Information conservation and Big Bang

[anorlunda]

I just finished Leonard Susskind's video course on cosmology. A question never raised nor mentioned in the course was conservation of information. I mean information in the sense of Liouville's theorem/Unitary operators in QM/Event horizon paradoxes ...

At one point Susskind said that the quantity of information and the number of photons in the universe are roughly the same thing. But I did not hear him say that the number of photons+particles was anywhere near constant during inflation and the big bang.

My question: Should we believe that information was conserved during the big bang and the inflationary period?

 

[Simon Bridge]

Do we have any reason to?
I don't think the question has been resolved yet.

 

[Chalnoth]

At one point Susskind said that the quantity of information and the number of photons in the universe are roughly the same thing. But I did not hear him say that the number of photons+particles was anywhere near constant during inflation and the big bang.

The number of photons+particles has varied greatly over the history of the universe.

My question: Should we believe that information was conserved during the big bang and the inflationary period?

No. Information is not a conserved quantity.

 

[Chronos]

Depends on how you define 'information'. That is not an easy task.

 

[Duhoc]

The universe as an information system for an outside user

Have you ever considered the idea that every single event that has ever happened in universal history is preserved as traveling light. We see the starlight of the past, but is it possible to resolve this light into physical events; or can a universe user simply access any universal event by inputting a coordinate of time and space? If entangled particles can defy the limitations of the speed of light, why can't other information?

 

[Chalnoth]

Have you ever considered the idea that every single event that has ever happened in universal history is preserved as traveling light. We see the starlight of the past, but is it possible to resolve this light into physical events; or can a universe user simply access any universal event by inputting a coordinate of time and space? If entangled particles can defy the limitations of the speed of light, why can't other information?

Quantum decoherence causes information to be lost to the environment, so even in principle all of the information of any given event just won't be accessible for any real observer.

If there were, somehow, a way to observe the full quantum wavefunction, then it would be possible to reconstruct any event. But that's not possible.

 

[Duhoc]

Well we might restrict it to processed information that has already been observed.

 

[Chalnoth]

Well we might restrict it to processed information that has already been observed.

I don't see how that improves things. It's simply not possible to retain information, of any sort, forever. The gradual increase in entropy prevents this.

 

[Duhoc]

If I had a big enough telescope could I or could I not receive the information from the past? Of course you could. We aren't interested in the possibility of all events just the history of events in which the wave function has been collapsed. But anyway my suggestion was to access the information by a cosmic computer that could process information faster than the speed of light.

 

[Chalnoth]

If I had a big enough telescope could I or could I not receive the information from the past? Of course you could. We aren't interested in the possibility of all events just the history of events in which the wave function has been collapsed. But anyway my suggestion was to access the information by a cosmic computer that could process information faster than the speed of light.

There's no such thing as a computer that can process faster than the speed of light.

You could certainly receive some information, but it's always going to be limited, and there's no guarantee you'll get information about every event.

 

[phinds]

... If entangled particles can defy the limitations of the speed of light, why can't other information?

Entangled particles do NOT transmit information faster than the speed of light. You need to read up on quantum entanglement. You are confused by the fact that the state of a particle changes instantaneously when its entangled pair is "read" but that does not transmit any information in any way that the term is normally used and certainly you can't communicate faster that c with it.

 

[anorlunda]

No, this is not a philosophical question, nor is it about everyday definitions of information. I recommend that you listen to Susskind's lecture 1 on Statistical Mechanics (available here on youtube.com) for some very educational and entertaining explanations of the basic definitions of systems and information.

At the quantum level it has nothing to do with superposition or decoherence. It means, for example, that if a quantum system has two states it will always have two states, not any other number. That is a sense of information.

Susskind calls conservation of information the -1th law of thermodynamics. That is what I was asking about relative to inflation and the big bang. If it did not apply back then, that would be remarkable.

 

[Simon Bridge]

Well the big bang is of so much interest precisely because it is remarkable.

At the quantum level it has nothing to do with superposition or decoherence. It means, for example, that if a quantum system has two states it will always have two states, not any other number. That is a sense of information.

At the quantum level, what information you can obtain about the ultimate origins of a particular bit of light has everything to do with superposition and decoherence ...

If a quantum system has two states wrt an observable, once you have measured one state, you have lost the information about the other one - whatever it may have been.

We see the starlight of the past, but is it possible to resolve this light into physical events...

... you cannot resolve the light into unique events. There is more than one way that a particular photon may have come into being. Consider light from the Sun.

I think you need to revisit Susskind's lectures and concentrate on the definition of "information".
His is one view, there are others.

 

[Duhoc]

I will.

 

[anorlunda]

Simon Bridge: If a quantum system has two states, that takes one binary bit to describe the possibilities. One bit is a quantity of information. Which state it is in is not the same thing as the quantity of information.

Two states, one bit. That's the conserved quantity.

By analogy, one character in this message takes 16 bits using Unicode. The quantity of information is 16 bits, which character I choose is a different question.

 

[Chalnoth]

No, this is not a philosophical question, nor is it about everyday definitions of information. I recommend that you listen to Susskind's lecture 1 on Statistical Mechanics (available here on youtube.com) for some very educational and entertaining explanations of the basic definitions of systems and information.

At the quantum level it has nothing to do with superposition or decoherence. It means, for example, that if a quantum system has two states it will always have two states, not any other number. That is a sense of information.

Susskind calls conservation of information the -1th law of thermodynamics. That is what I was asking about relative to inflation and the big bang. If it did not apply back then, that would be remarkable.

It sounds like he's talking about unitarity, which is a statement that if you could know the full wavefunction of the entire universe at one point in time, you could (in principle) compute the exact state of the universe at any other point in time.

Unitarity is pretty important, and rather interesting. But it doesn't have much of any application to the world we live in: we cannot observe, even in principle, the full wavefunction of the universe. In the real world, information is "lost" through thermodynamic processes all the time.

I put lost in square quotes because the information technically doesn't disappear, it just becomes inaccessible.

 

[Chronos]

This reminds me of the epic arguments over the black hole information loss paradox.

 

[Simon Bridge]

This is what I was thinking.

 

[anorlunda]

This reminds me of the epic arguments over the black hole information loss paradox.

Now we're zeroing in. In the original post at the top of this thread, I mentioned both Unitarity, and Black hole information paradoxes. I guess I wasn't clear. Sorry.

I mean information in the sense of Liouville's theorem/Unitary operators in QM/Event horizon paradoxes ...

Liouville's theorum from classical physics say's essentially the same thing; that in time evolutions the number of distinct states is conserved.

I mentioned all those things trying to distinguish this meaning of "information" from the everyday meaning of information as what we know and don't know.