Hacker News Reader

> I never understand this arguement. I hear it in the form "we should just regulate cars to be safer", why dont you just buy a safer car?

I don't understand your argument. Do you expect every potential car buyer regardless of tech literacy to go to a dealership with an SDR setup on-hand, asking if they could test-out the key fobs, capture some signals, try to figure them out and look for common vulnerabilities on the new models? Do you expect the seller to explain the technical details of the key fobs at a depth you can be sure there's no vulnerabilities?

Determining that the key fobs were implemented securely is not something the buyer or even the seller will be able to determine. I would imagine a regulator has the ability to inspect them as they're being designed/built.

> if you go to buy a car you should have some simple considerations [...] Is it common for this car to be stolen?

By the time vulnerabilities are found and people start stealing them, there should already be plenty sold and roaming the streets.

Or, google.

We should regulate cars to be safer because I don't get to decide which car (or, more likely, truck) the moron who puts me/my family in the hospital (or the grave) was driving. Given the trends in traffic and pedestrian deaths in the US, our regulators are grossly negligent in this regard.

Sure, in the context of broad safety you have a point. Of course, you are more likely to die in places where police are responsible for your protection. The safest places people inhabit are their home, their work, places with private security measures.

The negligence you speak of, how widespread would you say it is? Would you say the EPA dropped the ball preventing the east Palestine Ohio train derailment? Would you say that having some hard cutoff for how many chemical train cars until the train is reclassified? I imagine you would agree, that the chemical company sending 59 (oh w/e) train cars instead of 60 so that they are under easier rules, is kind of bullshit.

If the company is following the rules then they can largely get away with it when things go wrong; if that company had actual liability for whats in the train cars, then they would have insurance. That insurance would be prorated on the safety profile of the train, and it would fix the hazzard or pay the claimants their due. These regulations only exist to cement decrepit companies monopoly over you.

I have no faith that "negligence" in the public sector will improve. The roads, the enforcement of driving conduct, that will not improve. The safety, security, and reliability of cars will improve, because there is still competition from self aware drivers.

> The safest places people inhabit are their home, their work, places with private security measures.

Weird takeaway. Those aren't primarily places with security measures, they're places where trusted people outweigh untrusted ones.

Those are places where you only call the cops ('public security') because the situation has seriously degenerated. Their security measures are the ones you are already familiar with; locks and keys, the understanding that you need specific permission to be present, some sort of fridge policy etc.

I don't have much faith that our regulators are going to get their shit together either, but that doesn't mean I can't hold them responsible for the consequences of their negligence. As for the theory that the invisible hand of the market will drive better outcomes in traffic safety, that is obviously not happening in aggregate. It may even be having the opposite effect, given the unchecked proliferation of pickup trucks and large SUVs that are known to be more dangerous to pedestrians and the drivers of other vehicles.

I'm not super familiar with the East Palestine derailment, but based on what I do know about it I think your 59/60 train cars example is sort of disingenuous. American freight rail operators are well known for skimping on safety in multiple dimensions; issues like obsolete brake technology, stretched-thin crews, insufficient maintenance, and excessively long/heavy trains have all been discussed to death before and after that accident. The problem isn't some off-by-one miscalibration, but rather that our regulators (and I'm including lawmakers under that umbrella, especially since they're usually the ones taking the bribes AFAIK) are simply not willing to demand that these industries put public safety over profits in areas where there are well-understood systemic safety problems. AIUI, efforts to improve rail safety during the 2010s and earlier have been largely stillborn, sabotaged by rail company lobbying.

Just to illustrate what I'm talking about via another example: if the MCAS system that killed 346 passengers on two 737 Max aircraft within the span of a few months had been an honest design flaw rather than a hack devised to juice sales by dodging retraining requirements, cheaply and shoddily implemented ignoring the misgivings of Boeing personnel and rubber-stamped by the FAA, then it would have been a simple tragedy. As it actually happened, it's an atrocity—a failure on multiple levels by many people who ought to have known better, holes in the Swiss cheese all lining up, 346 lives sacrificed on the altar of avarice. That's where the invisible hand gets you when you let it take the wheel, but I suppose next you'll tell me that those people ought to have done their own safety audits of the aircraft that got them killed.

Depending on the make and model of a car, you might save money because insurance companies have determined those cars to be safer. If you have a lowjack, your location, your history, there are a lot of factors that go into insurance costs. Insurance companies adjust insurance prices based on risk. Why would you assume private insurance wouldnt also price their policies based on how safe a train is, what the risk is that it crashes, etc.?

Its about who is signing off on whats safe. The government has proven that thry are at best incompetent, they do not have the incentive. A private company has to maintain their reputation through quality of service, they dont have the seemingly endless faith of people because they are "government".

> Why would you assume private insurance wouldnt also price their policies based on how safe a train is, what the risk is that it crashes, etc.?

I don't assume this and I don't think I have even slightly implied that I do. It's obviously, trivially, verifiably false.

Your mistake is in assuming that cost is the primary motivating factor for consumers^[1], and that corporations and their stakeholders won't favor paying higher insurance costs (and fines, etc.) today over paying even more toward safety improvements that won't break even wrt. insurance costs for many years. As I mentioned above, the observable state of the real world explicitly disproves the claim that market forces will drive positive trends in safety outcomes.

^[1] Look how many huge, expensive pickup trucks are on the road in America today. Look at the monthly payments their owners are enduring to satisfy their vanity; look at their spotless paintjobs, empty beds (except perhaps the token equally spotless toolbox), and mud-free fenders.

You make a good point. The government should not impose rules about what companies do and also the government should hold companies liable for what they do, which is a different thing from having rules

Yes, im not saying you shouldnt be able to sue. Im saying the government shouldnt set a magic bar that if companies hurdle they are without fault.

Exactly. If an insurance company writes rules for a company to follow, it is a good thing. If the government writes those same rules it is a bad thing as the goal is not to reduce the number of thefts but rather make whole the people that get their car stolen and have the money to sue

Doing a security analysis of a car is a complex task that most laymen cannot do, so the argument for laws and compliance among specialists is quite reasonable.

Put it another way: if it's all market based, your choices in buying just get 2x more complex. Is this car easily unlocked? Do the brakes often fail? Can it resist a collision?

Or even another thought: since most people don't think about security, companies will flood the market with insecure cars. Want a secure car? That would be niche and cost you double.

You shouldn't have to consider if the car's locks work, because working locks should be a baseline in cars produced in a first world country. It's that simple and easy. You aren't nearly as smart as you think you are and having basic standards of how corporations should operate isn't coddling.

Then by the standards set by government, I guess the united states isnt a first world country.