Time?

[Knotted]

It's a metaphor, not a cop out! And it's a bit rich of you to complain. Didn't you say (of your own, determinist philosophy) The fact that it is problematic is no worry to me. But no amount of reductive analysis enables one to understand a movie, or a poem. I mean, cut me the same slack, huh?

If every random event was treated as just an empirical given not to be questioned or explained, then science would never have got off the ground. I'm just saying that there's no particular reason to give up yet!

You seem to hold that information can neither be created or destroyed. That's by no means a certainty is it? Readers may think about the planet now, compared to a few thousand million years back. There's a lot more going on, a lot more detailed events happening. Is there really no more information at play? That's not an easy thing to believe.

But the earth is not a closed system. During that time it has received a lot of low entropy energy from the sun. Life is a very strange thing and civilisation is even stranger, but in terms of thermodynamics they are unspecial.

I'm sure you have a clear idea of what you mean by information. But there is a sense in which the beholder ("reader") decides what is and what isn't information. Remember the old joke about punch cards -- the less card there is left, the more information it contains! More generally, the information contained in a binary file (for example) depends on how it is read. If it were a program file, it could, in theory, result in quite different programs when run on different systems.

This is a very good point that's difficult to answer. Yes there is a distinction between information and information that we can use, which is surely related to the amount of information we have about the type of information we are looking at. To understand what I am writing, you need to know how to read English for example.

I believe that there is enough information in the universe to predict its future states. I don't believe that all this information is accessible to us. I certainly do not believe that it is possible for the human race (presumably with the aid of powerful computers) to be able to know and predict everything about every particle in the universe.

Nah, this does not follow. It really depends on what you want your science to do. Anyway, consider that the addition of random noise to otherwise determinist processes would result in a non-determinist universe. The conventional interpretation is that phenemena such as radioactive decay do just that. You don't seem to accept that

Certainly scientists have no way of predicting the decay of an atom but I honestly don't know how many believe that it is not a deterministic process.

As I understand it, any experiment with quantum particles assumes that the particles in question are not entangled with their environment or rather that the entanglements 'average out'. Given that the experimenter is already applying a statistical assumption at the start of the experiment, it is unsurprising that the result has a random component.

Or what about meteorology? Chaos will always limit how far ahead we can make accurate weather predictions. But I don't see that as placing a pessimistic restriction on questions of meteorology. The fact we cannot minutely predict the weather in Brighton ten years in advance hardly matters. There's still a very great deal to learn about the dynamics of weather systems, on other planets too.

I'm not so concerned with the question of how much science can predict but how much science can explain. A chaotic system is, of course, perfectly deterministic there is no mystery here, but yes it is practically impossible to predict a month ahead nevermind a year.

If the weather were not just chaotic but random - ie there were processes which we cannot access or completely understand - then this is asserting that the explanatory value (as opposed to the predictive power) of science is absolutely limited. I don't think such pessimistic assertions are warranted.

Agreed, the operation of the brain is dependent on notions of time and space. That's what makes understanding time such an interesting project, and demands that consciousness be included.

Time does seem crucial to consciousness (whatever consciousness is). To fly off at a tangent, Hegel built a philosophical system using Zeno paradoxes in order to resolve the question of the object and subject in history. I've always thought of his version of the dialectic as a distorted version of the laws of thermodynamics. [There is a bit more too it than that but less than you might think.]

 

[Jonti]

You seem to hold that information can neither be created or destroyed. That's by no means a certainty is it? Readers may think about the planet now, compared to a few thousand million years back. There's a lot more going on, a lot more detailed events happening. Is there really no more information at play? That's not an easy thing to believe.

But the earth is not a closed system. During that time it has received a lot of low entropy energy from the sun. Life is a very strange thing and civilisation is even stranger, but in terms of thermodynamics they are unspecial.

Sure, of course I understand that all this information at play in the world's biosphere requires an energy source to drive it.

But we're talking about information. What is the relationship between entropy and information? The first inexorably increases -- the second, you say, can be neither created nor destroyed.

 

[Knotted]

Sure, of course I understand that all this information at play in the world's biosphere requires an energy source to drive it.

But we're talking about information. What is the relationship between entropy and information? The first inexorably increases -- the second, you say, can be neither created nor destroyed.

I must admit that I'm not entirely clear on the relation between Shannon entropy (information) and thermodynamic entropy, but physicists seem to believe the two are strongly related.

If you think of the sun as a vast machine converting small atoms such as Hydrogen into larger atoms such as Helium, Lithium, Carbon etc. As the atoms fuse into larger atoms the degrees of freedom of the system are reduced as there is no reverse procedure so the amount of information stored in the sun is reduced ie. entropy increases. We pick up some of this loss via radiation.

I have only objected to the idea that the amount of information in the universe increases. I think generally it should decrease in accordance with the second law of thermodynamics.

 

[Jonti]

Under conditions of strict determinism (an idea that only makes sense if exact simultaneous values can be assigned to all physical quantities) information could not be destroyed.

However disorderly the scene, the original info is still in there -- one needs only to run the tape backwards, so to speak, to recover it.

 

[Knotted]

There are systems which are deterministic but destroy information.

Consider a universe which is a Turing Machine and at the start its memory contains the last message from God (encoded into binary). The machine is set up to systematically go through the memory and wipe it - it does this by changing any bit storing '1' into a bit storing '0'.

Now of course you wind the clock back and recover the message but this is a process where information is added to the universe. The situation is even worse if God's last message is a 'random number' such as the number Omega from algorithmic information theory where there is no computational procedure to calculate it - so the poor Turing Machine would have to be retold all the original information bit by bit and the reverse process of inputting the information would not follow any deterministic rules even though the message is well defined.

Furthermore I don't think the second law of thermodynamics does destroy information. Recall that it is a macroscopic effect and not a microscopic effect. I believe all the information of a system is preserved microscopically but not macroscopically. Essentially the second law of thermodynamics does not describe physical reality but an exceptionally good statistical approxiamation of a physical reality where the exact mechanics of the particles involved are unknown to the observer - ie. random.

 

[Knotted]

and the reverse process of inputting the information would not follow any deterministic rules even though the message is well defined.

Thinking about it there are deterministic but non computable rules for computing Omega.

 

[Jonti]

Furthermore I don't think the second law of thermodynamics does destroy information.

Ah, that's what I thought , but here you said

I think generally (information in the universe) should decrease in accordance with the second law of thermodynamics.

Everyone knows about information. As Donald Knuth put it, it's "the meaning associated with data...". It's timetables, and shopping lists, and recipes. CDs and DVDs too. And stuff like DNA. All these are information, and are all meaningful only in association with a reader. Then there's the abstract mathematical concept of a scalar defined by log(number of states) of some system. There is a relationship between the two ideas, I'm sure. The point is that there's (at least) two quite different ideas here.

There is a perfectly good sense that we may assert the information in our DNA did not exist a billion years ago. We've been talking at cross-purposes, I think, interesting as it is

 

[Knotted]

Ah, that's what I thought , but here you said

Well I did say I wasn't sure. However the fact that the particles in the sun contain less information does not mean that information is being destroyed as the sun is chucking out a lot of low entropy energy.

Everyone knows about information. As Donald Knuth put it, it's "the meaning associated with data...". It's timetables, and shopping lists, and recipes. CDs and DVDs too. And stuff like DNA. All these are information, and are all meaningful only in association with a reader. Then there's the abstract mathematical concept of a scalar defined by log(number of states) of some system. There is a relationship between the two ideas, I'm sure. The point is that there's (at least) two quite different ideas here.

There is a perfectly good sense that we may assert the information in our DNA did not exist a billion years ago. We've been talking at cross-purposes, I think, interesting as it is

Yes.

By the way where did the Knuth quote come from? Just curious.

 

[Jonti]

It's something I'd heard before -- but I came across the attribution recently a piece called "Finite Nature". I think you might enjoy some of the ideas in that paper. I can't seem to run down an actual quote -- it's possible it's something he said, rather than wrote.

 

[Knotted]

It's something I'd heard before -- but I came across the attribution recently a piece called "Finite Nature". I think you might enjoy some of the ideas in that paper. I can't seem to run down an actual quote -- it's possible it's something he said, rather than wrote.

Thanks!

Glancing at the article its a bit disappointing that there is no real discussion on quantum computing but it looks interesting enough otherwise.

 

[Jonti]

You may care to take a look at Information: The New Language of Science written by Hans Christian von Baeyer (originally published by the Harvard University press). There's a section on Quantum Information.

According to a review in Nature it's "An accessible and engaging overview of the emerging role of information as a fundamental building block in science".