Skip to main content

View Diary: Running Against Paul Ryan is Exactly What Team Obama Wanted (Updated) (167 comments)

Comment Preferences

  •  so am I, "and yet, it does move." (digress alert!) (0+ / 0-)

    When I was a kid, I was a Newtonian mechanist and ran into this thing entirely unexpectedly.  There's no way in hell I would have believed any of it, but for first-hand experience including statistical testing with random-number targets.  And yes, I am fully aware of confirmation bias and similar errors of reasoning, and took steps to minimize them.    

    Here's a thought experiment for you:

    Two teletype machines connected with a simple communication circuit between them.  Milliwatt-hour meters connected to the power supplies for teletype 1, teletype 2, and the communications circuit.

    One punched paper tape containing ASCII characters.

    Run the tape through teletype 1 and observe what is printed out on teletype 2.   Record the power consumption of each element in the system.  

    When the tape is run through teletype 1 in one direction, what is printed out on teletype 2 is unintelligible gibberish.

    When the tape is run through teletype 1 in the opposite direction, what is printed out on teletype 2 is a Shakespeare play.

    The ASCII characters are identical.  The power consumption for both runs is identical.  Yet one run conveys semantic information and the other does not.

    QED: There is not a thermodynamic entropy penalty for the transmission of semantic information.  Semantic information, unlike Shannon information, is orthogonal to thermodynamics.  

    Vector time (directional time; "the arrow of time") arises from thermodynamic entropy.  However in kinetics, time is scalar (non-directional), on the classical scale as well as the QM scale.  Certain equations work whether the time variable is set to + ("forward" time) or to - ("backward" time).  (Apparently time is also reversible for optics, acoustics, and stochastic processes, and what I referred to as "kinetics" is more broadly considered as "mechanics":  )

    Yes, this thoroughly conflicts with "common sense," but eventually all of it (including the messy bits involving brains as detectors of information) will be explained in purely physical terms.  I'll also stick my neck out and say that it will not require "hard determinism" or "superdeterminism," though that position starts from a subjective bias in favor of free will.

    The thing to do when faced with an apparent anomaly, is not to try to wish it away, but instead to figure out the physical mechanisms involved.  That way of proceeding has a pretty good track record so far;-)

    "Minus two votes for the Democrat" equals "plus one vote for the Republican." Arithmetic doesn't care about your feelings.

    by G2geek on Mon Aug 13, 2012 at 01:31:01 AM PDT

    [ Parent ]

    •  Conclusion of thought experiment: (0+ / 0-)

      "information is not a physical property, but rather a description of configuration".

      Information is a description of an arrangement of the configuration of matter/energy from which deductive interpretations can be made. That is not a physical property like mass. One arrangement of a set of blocks may spell out "eat cookies", while another may not, but that doesn't mean that the physical nature of the blocks is in any way different. And, of course, if you measure the blocks, you find that the ones arranged into letters are identical in every way to the ones that are not, because the physical characteristics of the blocks aren't relevant to whether or not one carries information.

      It is therefore a gigantic and unwarranted leap to draw any conclusion about information by trying to apply laws relating to physical properties. And certainly there is zero reason to believe that information lives in a magical parallel universe where time is not a vector.

      We're not going to agree on this point. When it is published in Nature or Science and broadly agreed with by the consensus of cosmological physicists of the day, I'll sign on.

      Have a flagon and discuss the news of the day at the sign of the Green Dragon

      by Dracowyrm on Mon Aug 13, 2012 at 01:28:51 PM PDT

      [ Parent ]

      •  watch out for the goalposts: (0+ / 0-)

        "Theory translated to hypotheses supported by empirical data" is an objective standard.  Pre-publication peer review is a check on the quality of the original work.  Replication is validation.  All of those, so far, are objective criteria for the correctness and completeness of theories and hypotheses and experimental results.

        So yes, "published in Nature or Science" works for me, and I'll see you when we get there.  

        "Consensus of opinion" is a purely subjective standard.  So, no, the ability to hide in a like-minded crowd doesn't make one's theories "more correct" than otherwise.  


        Thermodynamic entropy increases as a function of the quantity of bits transmitted.  Shannon information consists of ordered bits as compared to random bits.  Semantic information consists of yet a further increase in the order of the bit-stream to convey semantic meaning.  

        There might arguably be an entropy penalty for creating the initial ordering, and it can be operationalized in terms of glucose consumption of the brain of the person who is doing the ordering (e.g. Shakespeare writing the play).  Here we would compare drop in blood glucose level over the course of producing a highly ordered bit-stream compared to just sitting quietly and producing nothing, or copying random characters from one page to another.  The difference will probably be insignificant.  

        In any case, having written the play, there is no further entropy penalty for transmitting that bit-stream an infinite number of times, as compared to transmitting an equivalent quantity of random bits.  That kind of multiplier effect is truly extraordinary.  

        One could also do the blood glucose test with the person who receives the bit-stream to ascertain if there's an increase in glucose consumption for reading semantic information compared to reading a random bit stream.  Once again, I'd predict the difference is insignificant.

        If as you say, configuration is not a physical property to begin with, then both Shannon information and semantic information are not physical properties.  In which case they also don't give a hoot about thermodynamics.  

        Meanwhile here sit we, respectively writing and reading semantically meaningful ordered bit-streams.

        Quite a bit of time humans spend engaged with something that might turn out to be, as you said, nonphysical.

        Though I'm still betting on it being physical.  And here I'm following from John Wheeler's "it from bit" theory (information as the fundamental substrate of the universe), made more comprehensible to me by someone I know who spent a decent amount of time working under Wheeler.  (Speaking of papers published in respectable journals;-)

        "Minus two votes for the Democrat" equals "plus one vote for the Republican." Arithmetic doesn't care about your feelings.

        by G2geek on Tue Aug 14, 2012 at 01:42:12 AM PDT

        [ Parent ]

Subscribe or Donate to support Daily Kos.

Click here for the mobile view of the site