On 1/12/2017 4:10 AM, clipka wrote:
>
> For me the question is not whether every conclusion in this video's
> train of thoughts is correct and irrefutible, but whether it leads to an
> interesting perspective on information and entropy that warrants to be
> pondered further.
>

Indeed I think it does... although maybe it's partially lost on me since 
I was lucky enough to first learn about entropy in 
semi-information-theoretic terms anyway.  Anyway, if you have the time 
there's a lecture I like by someone who actually *is* a top theoretical 
physicist related to these concepts and how they relate to black hole 
holography.  Maybe this is a good example of where you can get by 
pondering this stuff further (in addition to being substantially more 
precise about things): https://www.youtube.com/watch?v=2DIl3Hfh9tY

Another interesting related trend I've seen in physics is to start 
looking at the relationship between fundamental physical properties and 
computational complexity.  I know very little about this but it seems 
like a super neat connection between physics and information science.

>
> According to the Kopenhagen interpretation, the wave function does not
> describe a particular state. It describes the /probability/ of a certain
> state.
>

Oh man, the Copenhagen interpenetration is such a can of worms.  I 
actually like to think of it as actually not saying much of anything 
about the nature of reality, just as a way to tell you what you'll get 
as the result of a measurement of a quantum system.  Sort of the "shut 
up and calculate" interpretation of quantum physics.

But taking it as a serious model of reality, I have to admit that I've 
never understood precisely what sort of ontology or lack thereof is 
actually being proposed by this interpretation.  So it's hard for me to 
say much about it.

Nevertheless, I'll try anyway.  Even under the Copenhagen 
interpretation, you can resurrect a notion of state by simply looking at 
the amount of information needed to distinguish one wave function from 
another, and calling a physical instance of this information a "state". 
  It might have some strange more-epistemological-than-ontological 
status, but you can largely evade this to a large degree by focusing 
solely on the "state" as being just that which changes the probabilities 
of what you'll see as the result of an observation.

I think (but I have not done the math) if you do use this notion of 
state, then the information theoretic properties of it will match what I 
mentioned earlier where information can be destroyed but not created.

Anyhoo, you may not call that a "state", but I'm pretty comfortable 
doing so, in which case it's just a disagreement of whether a certain 
English word is appropriate to describe a particular concept, and the 
math is the same either way.

Post a reply to this message