The Big Bell Test

Discussion in 'Physics & Math' started by James R, Nov 30, 2016.

  1. James R Just this guy, you know? Staff Member

    Messages:
    29,915
    Want to take part in an actual physics experiment?

    Today only (30 November 2016), you can contribute to a test to see whether quantum mechanics is truly incompatible with local realism.

    This is being done by testing Bell's theorem using human-generated random data from the internet.

    To participate, and to find out more about this, go here:

    thebigbelltest.org
     
    danshawen likes this.
  2. Google AdSense Guest Advertisement



    to hide all adverts.
  3. cluelusshusbund + Public Dilemma + Valued Senior Member

    Messages:
    6,295
    I got to the 4th level but then my typin fangers wore out.!!!
     
  4. Google AdSense Guest Advertisement



    to hide all adverts.
  5. James R Just this guy, you know? Staff Member

    Messages:
    29,915
    I got through all the levels. But that's not the point.

    From the leader board on the site, it looks like some people have written bots to do it, which kind of defeats the purpose.
     
  6. Google AdSense Guest Advertisement



    to hide all adverts.
  7. cluelusshusbund + Public Dilemma + Valued Senior Member

    Messages:
    6,295
    I got a feelin that the results woud be the same whether humans only... or bots programed to randomly tap keys -- only took the bell test... just like if a human flipped a coin 1000 times or a robot flipped a coin 1000 times the results woud be about the same number of heads as tails... whether anybody was watchin or not.!!!
     
  8. James R Just this guy, you know? Staff Member

    Messages:
    29,915
    The problem with computers doing it is that most computers generate pseudo-random numbers. That is, they use a mathematical formula to calculate what the next "random" number will be, based on some initial "seed" value.

    The point of this experiment is to use what is assumed to be a better generator of randomness - or at least a theoretically more unpredictable one. Hence the human involvement.

    There are mechanical randomness generators based on various physical processes rather than on mere computation, and those are also being used in this experiment some of the time. But that's been done before. The unique feature of this one is using human-generated "random" data.
     
    danshawen likes this.
  9. sculptor Valued Senior Member

    Messages:
    3,801
    Can a (sane?) human "generate 'random' data"?
     
  10. cluelusshusbund + Public Dilemma + Valued Senior Member

    Messages:
    6,295
    Just as good as an "insane" person.!!!
     
  11. cluelusshusbund + Public Dilemma + Valued Senior Member

    Messages:
    6,295
    Are ther computers that can generate real random numbers.???
     
  12. Kittamaru Giving up this insanity for Lent - back ~ 4/15 Staff Member

    Messages:
    11,901
    You can write some pretty complex code that generates "pseudo" random numbers... but even then, it's still based on something, so it isn't truly random. Case in point, a lot of "random" generators will use the system clock and an algorithm of some sort to generate what appears to be a unique number (current Millisecond timestamp and some equation involving the current "step" of the processor cycle and bam, you have a number that appears random but that could actually be duplicated if the same input criteria was used). True random is, insofar, out of reach of computers.
     
  13. James R Just this guy, you know? Staff Member

    Messages:
    29,915
    I guess it depends on what you regard as random.

    The assumption of the Bell Test is that humans are unpredictable in lots of ways. It would be very hard to sort out why I chose to hit "1" rather than "0" in one of the Bell Test games at any particular time, even though, in principle, there might be deterministic reason.

    Any computer that uses a pure mathematical algorithm won't be producing true random numbers, because if you start the algorithm with the same initial conditions then the computer will produce the same string of supposedly-random digits.

    But there are various ways you can couple a computer to another physical system that produces better randomness. For example, the computer could look at the decay of a radioactive source (e.g. a geiger counter), which is totally random in certain respects, as far as we know. And a computer coupled to a random human brain might give good randomness...
     
  14. Kristoffer Giant Hyrax Valued Senior Member

    Messages:
    1,204
    I unintentionally double tapped at one point, which made it into something I didn't intend it to be.

    That's pretty random I suppose.
     
    sculptor likes this.
  15. sculptor Valued Senior Member

    Messages:
    3,801
    Could a machine also make a "random" mistake?
     
  16. cluelusshusbund + Public Dilemma + Valued Senior Member

    Messages:
    6,295
    If ther actualy is such a thang as random.!!!
     
  17. danshawen Valued Senior Member

    Messages:
    3,335
    Which means that at some point, even if this simulation demonstrates the validity of Bell's Theorem, someone will attempt to repeat the experiment with those pseudo-random numbers and expect a different result. Like picking either a red ball or a green ball out of a black bag, or an "up electron or quark" or a "down" electron or quark out of the same bag by means of a random process. If both states / colors / entanglement states are equally likely, what is that supposed to prove again? What is that definition of insanity everyone keeps quoting?

    Those states, if entangled, are different aspects of the same wave function, or like different sides of the same spinning coin. You can never know for certain if they were correlated or not without observing them for a long while, and even if you do that, the entanglement might decay at some instant, for reason(s) that may or may not be observable.

    But if you are one of two people riding a seesaw, you know very well whether you are up or down, and even blindfolded, can know with certainty what the state of the other person of equal weight riding the other end of the seesaw is, can't you? If you are one of those children behaving / misbehaving in that classroom, you can predict what the entangled child in the other or even the same classroom is doing, can't you? I don't see that there is even a problem to be solved here. Bell's Theorem, if anything, is what is insane. Of course there are hidden variables. Most of the events in this universe are hidden from a given observer, whether they acknowledge the fact of their ignorance or not.

    There is something significant about there being two entangled observers that is completely lost (apparently, along with parity) when there are three. Unless you simply acknowledge that the entangled pair is really a single observation, and then a third observer is actually only a second observer. That works in a consistent and reducibly complex manner, doesn't it? Pair up the misbehaving children and send them to the Principal to see to their backsides.
     
    Last edited: Dec 11, 2016

Share This Page