8 ms·
What really happened at Volkswagen
- rurban 9y agoOh no, another VW hit piece, attacking the biggest US competitor. It is definitely NOT what really happened with the emissions scandal, as it omits the most important facts, how this cheating scandals began, and how it spread out. The first emission cheating scandals happened with US carmakers, then the US and EU emission testing laws deviated, allowing the US to cheat with their drivers, giving national carmakers the advantage, whilst EU and Asia committed to reproducible tests. But only on one model out of R&D, and not every single engine coming out of production. And then of course the tricks began to detect the tests, probably at Audi (certainly not VW), with Bosch being the one who needs to take the blame. They implemented it. Then many others detected the same tricks, and either used the sophisticated cycle detection in the Bosch ECU or used simplier timers. In the last list I saw 50% of all tested engines cheated with the NOx emissions, and VW was not even at the top of the numbers.
- golergka 9y agoThat's an alternative view I haven't seen before, thanks. Do you have some links for further reading on this?
- goldenkey 9y agohttps://youtu.be/7t4paclIwuU https://youtu.be/7t4paclIwuU
- jokoon 9y agoYes, to me it really looks like those kind of things go well beyond VW, it seems it's mostly about national economic interests.
- matt4077 9y agoHello! Just the other day I argued that the EU's new antitrust case against Google isn't economic nationalism, so allow me to even the score :) VW's conduct both in implementing these cheats, as well as resisting the investigations into them (to this day!) was obviously, morally indefensible, and totally stupid. Health statistics put the most likely number of premature deaths caused by it somewhere in the 4-digits, in the US alone. The absolute amount of the fine must have punitive component large enough to make it obvious that such crimes don't pay. At the same time, it wold have been unconscientious to level a fine that would bankrupt VW. I believe the US did a pretty good job. It also shows a deep misunderstanding of the processes by which the Federal Government works. These fines are not determined by some Trump-like egomaniac impulsively acting on his emotions. Nor is the EPA known to be a fan club of US car manufacturers. Why only VW? It's quite easy: these manipulations are really only relevant to Diesel engines, which for some reason became a fetish for manufacturers in Germany only. VW sells more Diesel cars in the US than all other companies together.
- rurban 9y agoWhow, so many mistakes. "VW's conduct both in implementing these cheats": No Bosch implemented these cheats. If there was help or demand from outside, it was Audi, not VW. "Why only VW? VW sells more Diesel cars in the US than all other companies together." The fines should be on all manifacturers of the cheaters. From the list of the affected engines I see not a single one being targeted. The relevant VW engines are all manifactured by Audi, the relevant cheats were manifactured by Bosch, the other relevant cheaters involve everyone: Audi, Daimler, BMW, Skoda, Volvo, VW, Jaguar, Land Rover, Citroen, Mazda, Peugeot, Hyundai, KIA, Toyota, Ssangyong, Infinity, Fiat, Ford, Mitshubishi, Seat, Vauxhall, Chevrolet, Dacia, Honda, Nissan, MG, Porsche, Renault, ... you name it. https://www.theguardian.com/business/2016/apr/23/diesel-cars-pollution-limits-nox-emissions https://www.theguardian.com/business/2016/apr/23/diesel-cars... Only those 7 diesel cars meet the Euro 6 NOx limits, the other 97% not. * Audi A5 2014 * BMW 3 Series 2013 * Skoda Superb 2016 * Volkswagen Golf SV 2015 * Volkswagen Passat 2016 * Volkswagen Scirocco 2015 * Volkswagen Touran 2016 Note that this mostly consists of VW cars meeting the limits, all others not. A good technical overview is summarized here: http://www.theicct.org/sites/default/files/publications/ICCT_NOx-control-tech_09032015.pdf http://www.theicct.org/sites/default/files/publications/ICCT...
- Kluny 9y agoThe first couple paragraphs is light-on-facts shade throwing. I didn't read the rest. My first impression was not one of even-handed journalism.
- nitinreddy88 9y agoExactly my thoughts. I read almost all these articles published over the last one year and none of them revealed anything new. Same story writing in different way with same headlines.
- jchw 9y agoI especially dig the necessity to bring up Adolf Hitler. As if that has any relevance AT ALL. Especially when we know that the problem isn't even specific to VW, with many manufacturers accused of the same thing. If anyone really wants to learn more about what happened here, I can't recommend more the 33c3 talk "Software Defined Emissions," which tells you a ton more than this clickbait hack journalism could ever.
- moreentropy 9y ago> I can't recommend more the 33c3 talk "Software Defined Emissions," which tells you a ton more than this clickbait hack journalism could ever. https://media.ccc.de/v/33c3-7904-software_defined_emissions https://media.ccc.de/v/33c3-7904-software_defined_emissions
- smartbit 9y agoThe technical details of Dieselgate in 32C3 talk [0]. Highly recommended. At around 3260 secs. the famous acoustic function is revealed. [0] https://media.ccc.de/v/32c3-7331-the_exhaust_emissions_scandal_dieselgate#video&t=3260 https://media.ccc.de/v/32c3-7331-the_exhaust_emissions_scand...
- toomuchtodo 9y agoI hope Silicon Valley or Mr Robot steals "acoustic function" for use as a plot device of hiding something nefarious in your codebase.
- goldenkey 9y agoVW was barely the only one who cheated. They just got the brunt of the punishment so the other cheaters could evade just desserts. Watch this video: https://youtu.be/7t4paclIwuU https://youtu.be/7t4paclIwuU
- payne92 9y agoRegarding the other comments here: this is a book review, not a general analysis of emissions cheating. I am about halfway through the book right now (so far, very highly recommended). There is a lot more in the book than is captured in this review. VW's history is extremely interesting, and the company has unique organizational aspects (even by German standards). And related, an excellent, detailed technical summary of cheating implementations, including the historical evolution across manufacturers: https://cseweb.ucsd.edu/~klevchen/diesel-sp17.pdf https://cseweb.ucsd.edu/~klevchen/diesel-sp17.pdf
- Tomte 9y ago"Unique" is very right: https://en.m.wikipedia.org/wiki/Volkswagen_Act https://en.m.wikipedia.org/wiki/Volkswagen_Act
- wiz21c 9y agoFTA : >>> He delves into VW’s origins, when Adolf Hitler ordered the construction of a “people’s car”, or Volkswagen in German. VW set up shop in the German countryside. Wolfsburg bred a “headquarters mentality” that insulated the firm from outside influence. Unprecedented union power, handed over in the 1960s as the price the federal government paid for floating the firm on the stockmarket, and the sway of the state of Lower Saxony, which retained a 20% voting stake in the company, gave outside shareholders little say. So basically, union power and government involvement (and lack of private investors) lead to cheating ? come on...
- LoSboccacc 9y agono, it's in your quote, Hitler did it /s that's just some comtext blurb, you're extrapolating too much out of it
- wiz21c 9y ago>>>> Unprecedented union power, handed over in the 1960s as the price the federal government paid for floating the firm on the stockmarket Unless I'm missing something, this is roughly 15 years after Hitler disappeared. Or my english skills are truly fading away :-(
- msnower 9y agoThe "autocratic" management style seems to be the biggest cause of this issue. I'd be surprised if members of upper management didn't have a sense of what's going on. It reminds me of the events leading NASA Challenger disaster. The engineers and lower management employees were totally ignored by upper management in that scenario. Though, I should mention, I haven't read any books about the VW scandal yet.
- mowenz 9y agoI thought the same thing with the groupthink at NASA. It's hard to put in perspective just how badly NASA management screwed up though--probably much moreso than here. I would bet 90%+ of HN user's great grandmothers would not make the same mistake. Of course they wouldn't--it takes a special case of narcissistic manager who can both get himself to the top and also tell his engineers no because he cant admit a mistake and would rather risk blowing up a space shuttle (and it happened).
- Animats 9y agoThe "autocratic" management style seems to be the biggest cause of this issue. Unclear. The author of the book takes that position. The author seems to be hostile to worker ownership or state ownership. Companies with multiple classes of stock to deprive the stockholders of voting control can have a much worse autocratic leader problem. Google and Facebook are set up that way, of course. Here's the list of the 28 S&P 500 companies with restricted voting rights.[1] As a group, they underperform the full S&P 500. Some of those companies that aren't doing well face shareholder litigation. Viacom's minority shareholders did manage to get rid of Sumner Redstone, but he was 93 and really had to go. For many decades, no big company did this, because the NYSE didn't allow such companies to be listed. (Except for Ford and Hershey, which predated that rule; they were grandfathered in.) It's still rare, but now companies can get away with it. [1] https://s.thestreet.com/files/tsc/v2008/photos/contrib/uploads/a982bf14-396f-11e6-8a30-933f82343784.JPG https://s.thestreet.com/files/tsc/v2008/photos/contrib/uploa...
- blazespin 9y agoI wonder if there is a dropoff in returns for companies with voting control. Google and FB have been pretty good in their early years so far.
- asmithmd1 9y agoWhen I first heard about the cheating I saw VW engineers (both management and practitioners) as totally at fault. I thought they were directly detecting that the vehicle was undergoing a test and modifying the engine computer in response. I later found out they were changing the computer algorithm only when the inputs to the throttle and gearshift matched the test protocol. To me this seems like a legitimate way to "cheat." They are using TDD and are passing the test. The test writers should have thrown in a secret or random test as well as the pre-defined tests.
- nathanvanfleet 9y agoLaws don't work that way
- asmithmd1 9y agoIn this case the law is written precisely like a unit test: https://en.wikipedia.org/wiki/New_European_Driving_Cycle https://en.wikipedia.org/wiki/New_European_Driving_Cycle https://en.wikipedia.org/wiki/FTP-75 https://en.wikipedia.org/wiki/FTP-75 The problem with the testing is that it also followed the exact same unit test. I have not read the law, but they certainly violated the spirit of the law if not also the letter.
- kwillets 9y agoIf the law worked the way you think, they wouldn't need a test.
- kwillets 9y agoPart of the problem is that the old analog systems were never expected to match the tests in the real world either -- that's the origin of the YMMV phrase. But I suppose the tests were sufficient to keep them within some bound.
- gech 9y agoWhy do you want to encourage a race to the bottom?
- mnm1 9y agoLike VW is the only company that cheated on emissions? Fiat Chrysler, Daimler, GM, Mitsubishi, Renault, and the list keeps growing. The interesting story is why the whole industry is allowed to get away with cheating. And it's got nothing to do with Hitler.
- Judgmentality 9y ago> And it's got nothing to do with Hitler. What?
- fatbird 9y agoTFA starts by mentioning VW's origins in Hitler's creation of a group to create a "people's car", and cites a significant part of that as being an organization created and sheltered from outside influence, which apparently set up the conditions for a very autocratic CEO to create the internal culture that led to the cheating/fraud several decades later. It's a really weird, weak connection.
- macintux 9y agoIn fairness, the violations in the other cases I've seen (and I admit I haven't paid all that close attention) seem to be much more marginal.
- rasz 9y agoThere was a program on national French TV almost a full year before VW case blew up, they tested PSA, Renault and Fiat cars and _every single one_ was several times above the norm. also http://www.cnbc.com/2017/04/24/reuters-america-update-1-french-prosecutors-open-peugeot-diesel-cheating-probe.html http://www.cnbc.com/2017/04/24/reuters-america-update-1-fren...