00:00:00.000 So welcome to Topcast and two episode three of the science of Canon cards called information.
00:00:24.760 And this first episode is just going to be an introduction to the chapter, rather than
00:00:31.400 And I'm going to call this the science of information, you know, and then go through
00:00:35.760 some of what we know about information so far without really concentrating too much on
00:00:41.600 constructor theory or the new material that is in this particular chapter.
00:00:46.000 No readings from the science of Canon card today, but I'll have another episode out
00:00:50.320 very, very shortly, which does contain the readings from this chapter.
00:00:55.480 It has some really interesting insights about the link between physics and information.
00:01:01.000 These links have been made before, or I should say, links between physics and the theory
00:01:08.920 But here with constructor theory, there is a new window into seeing the way in which information
00:01:16.240 And of course, information is very closely connected, as we will see, to knowledge,
00:01:21.240 which is an area of interest of mine, which means that we have an interesting connection
00:01:26.000 as already mentioned in this series between physics and epistemology.
00:01:31.840 The work of David Deutsch, like I say, often focuses on knowledge more than information
00:01:36.960 in the beginning of infinity, for example, but how are knowledge and information different
00:01:42.400 Well, before I get to the correct answer, I think we should look at some ways of describing
00:01:47.280 the issue that are probably less than fruitful. Here, for example, is a popular meme type
00:01:58.120 I mean, we're getting some kind of insight here into what possibly the difference between
00:02:03.520 information and knowledge is, knowledge seems to be higher up the hierarchy, so to speak.
00:02:10.160 And in some of these, we're even getting a different species of this way of understanding
00:02:16.120 the world, namely data, or data, on the one hand, and wisdom on the other.
00:02:24.480 And how sharply can we distinguish between these different levels if indeed levels there
00:02:30.160 When I was a senior in high school, I could take a subject which was computer science,
00:02:36.160 And in that subject, we were taught that data was elemental, so to speak, and unorganized.
00:02:42.000 All information was still data, but it was organized now, so your list of data, just
00:02:47.920 raw numbers, could be organized into a table, which turns it into information.
00:02:53.640 So with this particular cartoon, we've got data, information, knowledge, wisdom.
00:03:01.040 Now, data, I used to be told, was just the unorganized version of information, as I said,
00:03:11.040 Maybe just a pose came down to something about it being useful in some way, that information,
00:03:17.120 but we were never really told why, and then wisdom, well, what is wisdom?
00:03:21.440 I think these words do mean something, and I think we could still use these words in
00:03:26.400 a rough way to capture realities about the world, for example, what I mean by that as well.
00:03:37.760 It's always there, it's ready to be captured, but you haven't yet captured it, perhaps.
00:03:44.160 And so the sun is shining right now, and that light that's falling down to the earth contains
00:03:50.800 evidence or data, which if only you knew how to interpret it, if only you knew how to
00:03:55.800 interpret the photons and the information being carried by those photons, then you would
00:04:00.920 be able to construct knowledge, but you would need to collect the data first, using some
00:04:06.240 kind of instrument, and you might not necessarily understand what that data means.
00:04:10.520 When you begin to have an understanding and you begin to put it into tables and graphs
00:04:14.840 and so on, then you have information, and once you finally figure out that there's a problem
00:04:19.840 with your data or with your information, or that that information does or does not agree
00:04:25.360 with a particular theory you previously had, then you start to create knowledge, then
00:04:30.400 you start to solve problems with that information, that's when it becomes knowledge.
00:04:34.440 And what's wisdom in this picture, although I would say that wisdom is basically a moral
00:04:38.120 claim, something is described as being wise, simply when it is morally good.
00:04:44.760 So it's wise to do that thing, means it's good to do that thing.
00:04:48.000 It's unwise to do that thing, means it's bad to do that thing.
00:04:51.040 And I don't know if wisdom has much more cash, beyond that.
00:04:54.800 In so far as it is a form of knowledge, it is a moral claim about the knowledge, how we
00:05:03.440 You have the knowledge perhaps of how to set off a bomb or how to create a virus, but
00:05:09.480 In other words, should you do it, why is just means good, I would suggest.
00:05:14.640 Now these days I would say that that is a kind of distinction which makes little difference.
00:05:20.720 Data is information, as we will come to see, anything one could say about a data fits
00:05:27.040 with what we say about information, what will come to say about information.
00:05:34.120 Well, following the work of David Deutsch, we have, I think, about three roughly equivalent
00:05:40.920 approaches to what knowledge is in terms of information.
00:05:45.520 And so these days I would say something like, knowledge is one useful information.
00:05:51.920 And you might very well ask, what do you mean by useful?
00:05:55.120 Well, that brings us to the second quality that knowledge has in terms of information.
00:06:04.440 And so this is what it means to be useful, essentially those two are kind of equivalent.
00:06:10.320 Three, knowledge is information that tends to get itself replicated.
00:06:17.520 It is the composition that the musician is trying to come up with that does not get thrown
00:06:25.360 It is the data, not filled with noise or errors and so on, that the astronomer, the geologist,
00:06:32.080 the zoologist keeps and preserves and then that ends up in the journal article somewhere
00:06:38.480 Now another way of formulating this, and it's a little bit more verbose, but I actually
00:06:44.320 But it means the same thing as information that gets itself replicated.
00:06:47.480 So I'll call this three A, knowledge is information that once instantiated in some physical
00:06:57.680 And so this is a remarkable feature of knowledge, knowledge is that kind of information
00:07:02.320 which once you've written it down on a piece of paper, well, there it is in a physical
00:07:06.880 You may very well have had that thought in your mind and it's in a physical substrate there.
00:07:11.600 But if you forget it, if it's utterly forgotten for the rest of your life, then it doesn't
00:07:15.480 really form part of your knowledge, let alone everyone else's knowledge.
00:07:19.560 So if you write it down now, it's in a more robust physical substrate and maybe, then you
00:07:24.520 go on to publish it somewhere and it remains there, not only instantiated in the original
00:07:29.080 physical substrate for some time, for so long as that physical substrate remains.
00:07:32.560 But the important thing is it can get copied over and over again.
00:07:37.480 And that brings us to what I would call three B. And this is following what we're about
00:07:42.480 to read in the signs of Canon card by Caramel Ito.
00:07:47.080 And she describes information here as knowledge which is resilient, it's resilient information.
00:07:53.840 And so that's got now something to do with design, it's very much echoing what we just
00:07:58.360 said in that three A definition, but it's the capacity of the system itself to maintain
00:08:05.720 And the very remarkable and deep thing here is that as Caramel explained, knowledge is
00:08:11.240 the most resilient, the most robust thing in the universe that seems to be able to just
00:08:15.280 maintain itself, often to the indefinite future, far more robustly than other physical objects
00:08:23.680 Other physical objects are subject to erosion and decay due to the second law of thermodynamics.
00:08:28.880 The issue there is, there's no error correction.
00:08:32.000 So if you really did want to preserve the rock, how do you go about doing it?
00:08:35.760 Well, the only way would be able to gather information about the rock and perhaps make
00:08:38.880 a copy of it, but of course that wouldn't be the original.
00:08:41.680 Okay, so there are some ways of circling around this knowledge.
00:08:47.360 Knowledge has, of course, as we learned in the beginning of infinity, great reach, explanatory
00:08:55.520 And all of this is very well, but we're still talking about knowledge.
00:08:59.960 We haven't actually gotten to information and the physics of information.
00:09:06.040 There are mathematical theories of information, just information broadly.
00:09:10.360 The most famous is Claude Shannon, which links information to uncertainty or something like
00:09:15.640 degrees of freedom, which means it is directly linked to entropy.
00:09:21.240 Entropy is a concept from thermodynamics, which is about disorder.
00:09:24.240 It tries to quantify the amount of disorder disorder, uncertainty, degrees of freedom.
00:09:29.800 These are all ways of speaking about the same thing in more or less precise terms.
00:09:34.360 Whatever the case, we can quantify how much information we have using this idea.
00:09:40.440 For example, before we toss a coin, we lack information about whether it's heads or tails.
00:09:45.320 When we learn its heads, we've gained some information.
00:09:48.520 If the coin is fair, the chance of it being heads is one and two.
00:09:52.720 To encode that, using information would take one bit, one binary digit.
00:09:58.480 In other words, a zero or a one, zero representing perhaps the head and one representing
00:10:04.120 So because it only takes one bit of information, we can actually say its entropy is one.
00:10:09.440 A way of looking at this is to ask, well, how many questions would a person need to ask
00:10:22.200 Is it a head and if the answer is yes, you know, it's a head and if the answer is no,
00:10:26.840 Hence the entropy or information content is, if you like, one.
00:10:31.600 Indeed, Claude Shannon provided us with a mathematical formula to tell you what the entropy
00:10:38.960 So maybe we'll use this formula just just for the intuitive example of the heads versus
00:10:48.040 And at the risk of losing everyone by explaining how logarithms work, I'll do the calculation
00:10:52.320 only once for this simple case, but not for any more difficult cases.
00:10:55.800 Okay, I know that many of my viewers are quite proficient in physics and mathematics.
00:11:03.160 So you can skip this part because I'm going to do a basic introduction to logarithms
00:11:09.120 and indices for the people who watch me who don't understand this staff.
00:11:14.080 Maybe had a bad experience at school, let's say, and in particular who want to understand
00:11:18.880 Shannon's formula that's about to come, which explains the quantification of information
00:11:23.960 and requires us to have some understanding of logarithms.
00:11:27.480 So anyway, a logarithm, which is something that looks like this, usually written in this
00:11:32.480 form here, is basically the same thing as an indices or an exponents, just two ways of
00:11:39.400 What this thing here means, the logarithm to base two of some number x equals eight means,
00:11:44.840 which number x do we need to raise two to the power of in order to get eight?
00:11:50.960 Or we can write it like this, two to the power of some number equals eight, what is that
00:11:55.800 Well, two to the power of one would be two to the power of two is four, two to the power
00:12:06.680 We're going to need negative indices for what I'm about to explain, so I thought it'd
00:12:09.920 be useful to go through a pattern which can explain, or at least force one to the conclusion
00:12:18.880 that negative indices end up giving us numbers that are less than one, a fractional numbers.
00:12:25.800 And so here are the ones that people are familiar with over here, two to the power of
00:12:29.360 one, two to the power of two, two to the power of three, two to the power of four.
00:12:32.440 So what two to the power of one, two to the power of two means is two times two, two
00:12:37.480 to the power of three is two times itself three times and two to the power of four is two
00:12:43.280 Whatever the case is, we always start with the number two because we're talking about
00:12:49.120 We're halving what the two is being multiplied by.
00:12:52.320 So here we're saying two multiplied by two times two times two times two, which is eight
00:13:01.720 Half of eight, that's four, okay, and so we get two times four, which is eight.
00:13:09.840 And so when we get down to two to the power of one, we're asking what is two times
00:13:16.040 The only reason I'm emphasizing this is if we're halving this number by which two is
00:13:19.960 being multiplied, then the next one logically would be half of one, which is itself a half.
00:13:28.120 The pattern here is we're going, the indices going down by one each time, four, three,
00:13:33.800 Well, the next one, we subtract one again and we get to zero.
00:13:37.280 And you can see what happens here, we subtract one and we get minus one, we subtract another
00:13:41.960 That's where the negative indices come in, but if we're following the pattern over
00:13:46.520 in this column here about the product, then halving the one, leading to a half, then
00:13:53.400 means we need to find half of what a half is, which is a quarter, and half of a quarter
00:13:59.320 All right, so that's that, so this is why this pattern of going down by one each time
00:14:04.200 leads to this pattern over here of halving the number by which two is being multiplied.
00:14:11.760 Well, everyone's familiar with two to the power of four, 16, half of that eight, half
00:14:15.680 of that four, half of that two, half of that one, two to the power of zero is one, which
00:14:20.200 is sometimes surprising to people, but that's simply a fact given the pattern here.
00:14:24.520 In fact, any number, three to the power of zero is also one, four to the power of zero
00:14:29.440 You might want to convince yourself of that if you're not already familiar with this
00:14:34.320 If we have one, because that's what we've been doing in this column here, 16, 8, 4,
00:14:43.200 So two to the power of minus one corresponds to a half.
00:14:48.160 This is the most important part of this table for what we need to understand next with
00:14:56.560 Continuing the pattern though, two to the power of minus two, that's one over four, and
00:14:59.880 the reason that's one over four, by the way, is because that's one over two squared.
00:15:03.480 So we could turn this into a positive indices, so just like this, but put one over that,
00:15:10.720 One over two cubed is one over eight, and so you can just imagine continuing the table
00:15:18.840 So the smaller and smaller the indices or the exponent, the smaller and smaller the actual
00:15:25.200 So this is what indices are about, and this connects then to logarithms, which are just
00:15:31.480 another way of writing, exponents or indices, just another way of talking about it.
00:15:36.760 So keep that in mind as we go forward from here on in.
00:15:40.920 But I think it's useful for people who might not be familiar with mathematics to link what
00:15:45.280 looks like a complicated formula to common sense.
00:15:50.840 Okay, well, H is going to represent H as a function of X. H is going to represent the quantity
00:15:57.400 Okay, so that's ultimately what we're looking for.
00:16:00.600 And that's going to equal the negative of sigma, sigma means the sum.
00:16:08.680 So from whatever our first thing is, all the way up to n, the total number of possibilities
00:16:14.400 Now, in the case of the coin, we're going to have two possibilities.
00:16:19.080 So n is going to be two, and basically that will mean that we're summing two terms
00:16:25.280 Okay, it's the sum of what the sum of, it's the probability of X occurring multiplied
00:16:29.440 by the logarithm of that probability, okay, to some base B. And in this case, we're
00:16:35.840 going to be using the base of two, because we have only two possibilities is the long
00:16:43.160 Okay, so let's go through this in plain English.
00:16:46.560 What this formula then means is that we've got the probability of heads happening, multiplied
00:16:52.760 by the logarithm to base two of the probability of heads occurring, plus because we're doing
00:16:58.520 a sum, so that was our first term, plus the probability of tails happening, multiplied
00:17:03.360 by the logarithm to base two, of the probability of tails occurring.
00:17:07.160 Okay, so that's the plain English way of understanding what that previous very abstract
00:17:15.320 Okay, so, and all I want to do here is to really show you that the formula does indeed
00:17:20.040 lead to the common sense notion that you need one bit of information in order to quantify
00:17:32.920 What is indeed the probability of getting a heads?
00:17:36.520 Well, it's one and two, okay, so there's a negative there, so negative in front of the probability.
00:17:43.200 So now we have that all being equal to the negative of what's the probability of heads?
00:17:49.120 Well, it's a half times the logarithm to base two of a half plus the probability of tails.
00:17:56.320 Again, it's a half, a logarithm to base two, of a half, there we go.
00:18:02.160 And that all equals negative, a half times now what is the logarithm to base two of a half?
00:18:10.480 What this means is you're looking for a number such that if you take two, that's how
00:18:15.200 base, and the reason why it's called the basis, because when you put it into exponential
00:18:19.760 form, okay, exponents and logarithms are the inverse operations of one another, much like
00:18:25.720 multiplication is to division or addition is to subtraction.
00:18:30.360 What we're going to do here is to convert our logarithm into an exponent, or an exponential,
00:18:35.920 we're going to take that two, and raise it to the power of x, and x is indeed the number
00:18:42.400 we're searching for, it is going to be the logarithm of the number.
00:18:45.640 And all of that equals a half, that's where that half comes in there.
00:18:48.920 So we've got two to the power of x equals a half.
00:18:52.640 Because that's going to be the solution to the number that we're actually after, well,
00:18:56.080 it happens to be minus one, because if you raise two to the power of minus one, you'll
00:19:04.800 We've got negative outside of one over two times minus one plus, but we're just going
00:19:10.720 Because whatever we just did there for the case of heads, we're going to do for the
00:19:16.800 It's going to be the probability of a half multiplied by logarithm base two of a half,
00:19:23.800 So this will reduce us to, it's going to equal minus, let's put a bracket there, minus
00:19:32.360 So we've got minus a half, plus minus a half, which reduces to, if you've got minus
00:19:36.440 a half, plus minus a half, or the plus and the minus.
00:19:39.520 The minus wins that battle there, so to speak, and so you end up with minus a half, minus
00:19:43.160 a half is minus one, minus outside of minus one, that's one, that's it, that's the answer.
00:19:49.920 So we've managed to prove that our common sense notion that the amount of information
00:19:54.640 in a coin flip, if you like, is one bit, using Shannon's mathematical description of what
00:20:03.080 information is to quantify the amount of information in something.
00:20:06.440 So we've recovered, so to speak, the common sense understanding using the mathematical
00:20:12.520 So I have up convinced you that the mathematical formula indeed has something to do with
00:20:18.200 But consider if you were rolling a fair dice now, then you've got six possible outcomes.
00:20:23.840 So we could go through this and do it all over again, the chance of rolling a two, for
00:20:28.080 example, chance of rolling any particular number is going to be one over six.
00:20:32.320 So if you wanted to go through and do the n equals six case for one over six and apply
00:20:38.680 the formula, you'll need a calculator for this one, you will end up getting an answer,
00:20:42.760 which is about 2.6. This means that, on average, you need to ask 2.6 questions, binary
00:20:50.120 yes or no questions, by the way, to get the right answer.
00:20:53.680 So you might ask, is the number when you roll the dice, is the number one, two or three,
00:21:01.400 And if they say no, then you would have to ask the question, is it four or five, for
00:21:05.600 example? And if they say no, well, then you know that it's six, so you've had to
00:21:08.800 ask two questions there, but if they say yes to four and five to four or five, then you
00:21:13.760 would have to say, well, is it four and so you've asked three questions.
00:21:16.680 So it's somewhere between two and three questions on average in order to get the amount
00:21:22.200 of information in a dice roll, but on average it's precisely 2.6.
00:21:28.320 And the point of all this, all the point here is that there is more information gained
00:21:32.120 on learning the outcome of a dice roll as compared to a coin flip.
00:21:36.520 One way I would put this in popular in terms is that we have ruled out more upon learning
00:21:43.640 what the dice ended up being as compared to the coin toss.
00:21:48.040 With the coin, we ruled out one possibility only, but with the dice, we ruled out five possibilities.
00:21:55.600 This links nicely to the way David sometimes speaks about the loss of physics.
00:21:59.320 They're in large part about what they rule out.
00:22:02.640 They say what is impossible, what cannot happen, and very good explanations, explanations
00:22:08.040 hard to vary, explanations with a lot of knowledge or information content rule out a lot
00:22:13.960 And all of that is fine and accurate, but we're going to refine it here today.
00:22:19.080 After all, what we've done there is we've talked about information entropy in such a
00:22:23.760 way that it takes very seriously the physical reality of probability theory, as if probability
00:22:34.240 But anyone who's been watching me for a while here or following the work of David Deutsch
00:22:38.040 would know the probability is not a fundamental part of physics, it's not a fundamental
00:22:46.360 So we have to reconcile these ideas about information with physics in some way, while not
00:22:52.720 taking probability seriously as an explanation of what is really going on in fundamental
00:22:58.720 And that, of course, is where constructive theory comes in.
00:23:02.240 We are going to understand what is going on realistically and fundamentally, without saying
00:23:07.280 that a coin has a probability of landing on heads of one half.
00:23:11.680 And on all this, and the motivation I would imagine for this particular chapter in the
00:23:15.880 science of Ken and Kant, is the paper by David Deutsch and Kiara Malito in the proceedings
00:23:20.920 of the Royal Society, a published in 2015, called the Constructor Theory of Information.
00:23:27.440 And I would absolutely recommend that for deep dives into this topic.
00:23:31.240 I'll just read a short part of this paper here at the beginning, quote, Deutsch and
00:23:40.600 In some respects, information is a qualitatively different sort of entity from all others,
00:23:46.200 in terms of which the physical sciences describe the world.
00:23:49.400 It is not, for instance, a function only of tensor fields on space time, as general relativity
00:23:55.040 requires or physical quantities to be, nor is it a quantum mechanical observable.
00:24:00.440 But, in other respects, information does resemble some entities that appear in laws of physics.
00:24:07.040 The theory of computation and statistical mechanics seem to refer directly to it, without
00:24:12.440 regard to the specific media, in which it is instantiated, just as conservation laws do
00:24:18.360 for the electromagnetic for current or energy momentum tensor.
00:24:22.320 We call that the substrate independence of information.
00:24:26.280 Information can also be moved from one type of medium to another, while retaining all
00:24:37.760 It is what makes human capabilities such as language and science possible, as well as the
00:24:42.960 possibility of biological adaptations that use symbolic codes, such as the genetic code.
00:24:49.000 In addition, information has a counterfactual character, an object in a particular physical
00:24:53.560 state cannot be said to carry information unless it could have been in a different state.
00:25:03.600 In communication theory relates not so much to what you do say as to what you could say,
00:25:09.960 end quote from weaver and end quote from Deutsch and my letter 2015.
00:25:14.600 So this is where we'll begin our book reading today, more or less, after a few more comments
00:25:22.560 It's part of the physical world and it has effects in the physical world, so physics
00:25:28.160 Again, this is a motivation for constructive theory because the dynamical laws and initial
00:25:32.400 conditions approach is largely about predictions.
00:25:36.520 But here with information, we have yet another case where in the physical world, we are
00:25:40.800 not necessarily most interested in what did happen and what does happen and what will
00:25:45.480 happen, especially that latter one, we cannot always know this.
00:25:50.000 We want to know what could possibly happen, what might have happened.
00:25:54.800 It's the physics of possibility and thus what could happen is a vast array of possibilities.
00:26:00.760 To know what they are, we need a physics of the possible and impossible, what transformations
00:26:11.600 Now, before we begin the reading, there is a couple of other things I'd like to go into
00:26:17.200 because anyone who studies physics to a sufficient depth encounters information at some point
00:26:23.800 and one of the places in which, even if you just read widely about physics cosmology
00:26:28.800 information, you're going to come across the black hole information paradox.
00:26:33.040 This is something that Stephen Hawking worked on and the whole idea here is that, well,
00:26:38.320 if you take quantum physics seriously, it's a description of reality, which we do.
00:26:42.680 We can say something like, there is a wave function of the universe, there's wave functions
00:26:47.800 of any given object, but there's a wave function of the universe and so the wave function
00:26:52.720 determines what happens at any given point in the future.
00:26:57.800 So the wave function right now is going to determine what happens in the future.
00:27:01.720 But the wave function takes account of all the information that happens to exist in a system
00:27:06.160 that the system is the entire universe, then that system, the wave function of that system
00:27:11.760 right now is going to determine what happens in the future.
00:27:15.000 But that wave function right now must take account of all the information that's going
00:27:18.960 on right now, the positions of the particles, the momentum of the particles, all these quantum
00:27:23.160 properties of particles, they're spin, they're mass, various things, okay?
00:27:28.040 So this can be the information of a system at a particular time is included in the wave
00:27:38.120 The black hole information paradox is, well, what happens in a black hole, black holes
00:27:42.040 are not only predicted by general relativity, but they have been observed.
00:27:46.720 Well, the only explanation for some of the observations we have is that black holes really
00:27:51.800 But our understanding of black holes from general relativity also says that they are a
00:27:55.600 they are a singularity and if they are a singularity then anything that falls into the
00:27:59.720 black hole has its information destroyed living behind, basically only the mass, some other
00:28:06.920 The point is, the information is supposedly destroyed, but if the information is destroyed,
00:28:11.400 then it can't be the case that the wave function at any given point in time is the only
00:28:15.840 thing that determines what happens in the future after all.
00:28:18.920 Some of that information right now, if it's falling into a black hole, is then vanishing
00:28:25.040 So this is a problem, it's called a paradox, but I would just say it's a problem.
00:28:30.240 Long theory, general relativity says the information is destroyed and the other theory, quantum
00:28:34.840 theory says that the information is required in order to determine the future state of
00:28:41.720 Well, again, this is just a problem that we've talked about before on this podcast
00:28:46.720 It's the difficulty of reconciling quantum theory and general relativity.
00:28:53.360 One way that people have tried to suggest that I always found interesting with this is that
00:28:59.120 and I think the movie interstellar tries to represent this to some extent as well and various
00:29:04.880 other science fiction notions have tried to represent this, which is that as an object
00:29:11.280 falls into the black hole, an image of that object remains permanently fixed on the
00:29:20.800 So a 2D image of the 3D object perfectly encapsulates the information that that object had
00:29:28.640 And so the black hole grows and grows and grows and the surface of the black hole serves
00:29:31.920 the event horizon, grows and grows and grows, preserving the information.
00:29:35.040 So the information doesn't get destroyed because it gets preserved at the surface of the
00:29:42.480 And this leads to weird things like holographic cosmologies where if it is possible for all
00:29:49.920 the information of a, let's say, quantum object or any object falling into a black hole
00:29:54.920 to be preserved on the 2D surface of the black hole, so a 3D object can be preserved
00:30:01.360 Perhaps our entire universe is kind of like that.
00:30:03.960 Our entire universe being a universe of three dimensions of space, one dimension of time,
00:30:09.200 can somehow be represented on a 2D surface of something else.
00:30:15.520 And so that is the way in which this 3D universe can exist inside of a holographic type
00:30:24.800 But anyway, this black hole information paradox, as far as I know, it's a real thing that
00:30:30.120 physicists are working on and it's yet another problem as to why quantum theory and
00:30:46.240 And perhaps constructive theory can help with this as well.
00:30:50.480 These various popular videos on YouTube, you can read about the quantum information paradox.
00:30:57.600 If you want to watch a video where someone says, there's no problem and every single solution
00:31:03.280 that's ever been suggested for this is completely fallacious and doesn't work.
00:31:07.360 Look at Sabine Haas and Phil, this video, Sabine Haas and Phil Ders, I'm a video.
00:31:14.160 So I was looking through popular accounts of this, I stumbled across a video, which of
00:31:18.400 course, as you can see by the title, is going to attract someone like me because it says
00:31:23.760 the black hole information loss problem is unsolved.
00:31:28.720 She likes that kind of thing, you know, it's a, it's kind of clickbaity.
00:31:33.960 A lot of other videos are kind of like this yet.
00:31:38.320 So if you just get to seven minutes, 20 of that particular video, she goes through one,
00:31:48.840 She goes through something like 10 different attempts to solve this and just dismisses
00:31:54.600 them by waiting her hands essentially, or kind of mockingly describing them, just listing
00:32:00.160 them one after another as if they're all equivalently silly, you know, number seven there
00:32:05.320 is a paper by by Gerard to hoof to you who's not a, you can easily just dismiss like that
00:32:11.480 as if he's writing nonsense. He is not someone who writes nonsense. In fact, I don't
00:32:17.640 know all the physicists who she's dismissing that easily, but this is a habit of hers.
00:32:22.800 I would say she likes to not necessarily present her own ideas about things, but present
00:32:29.880 other people's ideas and then say, what's wrong with all those ideas and usually not
00:32:34.120 in a substantive way, she just argues from incredulity, just says, well, I don't believe
00:32:39.560 this. I think this is either religion or nonsense or gobbledygook, as she likes to say.
00:32:45.520 She likes to talk about physics without the gobbledygook. That's in fact how she introduces
00:32:50.000 that particular video, but of course what she actually means is I'm going to make a video
00:32:55.280 about gobbledygook without really describing much of the interesting physics, at least
00:32:59.400 that's my feeling and watching some of her videos, indeed in this particular case, she
00:33:04.280 says, with regards to the ten different theories that she very quickly dismisses as not
00:33:09.800 possibly being solutions of the black hole information paradox, is that it's not a matter
00:33:15.600 of objectively choosing among the solutions. In fact, it's just arbitrarily choosing which
00:33:21.240 one you prefer, which one you like best. Now, of course, I don't understand all of these
00:33:26.480 ten theories, so I'm not in a position to try and objectively assess them, but I'm sure
00:33:31.840 if I sat down for a while and really studied it for a long enough time, I could figure
00:33:36.000 out that they're not all unequal footing and can't be that easily dismissed. This is kind
00:33:40.680 of her habit I don't understand until I'm doing that, so then, but she does have this habit
00:33:45.160 of what's during interviews and even seeing her in discussions with other physicists.
00:33:50.120 And she is quite critical, which is very good. It's a very good, popularion attitude to
00:33:55.600 have, but on the other hand, she tends to straw man and denigrate on the basis of her not
00:34:02.440 liking, not preferring, a particular theory, which is, of course, not popularion at all.
00:34:10.360 We need to criticize, but we need to criticize the strongest possible version of any given
00:34:15.120 theory. And we can only evaluate the theory when put in its strongest possible terms, which
00:34:21.640 I don't think she necessarily does. She tends to summarize the ideas of others in a way
00:34:26.840 that lacks generosity, and then on the basis of her personal summary, rather than the actual
00:34:32.080 theory itself, makes some sort of moral call, like, for example, such and such, as nonsense
00:34:36.960 or such and such, as gobbledygook or such and such is religious when it comes to certain
00:34:41.160 versions of the multiverse, let's say. And near the end, she actually claims that the
00:34:46.080 problem here with the black all information paradox is that too many theoretical physicists
00:34:51.200 think that physics reduces entirely to mathematics, and it doesn't do this. I agree. And
00:34:56.840 of course, somehow like this, of course, you need to have testable predictions in physics,
00:35:01.360 or you should hope to have testable predictions in physics. And she claims at the end of
00:35:05.560 the video that not only is there no data here, but there's even in principle, we can't
00:35:10.520 gather the data to observe Hawking radiation directly. And this is a problem inherently
00:35:15.800 for all practical purposes of ever figuring out which of these theories, if any is true.
00:35:20.400 And this is why she says it's insoluble. It's insoluble because in practice we can't directly
00:35:25.560 observe, we can't observe, the Hawking radiation, the very thing that would help us to figure
00:35:30.400 out how to distinguish to rule out some of these theories in favor of others. Now this,
00:35:35.760 to me, is, of course, rank empiricism. She wants to observe Hawking radiation in order
00:35:42.480 to rule out all the other theories. It exactly wreaks to me of wanting to observe dinosaurs
00:35:48.680 in order to establish that dinosaurs really exist or to try and figure out theories about
00:35:52.680 dinosaurs. All we have accesses to with fossils and fossils tell us all about dinosaurs.
00:35:58.280 Not everything we would want to know, but they tell us a lot about dinosaurs. We do not
00:36:01.400 need to travel to the center of the sun in order to understand stellar nuclear fusion.
00:36:07.040 We do not need to travel back in time, 13.7 billion years and observe the big bee there,
00:36:12.160 the big bang in order to know the big bang existed. We have other forms of evidence. There's
00:36:16.440 absolutely no reason why Sabine should rule out other forms of evidence that might arise
00:36:22.040 in order to rule out these particular theories. In fact, rule out all the theories except
00:36:26.680 for one, which actually explains not only Hawking radiation and the black hole information
00:36:31.160 paradox, but new evidence yet to be found. But there we go. I think this is a complete
00:36:38.000 misunderstanding of what observation is and what its purpose is in science. So while many
00:36:43.680 physicists are in her opinion too hooked on mathematical models, I tend to think a deeper
00:36:49.600 problem actually at times as too many physicists are empiricists. Certainly Sabine is no doubt
00:36:55.920 she's an absolutely competent physicist, no problem with that. Having watched many videos
00:37:00.120 of her, she is, I gather an instrumentalist of a kind, but then of course she's in good
00:37:05.200 company being an empiricist and an instrumentalist. I think I'd like maybe to do a reaction
00:37:11.120 video one day of her, the trouble with many worlds video. It's deeply misconceived, but it's
00:37:18.240 one of those cases where you begin listening in almost every sentence is something one
00:37:21.840 can object to and reveals deep misunderstanding. So I'm not sure how illuminating it would be
00:37:27.280 is the first thing, but the second thing is I don't know how fun it would actually be
00:37:31.360 from the ultimately. The reason for that is she just seems kind of angry when she doesn't
00:37:35.720 many of her videos, she gets very frustrated with other physicists doing work or coming up
00:37:42.040 with ideas that are new and creative and she just misses them. Now for one thing in this
00:37:46.200 particular video, she talks about the universe is splitting, which as we know is wrong and
00:37:51.280 we've spent a lot of time explaining why that is wrong here. She thinks that probability
00:37:55.720 is fundamental in some way, including in the many worlds interpretation and somehow her
00:38:00.040 eight minute video of which I must say only about three minutes are actually devoted
00:38:04.480 to discussing the many worlds is supposed to be a complete refutation of, well, this entire
00:38:10.160 book by Wallace, among other things. So like I say, it's kind of a straw man, which might
00:38:17.360 be ungenerous of me to say, but that's a theme in the videos of hers that I've watched
00:38:22.120 and her interactions with other physicists sadly. And I should say she does go through other
00:38:27.320 interpretations as well, but she dismisses them all likewise, you know, none of them
00:38:33.080 are satisfactory to her. And this is again, she's in good company. This means that she
00:38:37.360 falls back basically on instrumentalism of a kind. Okay, so that's enough of that.
00:38:42.200 And that's enough of my introduction to the chapter, the science of information, if you
00:38:48.360 like. Next episode, we will do some actual reading from the chapter information and find
00:38:58.200 ways in which constructive theory comes directly to bear on this question of what information
00:39:03.440 is, how it, how it interacts with physics. Until then, bye bye.
