• Urist@lemmy.ml
    link
    fedilink
    English
    arrow-up
    18
    arrow-down
    5
    ·
    edit-2
    10 months ago

    Well no. You can try to count every real number forever and you will miss infinitely many still. Some infinites are larger than others, hence I do not see any reason why “infinite time” would cover “every possibility happening”. On the other hand, if you do have a mathematical proof you could refer to, I would be most grateful.

    EDIT: To write out my example, let us consider a machine that picks a random number between 3 and 4 every second. Then there is every second a nonzero chance that this machine (assuming true and not pseudo randomness) will pick, say pi. The range of numbers picked constitute the image of a function from the whole numbers to the real numbers (up to isomporphism), which cannot be surjective. Hence there are numbers not picked even though there was a > 0 chance of picking them every second for an infinite time.

    • xigoi@lemmy.sdf.org
      link
      fedilink
      arrow-up
      11
      ·
      10 months ago

      Then there is every second a nonzero chance that this machine (assuming true and not pseudo randomness) will pick, say pi.

      No. The probability of picking any particular number from a uniform distribution is 0.

      On the contrary, since the works of Shakespeare are a finite string over a finite alphabet (I can formalize this argument if you want), the probability of typing them out after some fixed large number of keystrokes is some nonzero number 𝑝. With 𝑛 monkeys, the probability that at least one will type out the works is 1 − (1 − 𝑝)ⁿ, which goes to 1 as 𝑛 → ∞.

      Now, you are right that this does not mean that the works are guaranteed to be typed out. However, it has probability 1, so it’s mathematically “almost certain”.

    • Dran@lemmy.world
      link
      fedilink
      arrow-up
      10
      ·
      10 months ago

      I don’t think I understand your example but I feel like people downvoting you without arguing the math is something that should be left to twitter and reddit.

      • Urist@lemmy.ml
        link
        fedilink
        English
        arrow-up
        6
        ·
        10 months ago

        Thanks. It was a bit poorly worded, but I do think the original statement is wrong and just wanted to sketch an idea of why.

    • ඞmir@lemmy.ml
      link
      fedilink
      arrow-up
      4
      ·
      10 months ago

      Even funnier in your example is that the chance of any real number ever being picked is infinitesimally small, instead of guaranteed.

      • Urist@lemmy.ml
        link
        fedilink
        English
        arrow-up
        4
        ·
        edit-2
        10 months ago

        Yep! Relatively speaking almost none of them will be picked. The same is also true even if one had a countable infinite amount of machines trying to pick these numbers.

    • GladiusB@lemmy.world
      link
      fedilink
      arrow-up
      4
      ·
      10 months ago

      I hear what you are saying and agree. I never took the monkey Shakespeare theory seriously. It sounded a bit too poppy and quite honestly the guy that told me was a douche and pronounced giblets wrong. But as a theory you could get anything in a long enough time span and infinite amount of resources. Why or how that matters? Well I just don’t see it.

      • Urist@lemmy.ml
        link
        fedilink
        English
        arrow-up
        3
        ·
        edit-2
        10 months ago

        Of course I am not denying that anything possible could happen. That is contradictory to the assumption it was possible in the first place. What I am saying is just that not all that is possible will happen, even if given an infinite time to do so.

        EDIT: Unfortunately, given a setup like this the math says monkey Shakespeare will almost surely happen due to there only being finite variations.