4 ms·

I haven't done the math but I'm prepared to bet that when it actually comes to the crunch you aren't as confident as you say. 'it can't happen' implies 0% proba

by dejb 17y ago

I haven't done the math but I'm prepared to bet that when it actually comes to the crunch you aren't as confident as you say. 'it can't happen' implies 0% probability. Well I'll go you one better as I'm prepared to bet on it at 0.01% for $1. I'll even send you the dollar as long as you agree to the bet. So that's my $1 vs your $10K. Talk is cheap. What I'm trying to demonstrate is how certainty can vanish when faced with actual consequences.

scott_s 17y ago
When someone says "it can't happen" in a scientific context, I don't take it to mean "I believe it is impossible to happen." I take it to mean "according to our current understanding, it's not possible." The implication being that while it's possible, if it were true it would require fundamental changes in our understanding. This is so unlikely to be incalculable.
- dejb 17y ago
  Well according to the constructs of the English language and logic, "it can't happen" does actually mean "I believe it is impossible to happen.". (Can't == can not == impossible). Your statement implies the 'scientific context' is bullshit which I certainly do not agree with. So in your alternative world of logical meaning what does "it can't happen" equate to anyway? Does it simply mean less than 50% probability or what? If you aren't prepared to put an actual mathematical meaning to probabilistic statements then they are worthless. I certainly am and I am prepared to back them up with cash. What about you? Do your statements mean anything or are they just worthless oratory?
  - scott_s 17y ago
    The intended meaning always depends on context. People are rarely precise with their words; it's takes too much time and doesn't improve communication if all parties have a shared understanding. At the start of my thermal physics class, my professor spent a lecture to define what a thermal physicist means when they say "never." We spent time calculating things like the likelihood of Hamlet emerging from 6 billion monkeys banging on keyboards for 13 billions years. This sort of understanding is important when talking about entropy and the laws of thermal dynamics. For example, a thermal physicist would be comfortable saying all of the molecules in a room will never spontaneously exist on only one side of that room. With simplifying assumptions, we can calculate the probability of that happening. While it's non-zero, it's so infinitesimal that physicists are comfortable saying "never", and we have the second law of thermodynamics. Your insistence on putting a dollar number on my confidence is strange, since even in your case it's rhetorical. No one is asking you to put money up, nor does the infrastructure or even rules exist for how to handle it. Further, my entire point was to demonstrate that certainty in science is impossible. All we have is confidence in something.
    - dejb 17y ago
      > All we have is confidence in something. Yes but if you can't define what confidence is what does it mean? When I studied physics we learned to provide confidence intervals for each measurement and subsequent calculation so every variable was of the form X +/- Y. From memory the confidence was a specific number of deviations from the mean. That meant something. Saying 'roughly X' doesn't. All the examples you provided are so improbable that 'never' is of course appropriate in most contexts. But what do you think 'it can't happen' means in a case like the existence of a phenomena like EM-sensitivity? 10% probability, 1%, 0.1%, 0.01%, 0.001%? None of these are at all appropriate to describe as 'it can't happen' in this context. I think that in your heart of hearts you know that chances of EM-sensitivity existing could well be better than 1%. > No one is asking you to put money up, nor does the infrastructure or even rules exist for how to handle it. Check out http://longbets.org http://longbets.org . Otherwise I'd settle for a proof of ID from whomever wants to take the bet and a public and signed statement to the effect that they'll pay up if the conditions are met within a certain time period. I'm pretty sure this type of bet is legal in most jurisdictions.
      - scott_s 17y ago
        A confidence interval is a different beast than saying that you are confident that current understanding is correct. One can be quantified, the other cannot.
        dejb 17y ago
        > One can be quantified, the other cannot. I've just described a method for quantifying probability. How hard is this to understand? Neither really exist. The measure of X is some exact amount just as a clearly defined proposition is either true or false. You can actually simplify the confidence interval case to a boolean proposition (e.g. is X greater than 7).
        scott_s 17y ago
        I must not have been clear in my previous post: there is a difference between an objective confidence (such as from experimental measurements) and subjective confidence (such as my own personal confidence that something is correct). Objective confidences can be meaningfully quantified. Subjective confidences cannot. For example, I am confident that special relativity is correct and that the speed of light is the unreachable-ceiling for the speed of matter with mass. A century's worth of observations and experiments support this. However, trying to quantify that subjective confidence is meaningless.

by dejb 17y ago

scott_s 17y ago

When someone says "it can't happen" in a scientific context, I don't take it to mean "I believe it is impossible to happen." I take it to mean "according to our current understanding, it's not possible." The implication being that while it's possible, if it were true it would require fundamental changes in our understanding. This is so unlikely to be incalculable.

dejb 17y ago
Well according to the constructs of the English language and logic, "it can't happen" does actually mean "I believe it is impossible to happen.". (Can't == can not == impossible). Your statement implies the 'scientific context' is bullshit which I certainly do not agree with. So in your alternative world of logical meaning what does "it can't happen" equate to anyway? Does it simply mean less than 50% probability or what? If you aren't prepared to put an actual mathematical meaning to probabilistic statements then they are worthless. I certainly am and I am prepared to back them up with cash. What about you? Do your statements mean anything or are they just worthless oratory?
- scott_s 17y ago
  The intended meaning always depends on context. People are rarely precise with their words; it's takes too much time and doesn't improve communication if all parties have a shared understanding. At the start of my thermal physics class, my professor spent a lecture to define what a thermal physicist means when they say "never." We spent time calculating things like the likelihood of Hamlet emerging from 6 billion monkeys banging on keyboards for 13 billions years. This sort of understanding is important when talking about entropy and the laws of thermal dynamics. For example, a thermal physicist would be comfortable saying all of the molecules in a room will never spontaneously exist on only one side of that room. With simplifying assumptions, we can calculate the probability of that happening. While it's non-zero, it's so infinitesimal that physicists are comfortable saying "never", and we have the second law of thermodynamics. Your insistence on putting a dollar number on my confidence is strange, since even in your case it's rhetorical. No one is asking you to put money up, nor does the infrastructure or even rules exist for how to handle it. Further, my entire point was to demonstrate that certainty in science is impossible. All we have is confidence in something.
  - dejb 17y ago
    > All we have is confidence in something. Yes but if you can't define what confidence is what does it mean? When I studied physics we learned to provide confidence intervals for each measurement and subsequent calculation so every variable was of the form X +/- Y. From memory the confidence was a specific number of deviations from the mean. That meant something. Saying 'roughly X' doesn't. All the examples you provided are so improbable that 'never' is of course appropriate in most contexts. But what do you think 'it can't happen' means in a case like the existence of a phenomena like EM-sensitivity? 10% probability, 1%, 0.1%, 0.01%, 0.001%? None of these are at all appropriate to describe as 'it can't happen' in this context. I think that in your heart of hearts you know that chances of EM-sensitivity existing could well be better than 1%. > No one is asking you to put money up, nor does the infrastructure or even rules exist for how to handle it. Check out http://longbets.org http://longbets.org . Otherwise I'd settle for a proof of ID from whomever wants to take the bet and a public and signed statement to the effect that they'll pay up if the conditions are met within a certain time period. I'm pretty sure this type of bet is legal in most jurisdictions.
    - scott_s 17y ago
      A confidence interval is a different beast than saying that you are confident that current understanding is correct. One can be quantified, the other cannot.
      - dejb 17y ago
        > One can be quantified, the other cannot. I've just described a method for quantifying probability. How hard is this to understand? Neither really exist. The measure of X is some exact amount just as a clearly defined proposition is either true or false. You can actually simplify the confidence interval case to a boolean proposition (e.g. is X greater than 7).
        scott_s 17y ago
        I must not have been clear in my previous post: there is a difference between an objective confidence (such as from experimental measurements) and subjective confidence (such as my own personal confidence that something is correct). Objective confidences can be meaningfully quantified. Subjective confidences cannot. For example, I am confident that special relativity is correct and that the speed of light is the unreachable-ceiling for the speed of matter with mass. A century's worth of observations and experiments support this. However, trying to quantify that subjective confidence is meaningless.