News for hackers. Stuff that matters.
Re:A near-disaster at a nuclear weapons lab takes toll on Am... by kragen on Sunday October 02, 2022 @09:54PM (#33062584)
Maybe they're all being counterproductively paranoid.
Re:A near-disaster at a nuclear weapons lab takes toll on Am... by seventytwo on Sunday October 02, 2022 @10:56PM (#33063006)
More nuclear nuclear accidents is exactly what we need to convince the public to use nuclear energy production…
Re:A near-disaster at a nuclear weapons lab takes toll on Am... by kragen on Monday October 03, 2022 @02:29PM (#33071661)
I want to give people the information they need to construct the knowledge to make good choices, and the freedom to put those choices into practice. Convincing people to do things is very rarely useful in this context.
Re:A near-disaster at a nuclear weapons lab takes toll on Am... by bumby on Sunday October 02, 2022 @10:38PM (#33062904)
I can’t tell if you’re being sarcastic?

I don’t think that claim can hold water in an area that 1) has the severity of this particular industry and 2) the org in question was written about because they have documented instances of very bad practices.

If it’s not sarcasm, do you think the Pu risks are overblown?

Re:A near-disaster at a nuclear weapons lab takes toll on Am... by kragen on Sunday October 02, 2022 @11:13PM (#33063098)
It's not sarcastic. I do think the Pu risks are overblown, and I think the numbers I've posted in this thread back that up sufficiently that I don't see how anyone could disagree. Can you explain further?
Re:A near-disaster at a nuclear weapons lab takes toll on Am... by bumby on Sunday October 02, 2022 @11:20PM (#33063130)
Remember, in the US it is a codified right to have a safe workplace. Meaning, employers have a legal responsibility to ensure risks to employee health are properly mitigated.

The article demonstrates a culture a LALN that doesn’t seem to do so, while the other DOE sites seem to do a better job.

Re:A near-disaster at a nuclear weapons lab takes toll on Am... by kragen on Sunday October 02, 2022 @11:28PM (#33063180)
I don't think that LANL or any other DoE site approaches the level of risk of steel mills, farms, garbage trucks, and as you pointed out, construction sites. If there are people willing to undertake such risks to life and limb to grow food or collect garbage, why not to produce nuclear fuel?

Of course it is important that those risks are within their control, not imposed on them by mismanagement or lack of information.

Re:A near-disaster at a nuclear weapons lab takes toll on Am... by bumby on Sunday October 02, 2022 @11:41PM (#33063238)
I don't disagree that, in general, the risk is lower than many other jobs. But that missed the point. It also doesn't mean it would be okay for teachers to have an unsafe work environment, as long as it's safer than a steel mill. Those comparisons are poor measures because it's about mitigating the risk to a reasonable level. In the context of the article, risking a criticality event is not a reasonable level.

Whether or not anyone is willing to take on additional risk also misses the point. Suppose I find people who are willing to live on an aquifer that is polluted by industrial waste because it's cheap land. Does that fact make it okay for the factory in your local area to disregard EPA regulations? I say no, because you have a codified right to certain standards of clean water. It's the same with safe job sites. You'll be able to find people on construction sites willing to work unsafe conditions; that doesn't alleviate their employer from legal responsibility.

What the article is demonstrating is the inability of LANL to implement the proper mitigations, whether those are engineered, administrative, or protective equipment.

Re:A near-disaster at a nuclear weapons lab takes toll on Am... by kragen on Sunday October 02, 2022 @11:55PM (#33063310)
LANL has had three criticality accidents in 80 years, which have killed three people, two of them during WWII (after which things presumably improved) and the third in 01958. That's out of what is now 12000 people, which would lead you to expect about 40 deaths over that period of time, though presumably the workforce size has fluctuated over the years (but it was about 10000 by the end of 01946). I think that by any standard that's a reasonable level of risk.

I don't hold with any of this "codified right" nonsense. What's right and wrong determines what the laws should be; the laws can't determine what's right and what's wrong. We have to determine what's right and what's wrong independently of the laws in order to figure out how they should be changed. Factories poisoning people was wrong before the EPA existed, is still wrong where the EPA doesn't have jurisdiction, and would still be wrong if the EPA allowed it.

Re:A near-disaster at a nuclear weapons lab takes toll on Am... by bumby on Monday October 03, 2022 @09:10AM (#33066982)
I don't hold with any of this "codified right" nonsense.

Forgive me, but the tone of this comes across like talking to someone from the sovereign citizen movement. Maybe you don't think workers have the right to a safe workplace, but the majority of US society does and has passed laws to ensure it. People don't believe in workers rights because there are laws; there are laws because people believe in workers rights.

It's worth discussing if you think the law is wrong, but outright dismissing it as if it doesn't exist is a strange take. I hope for your sake you aren't a supervisor.

So if I understand your position, it's that workers should not be guaranteed a safe workplace. You point out deaths, but overlook the other non-lethal incidents. So assume their bad practices lead to a criticality event that subjected other employees to radiation exposure. Employees that had no input into the decision or actions that lead to the risk. Is that ok in your opinion as long as the number of workplace injuries is below a threshold? Is your stance that the employer can just hand wave away any responsibility by essentially telling the employee "Welp, at least you're not working in a steel mill." By contrast, worker safety regulations define workers safety in this case on a radiation dose basis, irrespective of what goes on in a steel mill. I would say that's a much more objective measure of managing worker risk.

The main point I'm making is that your comparison to the aggregate average workplace injury rate is the wrong perspective. The way organizations like OSHA look at employers is by comparing similar industries. So the safety record of a chemical processing plant isn't compared to something like a pharmacy because the nature of the risks are different. The article clearly states that LANL has a much worse safety record to comparable facilities. I agree that safety regulations can go too far, but you're selecting the wrong benchmark and using it to justify erosion away employee rights with a fairly subjective and arbitrary definition of what is determined reasonable risk.

Re:A near-disaster at a nuclear weapons lab takes toll on Am... by kragen on Monday October 03, 2022 @10:24AM (#33067945)
I do think workers have the right to a safe workplace, or more precisely they have the right to not be coerced into taking risks, but that that right exists inherently; it does not come from laws or what the majority of US society thinks. So those things are irrelevant to determining whether a given workplace is too dangerous or too paranoid.

You say, "Employees that had no input into the decision or actions that [led] to the risk." To the extent that that's true, that's not okay under any circumstances, even if OSHA approves it, and even if they're working in a country with no equivalent of OSHA. But it's important that the risk in question is really a risk that is either important to them or would be important to them if they knew what someone else knows.

Re:A near-disaster at a nuclear weapons lab takes toll on Am... by bumby on Monday October 03, 2022 @11:10AM (#33068576)
I would tend to agree, if those risks are formally communicated and accepted. If they don't accept those risks, the employer is still legally bound to provide a safe working environment. What you see in many of the higher injury-rate industries is that the injuries are often traced back to a lack of diligence in that responsibility. What I've seen in practice is that risks are often downplayed due to cognitive biases or outright gaslighting. What tends to happen is that people want to either 1) push the risk down to a different person, who may not be in a position to either understand or accept that risk or 2) informally accept that risk so that if something goes wrong they have plausible deniability. This gives them all the productivity upside with none of the responsibility.

I don't think most people assume laws create rights (at least, those of the inalienable variety) but they create a framework to protect those rights. That framework is what allows people to objectively determine the safety of workplace and are relevant. In the case of something like OSHA, that framework gives the employee grounds to rebalance the power dynamic with management in terms of understanding and mitigating that risk. Sometimes I think HN is so skewed towards tech that many are unable to understand that people in other jobs don't necessarily have the same leverage with their employers. Laws help ensure they are still protected.

Re:A near-disaster at a nuclear weapons lab takes toll on Am... by kragen on Monday October 03, 2022 @02:26PM (#33071620)
Probably in this case the line employees are, on the most part, better informed than the management: https://news.ycombinator.com/item?id=33067293

And the fact, at LANL, that the safety team left en masse when the management wouldn't let them implement what they considered to be proper safety measures, suggests that the employees in question were in a good position to not only understand but also reject the risk.

In general, line workers are the ones whose incentives are best aligned with safety, since they're usually the ones who die (or, as you pointed out, suffer serious injuries). Sometimes they make bad decisions (as seems to have happened in this case) and sometimes they lack the information necessary — often, as you point out, because they're being lied to. Sometimes regulation helps improve that situation (it has been very useful in extorting MSDSes from material suppliers, for example) but I think much more often it results in the sort of time-wasting box-checking exercise referenced in the other comment, motivated by management ass-covering, not real safety improvements.

Usually, what improves workplace safety is decentralization of control (devolution of decision-making power down to line workers), freedom of information, high-quality coworkers, and an overall workplace culture that values safety and, as I said at the beginning of the thread, ceaselessly seeks ways to improve it. And if that makes me sound like a "sovereign citizen", so be it — even a stopped clock is right twice a day.

For the record, I've only experienced a significant workplace injury once, and the resulting disability was small enough I don't notice it most days, though the scar is obvious. I got fired for it.

Re:A near-disaster at a nuclear weapons lab takes toll on Am... by bumby on Monday October 03, 2022 @03:31PM (#33072483)
It's interesting how wildly differently we can interpret the same information.

>"Probably in this case the line employees are, on the most part, better informed than the management: https://news.ycombinator.com/item?id=33067293"

For instance, I would not consider 2 days of training "highly trained". I believe it's true that the LANL area has one of the highest concentrations of PhDs, but I don't necessarily think that translates to being informed about all the risks. A PhD often means a lot of knowledge, but in a very narrow field. Sometimes, that level of education can be a detriment if it fosters a false sense of general ability. I've had many experiences of Masters and PhD chemists and engineers who felt, "Hey, I'm smart. I've written some cool VBA macros. I'm sure I can program this thing. It's just software." Smash-cut to hazardous releases that cause people to have to shelter in place or firefighters being called. They were subject matter experts in one area, but not smart enough to recognize their own ignorance outside their specialty. It didn't matter that they were experts in their field or front-line workers. I also think it's wild to me that your takeaway from the article is that blatantly disregarding best practices is acceptable. It's a bit of a jump to look at the evidence in the article and assume it must be okay, 'cause, you know, they're first line employees so they must know what they're doing so best practices need not apply. As someone who used to work in safety-critical software, it scares me when I read about this same cavalier SV attitude pervading into safety-critical design like autonomous vehicles.[1] Which relates to,

>"In general, line workers are the ones whose incentives are best aligned with safety"

I think we agree that their incentives are aligned. I think where we disagree is that people on the front lines aren't always acting as rational agents. I don't think they are because of a host of competing incentives and cognitive biases. Major catastrophes bear this out. Good examples are the Challenger or Columbia disasters at NASA. To your point, there was disagreement between management and front-line employees regarding the risk. Both groups were wrong because of these biases. We just aren't wired to estimate low probability events well, especially when there are less abstract competing risks like budget and schedule. That's why NASA created a separate safety arm that was distinct from the production workers. They felt it was necessary to get an objective assessment. Note this lack of impartial judgement is also a finding in [1].

I don't disagree that there's a lot of unhelpful safety theater. But there's also an awful lot of push back against safety that's borne out of a combination of ignorance and arrogance.

[1] https://spectrum.ieee.org/ntsb-investigation-into-deadly-ube...