Hacker News Reader

Comment by joshmoz | Hacker News Reader

When I was car shopping the VW diesel numbers, for tdi sportwagen in particular, were impressive. Nothing else came close to VW's combination of power, space, and mileage. Now maybe there are people who know more about cars than I do and can dispute that, but that was my perception a couple of years ago.

If I were the other car companies, I'd want to know exactly how VW was pulling that off. They must have looked into it, and surely they're not as easily misled as I apparently was (how would I know if VW was outright lying about the car?).

This suggests to me that the other car companies must have known that VW was doing something wrong. The fact that they didn't rat VW out suggests to me that they were either doing the same thing (maybe not as aggressively as VW) or they hoped to get away with the same thing. The former seems more likely, and if that's true then I expect this to spread beyond VW soon.

View on HN

VW has also built a huge reputation on diesel. They go and win Le Mans with diesels. They have a giant budget and research ability. They've pushed diesel forward in a lot of ways.

Maybe the others knew they were cheating, I can't imagine that it wouldn't leak out some how. I also can't imagine how you wouldn't go buy a hundred VWs and meticulously take them apart and understand them after getting brutalized in the diesel market.

With mid-sized and heavy trucks, there is an entire subculture of guys that mod them for "more power" and such. There is a little industry built on it and nearly a religion surrounding the "better mileage" and "more power." Some of the systems and devices are sophisticated enough that they have integrated on/off switches for passing smog tests and such. If we really really cared about it, that would be illegal, there would be much more stringent emissions testing more frequently.

I'm of the belief that the regulators knew or suspected there was some cheating but it's political to make waves.

I also can't imagine how you wouldn't go buy a hundred VWs and meticulously take them apart and understand them after getting brutalized in the diesel market.

every miniscule detail of the VW car is patented. if you infringe one of those patents accidentally and VW sues you, you'll get slammed with damages proportionate to money you have made or VW has lost (thanks to you). now, you basically cannot infringe a patent incidentally after you've seen it at work. competition buying a bunch of VW cars to take them apart would be an invitation for VW to sue them for willful patent infringement and the accompanying treble damages.

you can read accounts on the internet of programmers working for major corporations (Sun Microsystems among others IIRC) who were prohibited from reading others' patents for exactly this reason.

it's a thouroughly corrupt system.

edit: "the holder" -> VW + some styling

> competition buying a bunch of VW cars to take them apart would be an invitation for VW to sue them for willful patent infringement and the accompanying treble damages.

As someone noted above, every auto manufacturer is buying the competitor's cars, tearing them down, figuring out they work, rebuilding them, and benchmarking.

The last point is key - you can run your internal tests against their vehicles and see how you compare in your tests.

You can also get a feel for how many molds/stamps/etc that your competitor has in their factory by looking at the mold IDs. You can figure out how tight their weld tolerances are by X-Ray'ing the welds. There is a wealth of information inside a competitor's product that goes beyond the IP.

> You can also get a feel for how many molds/stamps/etc that your competitor has in their factory by looking at the mold IDs.

Also known as the German Tank Problem[0]

Allied intelligence, trying to estimate the extent of Panzer production, used estimates based on serial numbers of various components in tanks that fell into their hands.

From Wikipedia: "Estimating production was not the only use of this serial number analysis. It was also used to understand German production more generally, including number of factories, relative importance of factories, length of supply chain (based on lag between production and use), changes in production, and use of resources such as rubber."

[0]: https://en.wikipedia.org/wiki/German_tank_problem

The auto industry is so incestuous as it is. If patents came to play at all, which they didn't, then the big money play would be for GM, Ford, etc.. to buy VW motors and or license the technology. Either build as good a diesel or if you somehow couldn't then buy it or the technology. There is tons and tons of precedence for it. (Heck, the Ford "PowerStroke" diesel wasn't built by Ford for the first half of its life) Auto competitors work together on tons of stuff and with something like a motor that has fuel demands, it's mutually beneficial, if Ford and GM sell small diesel cars, it's more likely there will be diesel at your nearest pump. GM and Ford build heavy diesel too, there is very very real demand if they can improve power, improve mileage and meet emissions and VAG doesn't compete in that market.

None of that happened, I think the competitors knew and I also think the EPA knew in some capacity and it wasn't until some "small time" researchers got attention that it came out.

Or: mutually assured destruction. They all have something on each other.

Perhaps it's just the US with the throw weight to go after Toyota and VW in the most widespread, newsworthy issues.

It's also helpful that these are foreign car companies.

I also can't imagine how you wouldn't go buy a hundred VWs and meticulously take them apart and understand them after getting brutalized in the diesel market.

This is exactly how the auto industry works, contrary to the uninformed person(s) who modded you down. People should refrain from moderating posts from users who actually know what they're talking about.

For example, the first thing GM did when they began work on the current-generation Corvette was buy a Porsche 911 (from Volkswagen, no less) and study it in detail. This is an objective fact by GM's own admission (http://www.edmunds.com/porsche/911/2013/comparison-test.html). Competitive analysis is a key engineering strategy, no less important than any other.

It's inconceivable that other manufacturers weren't aware of exactly how VW's seemingly-impossible engineering worked. The only question is why they didn't rat them out to the EPA.

  > The only question is why [other manufacturers] didn't rat 
  > them out to the EPA.

Speculation: Ratting out VW carries some degree of risk that the public will become aware that there are better ways to test the emissions of vehicles. The public will then insist that the government legislate newer/better emissions testing, and the manufacturers will have an expensive new problem to solve. Better to keep quiet and let the existing, implemented and paid for (if flawed) process keep running.

> It's inconceivable that other manufacturers weren't aware of exactly how VW's seemingly-impossible engineering worked. The only question is why they didn't rat them out to the EPA.

Everyone cheats, and preserving status quo is in the best interest of everyone.

What worries me is that if this explodes across the industry, the common refrain will become that EPA guidelines are unrealistic and that's why everyone is cheating. Just what we need as the world is coming around to direly needed environmental regulation.

Isn't it better for any individual/employee to have been a whistleblower under the False Claims Act - https://en.wikipedia.org/wiki/False_Claims_Act ?

According to the Wikipedia page "Persons filing under the Act stand to receive a portion (usually about 15–25 percent) of any recovered damages."

Here is an article with a tour of the GM teardown facility, so automakers unquestionably do it.

http://www.slashgear.com/the-auto-vivisectionist-inside-gms-...

That said, there is pretty much no US market for diesel outside the European imports. So they may not be focused on their diesel competitors enough to have done the teardown/research/testing to discover VW cheating. I find that more plausible than everyone cheats and everyone knows everyone else cheats.

> have integrated on/off switches for passing smog tests and such. If we really really cared about it, that would be illegal, there would be much more stringent emissions testing more frequently.

Actually it is illegal.

And inspections in the US are supposed to check if there are any altered emissions components.

It doesn't help if someone removes and reinstalls it each year, but they do try to check for such things.

> The fact that they didn't rat VW out suggests to me that they were either doing the same thing (maybe not as aggressively as VW) or they hoped to get away with the same thing.

Or (only hypothesizing) they realized they're all in the same industry and are necessarily treading softly wondering what the best course to accuse an 800lb gorilla of cheating.

It will be interesting to see if there are other cheaters, and to what degree, and also if this was an open secret in the industry. Nothing would surprise me.

I'm now searching for a story I'm sure happened, which was Mitsubishi faking crash test data... but damned if I can find it. Best I see is this[0]. Am I imagining a scandal w/ Mitsubishi where they were rigging crash test data ?

[0] https://en.wikipedia.org/wiki/Mitsubishi_Motors#Vehicle_defe...

It was a researcher from West Virginia who tested it.

Researcher is on the record as "I didn't believe the results" and "I thought I made a mistake".

Also, lol @ efficient market thesis because the results of his research were posted early 2014. EPA took a look at them, and being the slow bureaucracy that it is... took a year and a half to levy charges against Volkswagen.

So this entire "surprise" was over a year in the making. It's been discussed publicly in research groups for a year and no one seemed to care till now.

Slight correction, EPA was not slow to act. When they caught wind of this they approached VW who had one excuse after another:

http://www.washingtonpost.com/news/wonkblog/wp/2015/09/22/an...

The EMH predicts that unless the net probability of effects where totally zero there would be a move. I don't see evidence of there having been one.

Reply on news.ycombinator.com