4 ms·

> I have news for you, if you think you have it you got it This is a ridiculous thing to say after talking about test accuracy. You can't improve what you can

> I have news for you, if you think you have it you got it This is a ridiculous thing to say after talking about test accuracy. You can't improve what you can't quantify. Also your assumption that so many people have a job they can do from home or over email is ridiculous.

altcognito 7y ago
In fact, you won’t know if your test is inaccurate if you don’t test! Over and over again I hear people talk about how it is a waste of resources to test. Tests are not expensive in time or material. They will pay off. No, they don’t help you cure anything. We dont have a cure! Tests help you quarantine and mitigate.
ajuc 7y ago
> You can't improve what you can't quantify. That's obviously false.
- TheSoftwareGuy 7y ago
  It may not be 100% true, but quanitfying the accuracy certainly helps.
- bumby 7y ago
  Sure, you can randomly shotgun a bunch of potential solutions and hope for the best, but any systematic improvement process relies on quantitative measures and root cause analysis.
- BubRoss 7y ago
  How would you know if it was?
  - ajuc 7y ago
    Random changes can be positive and you don't need to quantify anything to do them. Also quantifying after the fact isn't what we're talking about so I could test and verify that indeed you did improved something 10 years ago and you made that improvement without quantifying anything yourself.
    - BubRoss 7y ago
      How do you know if something is positive without being able to quantify it?
      - ajuc 7y ago
        You don't need to know about something for it to be true.
        BubRoss 7y ago
        You don't know if something is true if you can't measure or test it.
        ajuc 7y ago
        When you throw dice 1000 times without looking you don't know when you got a 6 but you know you did.
        BubRoss 7y ago
        Too bad we don't have 1000 societies to randomly try things on. Also you don't know that you rolled a six, the probability is high that you did, that's very different. If you don't know where you rolled the number you wanted, how can you repeat it? If you can't repeat it how can you build on top of that?
        ajuc 7y ago
        > Also you don't know that you rolled a six, the probability is high that you did, that's very different. By that standard you don't know anything about the world at all - gravity may stop working the next minute - it's unlikely but not impossible. We don't have a proof it won't - we only have assumptions(models) supported by statistics. For me the standards that physics use is good enough - I say I know gravity works and I know I rolled a 6 when I rolled 1000 d6. You might call that differently but I'm pretty sure you won't jump out of a cliff and you won't bet money on NOT rolling a 6 in 1000 tries. > Too bad we don't have 1000 societies to randomly try things on. That's exactly why we shouldn't wait for results. We have data from 1000s of people though and several countries that tried various strategies. Social isolation works best so we should do that, immediately. With exponential growth doing the right thing NOW is more important than making sure you're doing it in an optimal way but a week later. > If you don't know where you rolled the number you wanted, how can you repeat it? If you can't repeat it how can you build on top of that? In the case of this disease we know how to prevent it overwhelming our healthcare systems so that less than 1% of population dies instead of over 5%. That's the most important thing, producing the vaccine and drugs, calculating the exact number of people sick, discovering the exact properties of the sickness - these things come after the fact. We won't have a vaccine nor drugs in time anyway. People arguing to measure and then act in case of a disease that infects 1000 times more people every month are insane. In 3 months the whole world population would be sick and most of it AT ONCE. Who will treat the 5-10% of patients that need help? Nobody - that's who. So they will die.

altcognito 7y ago

In fact, you won’t know if your test is inaccurate if you don’t test! Over and over again I hear people talk about how it is a waste of resources to test. Tests are not expensive in time or material. They will pay off. No, they don’t help you cure anything. We dont have a cure! Tests help you quarantine and mitigate.

ajuc 7y ago

> You can't improve what you can't quantify. That's obviously false.

TheSoftwareGuy 7y ago
It may not be 100% true, but quanitfying the accuracy certainly helps.
bumby 7y ago
Sure, you can randomly shotgun a bunch of potential solutions and hope for the best, but any systematic improvement process relies on quantitative measures and root cause analysis.
BubRoss 7y ago
How would you know if it was?
- ajuc 7y ago
  Random changes can be positive and you don't need to quantify anything to do them. Also quantifying after the fact isn't what we're talking about so I could test and verify that indeed you did improved something 10 years ago and you made that improvement without quantifying anything yourself.
  - BubRoss 7y ago
    How do you know if something is positive without being able to quantify it?
    - ajuc 7y ago
      You don't need to know about something for it to be true.
      - BubRoss 7y ago
        You don't know if something is true if you can't measure or test it.
        ajuc 7y ago
        When you throw dice 1000 times without looking you don't know when you got a 6 but you know you did.
        BubRoss 7y ago
        Too bad we don't have 1000 societies to randomly try things on. Also you don't know that you rolled a six, the probability is high that you did, that's very different. If you don't know where you rolled the number you wanted, how can you repeat it? If you can't repeat it how can you build on top of that?
        ajuc 7y ago
        > Also you don't know that you rolled a six, the probability is high that you did, that's very different. By that standard you don't know anything about the world at all - gravity may stop working the next minute - it's unlikely but not impossible. We don't have a proof it won't - we only have assumptions(models) supported by statistics. For me the standards that physics use is good enough - I say I know gravity works and I know I rolled a 6 when I rolled 1000 d6. You might call that differently but I'm pretty sure you won't jump out of a cliff and you won't bet money on NOT rolling a 6 in 1000 tries. > Too bad we don't have 1000 societies to randomly try things on. That's exactly why we shouldn't wait for results. We have data from 1000s of people though and several countries that tried various strategies. Social isolation works best so we should do that, immediately. With exponential growth doing the right thing NOW is more important than making sure you're doing it in an optimal way but a week later. > If you don't know where you rolled the number you wanted, how can you repeat it? If you can't repeat it how can you build on top of that? In the case of this disease we know how to prevent it overwhelming our healthcare systems so that less than 1% of population dies instead of over 5%. That's the most important thing, producing the vaccine and drugs, calculating the exact number of people sick, discovering the exact properties of the sickness - these things come after the fact. We won't have a vaccine nor drugs in time anyway. People arguing to measure and then act in case of a disease that infects 1000 times more people every month are insane. In 3 months the whole world population would be sick and most of it AT ONCE. Who will treat the 5-10% of patients that need help? Nobody - that's who. So they will die.