5 ms·

Large and complex systems are fundamentally unpredictable and have tradeoffs and consequences that can't be foreseen by anybody. Error rates are never zero. S

Large and complex systems are fundamentally unpredictable and have tradeoffs and consequences that can't be foreseen by anybody. Error rates are never zero. So basically anything large enough is going to kill people in one way or another. There are intelligent ways to deal with this, and then there is shooting the CEO, which will change nothing because the next CEO faces the exact same set of choices and incentives as the last one.

BrenBarn 9mo ago
Well, given what you said, one obvious mechanism is to cap the sizes of these organizations so that any errors are less impactful. Break up every single company into little pieces.
- terminalshort 9mo ago
  That doesn't really help because the complexity isn't just internal to the companies, but also exists in the network between entities that make up the industry. I may well even make it worse because it is much harder to coordinate. e.g. If I run into a bug cause by another team at work, it's massively easier to get that fixed than if the bug is in vendor software. In terms of health insurance, which is the industry where the CEO got shot, we can pretty definitively say that it's worse. More centralized systems in Europe tend to perform better. If you double the number of insurance companies, then you double the number of different systems every hospital has to integrate with. We see this on the internet too. It's massively more centralized than 20 years ago, and when Cloudflare goes down it's major news. But from a user's perspective the internet is more reliable than ever. It's just that when 1% of users face an outage once a day it gets no attention, but when 100% of users face an outage once a year everyone hears about it even though it is more reliable than the former scenario.
asdff 9mo ago
Pretty predictable what happens when you deny coverage for a treatment someone needs
- terminalshort 9mo ago
  But do they need it? How do you know? And don't say because the doctor said so, because doctors disagree all the time. When my grandfather was dying in his late 80s, the doctor said there was nothing he could do. So his children took him to another doctor, who said the same. And then another doctor, who agreed with the first two. But then they took him to a 4th doctor, who agreed to do open heart surgery, which didn't work, and if anything hastened his inevitable death due to the massive stress. The surgery cost something like 70 grand and they eventually got the insurance company to pay for it. But the insurance company should not have paid for it because it was a completely unnecessary waste of money. And of course there will be mistakes in the other direction because this just isn't an exact science.
  - asdff 9mo ago
    At that point, why cover anything at all if the doctor could always be wrong?
    - terminalshort 9mo ago
      Stupid question. If you have a better way to make decisions on insurance coverage then state it.
      - asdff 9mo ago
        Why is it on me to come up with a new model for healthcare? I can acknowledge shortcomings of the present system without having to come up with solutions for them.
        latency-guy2 9mo ago
        > Pretty predictable what happens when you deny coverage for a treatment someone needs Other poster demonstrated that you have no idea what "need" is. So you also have no idea what a "shortcoming of the present system" is either, because how the hell would you even know?
imiric 9mo ago
I'm not talking about unpredictable tradeoffs and consequences. I'm talking about intentional actions that lead to deaths. E.g. [1] and [2], but there are numerous such examples. There is no plausible defense for this. It is pure evil. [1]: https://en.wikipedia.org/wiki/Tobacco_Institute https://en.wikipedia.org/wiki/Tobacco_Institute [2]: https://en.wikipedia.org/wiki/Purdue_Pharma https://en.wikipedia.org/wiki/Purdue_Pharma
- terminalshort 9mo ago
  Well those get handled. Perdue was sued into bankruptcy and the Tobacco Institute was shut down when the industry was forced to settle for $200 billion in damages.
  - imiric 9mo ago
    So human lives have a price tag, and companies can kill millions for decades as long as they pay for it. Gotcha.

BrenBarn 9mo ago

Well, given what you said, one obvious mechanism is to cap the sizes of these organizations so that any errors are less impactful. Break up every single company into little pieces.

terminalshort 9mo ago
That doesn't really help because the complexity isn't just internal to the companies, but also exists in the network between entities that make up the industry. I may well even make it worse because it is much harder to coordinate. e.g. If I run into a bug cause by another team at work, it's massively easier to get that fixed than if the bug is in vendor software. In terms of health insurance, which is the industry where the CEO got shot, we can pretty definitively say that it's worse. More centralized systems in Europe tend to perform better. If you double the number of insurance companies, then you double the number of different systems every hospital has to integrate with. We see this on the internet too. It's massively more centralized than 20 years ago, and when Cloudflare goes down it's major news. But from a user's perspective the internet is more reliable than ever. It's just that when 1% of users face an outage once a day it gets no attention, but when 100% of users face an outage once a year everyone hears about it even though it is more reliable than the former scenario.

asdff 9mo ago

Pretty predictable what happens when you deny coverage for a treatment someone needs

terminalshort 9mo ago
But do they need it? How do you know? And don't say because the doctor said so, because doctors disagree all the time. When my grandfather was dying in his late 80s, the doctor said there was nothing he could do. So his children took him to another doctor, who said the same. And then another doctor, who agreed with the first two. But then they took him to a 4th doctor, who agreed to do open heart surgery, which didn't work, and if anything hastened his inevitable death due to the massive stress. The surgery cost something like 70 grand and they eventually got the insurance company to pay for it. But the insurance company should not have paid for it because it was a completely unnecessary waste of money. And of course there will be mistakes in the other direction because this just isn't an exact science.
- asdff 9mo ago
  At that point, why cover anything at all if the doctor could always be wrong?
  - terminalshort 9mo ago
    Stupid question. If you have a better way to make decisions on insurance coverage then state it.
    - asdff 9mo ago
      Why is it on me to come up with a new model for healthcare? I can acknowledge shortcomings of the present system without having to come up with solutions for them.
      - latency-guy2 9mo ago
        > Pretty predictable what happens when you deny coverage for a treatment someone needs Other poster demonstrated that you have no idea what "need" is. So you also have no idea what a "shortcoming of the present system" is either, because how the hell would you even know?

imiric 9mo ago

I'm not talking about unpredictable tradeoffs and consequences. I'm talking about intentional actions that lead to deaths. E.g. [1] and [2], but there are numerous such examples. There is no plausible defense for this. It is pure evil. [1]: https://en.wikipedia.org/wiki/Tobacco_Institute https://en.wikipedia.org/wiki/Tobacco_Institute [2]: https://en.wikipedia.org/wiki/Purdue_Pharma https://en.wikipedia.org/wiki/Purdue_Pharma

terminalshort 9mo ago
Well those get handled. Perdue was sued into bankruptcy and the Tobacco Institute was shut down when the industry was forced to settle for $200 billion in damages.
- imiric 9mo ago
  So human lives have a price tag, and companies can kill millions for decades as long as they pay for it. Gotcha.