immediately for violating the US Constitutional restriction against taxation without representation, and VW should be unconditionally pardoned.
Car's failing smog? Mod the chip... problem solved - forever. How long before we see these shipping from China?
It depends. If it's a workgroup that's basically working on some side project or research project, they probably don't have any real firm requirements, besides a single manager who has a list of some things he wants to do or try out. Not everyone is working on an actual product.
Also, if it's a support group (like "applications engineering"), there's no actual requirements besides "the customer called today and wants such-and-such ASAP!".
For critical, flagship product development (especially in a more mature industry/company), requirements are usually very well formalized, but a lot of things just aren't like that.
Engineers are usually smart enough to know they'll be the ones thrown under the bus first.
So they tend to not do things like this.
Plus getting caught. You're no longer employable...
Cxx and PHB's not so much. They get caught and go on to another company without a problem.
"Seems like all the programmers on Slashdot have never worked in a company larger than 10 people." ...while others assume that all companies larger than 10 people operate the same way.
So, still no proof. This is not news, it's another guess, and not a good one at that. Three words prove that: "...more than likely..."
Move on... there's nothing to see here.
If you get a requirement that "If the test equipment is plugged in then switch to test mode", would that really sound illegal to you?
Would it even occur to you that the government test would be done by plugging into the test connector, at all? Instead of measuring what was actually coming from the tail pipe?
The real problem was that the testers were lazy and did the test in the worst possible way. But no one wants to mention that...