Future of Science and Technology Q&A (April 12, 2024)

Foreign. Hi, everyone. Welcome to another episode of Q and A about history of. Sorry, future of science and technology. So I see a few questions saved up. I hope there'll be a bunch more here. So Johnny asks, what features will humans evolve to have in the future? Will we one day be able to look at the sun with our naked eyes, for example? Relevant for those who saw the eclipse. There's some last this this week. Well, let's see. The thing to understand, I think I've talked about this before, is that the whole process of genetic selection is something that is. It depends on what organisms produce progeny, so to speak. And in natural selection, it's the environment is killing off certain organisms that are less fit and allowing more fit organisms to prosper and have more descendants, so to speak. And that process very quickly, sort of exponentially has the sort of. The ones that are considered fitter relative to the environment and where fitter means have more children, so to speak. So in the sort of natural course of, of biological evolution, the only way that really operates is the who gets to have more children. Now, in the case of breeding and artificial selection, some breeder is determining what, you know, type of, I don't know, dog, snake, whatever else you're breeding or plant for that matter, gets to be the one that has the progeny. In natural selection, it's the world at large and the environment that's determining that. For humans, we've pretty much thought it's a really bad idea to do the equivalent of artificial selection for humans. That was 100 years ago. People were thinking about eugenics, breed the perfect human. Everybody realized that was a really bad idea. And so we don't do that for humans. We do that for farm animals, we do that for pets and so on, but we don't do that for humans. And so with humans, the only way we determine what traits get to be the ones that are more common in the population is by some kind of natural selection. And that natural selection is kind of complicated in the world as it is today because things that were sort of medically devastating in the past can now be medically worked around. So those humans who in the past wouldn't have survived very long wouldn't have had children. Now they can. That's one thing. Then there are other effects which are more social kinds of effects of, oh, it's so expensive to have children in Country X, I'm not going to do it, which is a different kind of sort of selection. And I think, you know, that that makes it, that that makes it have a different kind of dynamic. So, so insofar as, let's say, personality traits or something are. Are genetically passed down, which they are to some extent at least the, you know, if it turns out that all the people who have this personality trait that causes them to be, you know, celibate monks, they who don't have children, they. That those personality traits will tend to decay in the population. Whereas the ones who are like, for whatever reason of personality traits are like, I'm going to have as many children as possible, those traits will expand in the population. So that's the main dynamic of sort of natural selection on humans at the present time. We can certainly imagine other situations where we say, okay, let's do some gene editing on this human. Let's say, oh, you know, somebody might decide that if it was thought to be societally, you know, sort of societally allowed. And some of that already is. It's. Let's, you know, let's edit the fertilized egg cell to change the genetics, to give the kid green eyes or whatever, or to make the kid taller or whatever else, and. Or let's even do artificial selection. Let's say, as has happened in some parts of the world, the embryos, the male embryos, let's select for those. The female embryos, let's select against those. Seems like a bad idea to me, but it's something that's certainly been done. So there's mechanisms like that where you can kind of mess with the genetics of humans. And there's sort of a question of what if we ask what traits will either be naturally selected for or traits that we will consider it societally acceptable to kind of artificially select for? Those are both reasonable questions in terms of the natural selection. I think that's a very complicated issue because I think it relates as much to society as it does to kind of the environmental situation, as it does to other things. I mean, I. I don't think that the sort of. Insofar as, you know, might be the case that if the climate got colder, let's say somewhere, then the people who could survive the cold better, you might think, would survive longer, have more children, et cetera, et cetera, et cetera, but given that we mostly live in houses, it's just not really a factor anymore. You might say the people who can. I don't really know. I mean, I don't know what determines the. I mean, you know, one point of view would be depending on your sort of governmental setup, you know, it could be if having children is really Expensive financially. It's like the people who are better off financially will have more children. Let's say it's often a somewhat bimodal distribution because of. Social support, kinds of things that support, you know, that depending on exactly how the sort of breakpoints work and so on, it can be. So it's like, let's imagine that you had a situation where it's really expensive to have children and there's no other source of sort of support. If you don't make the money yourself type thing. Then you could imagine a situation where kind of a weird situation where sort of there's a natural selection in favor of traits that cause you to make more money in the world. You know, I don't think that's quite what's happening because of the fact that sort of the government and society and so on supports people in different ends of that spectrum. But that will be one weird type of selection that that could occur, I think, kinds of. I mean, another type of selection, I suppose, is if there are environmental toxins that are becoming more common. You know, when people have all kinds of theories about what sorts of things, about, you know, different kinds of food that are generated in the modern times as compared to ancient times, different kinds of, you know, bathed in electromagnetic radiation, all kinds of other things about the environment that are different. And you could certainly imagine that people who are less robust to those things would somehow have a slightly lower chance of surviving longer, having children, et cetera, et cetera, et cetera. And so there would be a tendency for people to become more robust to those kinds of things. But I think it's a little bit hard to tease those things out. And it tends to be one of these things where maybe in retrospect one will be able to say, oh, yes, this is why this happened. But it's a little hard to predict what will happen there. As far as the whole question about sort of editing the germline, the genetics which doesn't just affect you, it affects your children and so on, and your children's children and so on. At what point do you say, oh, yeah, we're going to change these genetics or allow those genetics to be changed, and in what direction? It's sort of an interesting question. And I think there's. Well, there's sort of what happens here on Earth and, you know, if we end up with a great big Mars colony, for example, what happens there? Because in that case, sort of the ethics becomes even more complicated because it could very well be the case that there are traits that you really kind of have to have if you're going to be happily living on Mars. And it could be that those are traits that can be generated by changing the genetics from the beginning, so to speak. Maybe you need, I don't know, you need to be able to survive at lower oxygen levels. You need to work well at lower gravity, Things like this. Maybe in the case of gravity, just by virtue of the way that bones grow and things like this, people who grow up in Martian gravity would end up being taller or whatever else that might happen as an epigenetic phenomenon. But one could also say maybe there are traits that one plainly knows are important if one's going to live on Mars, which in perhaps after many generations of living on Mars, might naturally evolve there. But one could imagine saying, oh, gosh, we're going to have this first generation of colonists on Mars. Oh, we should change the genetics so it's easier for them to live there. And that's a very ethically messy thing because it's kind of like, to what extent do you determine kind of the lives of your children by modifying their genetics? And I think it's very questionable that people would think that was a good idea. I mean, there are plenty of situations like this. There are plenty of ethical messes associated with this. There's another one that's kind of obvious, which is if we're going to do interstellar travel and we're going to have spacecraft that go out, and 20 generations later, they're going to have population of people living on the spacecraft, and in 20 generations, the spacecraft will arrive at an exoplanet and everybody will get off and stake a claim and lead sort of a diaspora of our civilization, so to speak, to some exoplanet. But if you imagine the dynamics of, you are a fourth generation person who was born in the spacecraft, and your great great great grandparents made the decision to that they're going to go on the spacecraft, and 20 generations in the future, they're going to, you know, you're stuck in a spacecraft in interstellar space, there's nowhere to go, so to speak. I don't think that our species and sort of human psychology will survive that well. And I would be amazed if at the end of the 20 generations, you know, despite whatever would have happened in the kind of whatever, you know, religious like development of what people think purpose of things is because it's important for us that we sort of imagine there's a purpose for life. What are we doing, why are we doing it, et cetera, et cetera. Et cetera. If you're kind of locked in place in this sort of spacecraft for your whole life, and you know that, and you know that many generations beyond you will be similarly locked in place. It's a very different sort of form of life than we typically have today. Now, I mean, one could argue that, that in many societal situations, people are born into a certain situation in society, and they're sort of locked in place in that situation in society. But I think there's at least, if one doesn't like that situation, that there's at least a hope of breaking out of it. Whereas if you're stuck in an interstellar spacecraft, you're stuck and there's sort of no way out, there's no social mobility, so to speak, you're in a trajectory going to an exoplanet, and that's the only thing that can happen in the course of your life type thing. Anyway, thoughts about that question here from Paula. What does that say about the future of philosophy? It seems like such an ancient study. Well, gosh, the arc of philosophical development is sort of interesting. What is notable about it is that there is a continued discussion of what was said 2,000 years ago. Aristotle, Plato, all of these folks. There's also a discussion of certain more recent waypoints. Kant, Hegel, Locker, whoever else, whoever your favorite philosopher is. It's a field where there's much more attention paid to the things of the past than to the things of the current, at least by many people. And that's sort of an unusual thing. I mean, I suppose it happens in theology and religion a lot. It happens a bit in literature. One could argue that it should happen in literature. You should be studying Shakespeare and so on, rather than only studying the latest book that was cool based on some bestseller list or something. It happens a bit there. But in philosophy there's quite an emphasis on sort of what's been said before. Now, part of the reason for that is because there were things that first got written down by those people. And it's still the same things we're discussing today. Questions about sort of our place in the universe, our place in society, the nature of human condition, all these kinds of things. While we have the human condition as the human condition is, we're kind of stuck discussing those things now. In the course of time, some things change in the world. The main things that are drivers of change in the current time are technology. And there are situations that are sort of the computational philosophy and things that we get to think about based on kind of the ideas of the ideas and actualities of technology are a bit different. Sort of a precursor to this, because math and kind of formal thinking was something that definitely has informed philosophy. And I think when Plato was talking about things, if he hadn't had sort of geometry as one of. And sort of the idea of deduction and necessary things that you could figure out in geometry as sort of a backdrop to some of what he was talking about, he would have had different things to talk about. And similarly, when we talk about sort of the nature of our human experience in the context of our perception of what's going on in the world, science has been important in that. I think science that we've done in recent times with our physics project and the things that go beyond that is actually a pretty major, I think, contribution to different ways to think about our relationship with the world and our realization from the physics project and so on, that the laws of physics that we perceive are not kind of just wheeled in arbitrarily from the outside, but they are necessary for observers who have the characteristics that we have. That's sort of a major, I would say, philosophical point that has emerged from a piece of science. And I think, for example, if you ask what the sort of future of philosophy will be, I would like to think, and it seems like there are a bunch of people really interested in this, that studying that view of the nature of reality, the kind of the ruliad and our perception of it and so on, that that's a definite direction that untangles quite a lot of things that have been thought about for a long time in philosophy. There are also things about sort of linguistic philosophy and how we think about thinking, so to speak, which some modern AI developments certainly help inform. When you say what's involved in thinking about things, is it important that we have language as a structuring of thought? Well, we can kind of really address that question by starting to look at what happens in sort of linguistically driven AI situations. And not linguistically driven. I've looked at this quite a bit now, and it's something where, again, we've got sort of this new tool for thinking about philosophy. And I think that another thing in terms of the development of philosophy is when it comes to political philosophy. There are things that are sort of now possibilities as a result of AI and connectivity of the world and all those kinds of things and sensors about what's actually happened in the world that are different from what there's been before. And so sort of the backdrop in which one can discuss philosophy has Changed. I mean, it tends to be the case that people will use kind of the ambient kind of way of talking about the world as sort of a backdrop for philosophy that they do. I mean, people used to say, I remember that when talking about things like necessary truths, truths that are just a consequence of the definition of terms. And don't rely on some coincidental fact about the universe. An example of something that sort of was supposed to rely on a coincidental fact about the universe is the number of planets in our solar system is 9. Or it was when I was a kid, at least before Pluto was ejected. Not ejected from the solar system, but ejected from the descriptor of being a planet, so to speak. So it is very common for people to use kind of the ambient kind of sort of ambient things that exist in the world as a backdrop, as a sort of raw material for describing philosophy. And as those things change, so there are different questions that get asked and so on in philosophy. I do think that some aspects of philosophy that have been sort of just a matter of debate for a long time are in the process of being sort of turned into things where, okay, you've had an ethical debate, but now you have to write code to decide what the system is going to do and to actually sort of make a definite decision about that thing that you had previously just written essays about. And so I think that the sort of actualization of a lot of philosophy is upon us, so to speak. And one could say that the sort of the tradition of mathematical logic and computation and so on was an actualization of some things that came out of kind of what amount to philosophical logic and the philosophy of mathematics and so on. And they were sort of actualized in computation as we have it today. And there are other things about ethics and so on and ontology and epistemology and so on that have to be actualized in the world today. I mean, I think it's sort of notable that when we think about designing Wolfram language, our computational language, when we think about designing that, a lot of the questions that come up are really quite philosophical questions about how do we characterize a thing versus variance of a thing versus whatever else. And, for example, the symbolic discourse language that we're working on, a way to sort of formalize everyday discourse about the world. That project is deeply connected to philosophy. And it's really, again, an actualization of a lot of things that people have discussed for a long time in philosophy. So I think it's sort of to attempt to summarize my immediate thoughts about sort of the future of philosophy. I think a lot of kind of actualization of things, a lot of. And now we turn it into some formalized system rather than just write essays about it. That's very much coming again. Sort of the kind of the things that you can do philosophy about are changing as a result of our different understandings of science, our different types of things that happen in the world as we built the world and as technology has set things up for us in the world. Let's see. There's a question from David. If empirical evidence indicates that there's a finite digital physicality to the universe, will the practice of philosophy and mathematics undergo huge changes? Well, it's an interesting question. I think sort of the precursor story is probably Copernicus and the question of whether you could really do natural philosophy just from your common sense to say how the world works, or whether you had to sort of trust the formalism, you know, before sort of the Copernican revolution, so to speak. It's like, well, it's common sense. The Earth is standing still. We're right here. It must be at the center of the universe, and things we see going around must be going around us. And then it was like, well, actually, the math works out. And we think a better way to describe it is the sun is sort of fixed in some sense, of course, not really fixed, because it's moving relative to the universe, the rest of the universe, but that's fixed and we are orbiting around the sun. And so the main sort of meta realization from that was just because the common sense says you're standing still, doesn't mean that's really the correct scientific formal description. And now go. There are things you can figure out by just saying, forget what we common sense think. Just use the formalism. Just trust the math. And that's been kind of a big tradition for the last 300 years. And it's something where people, I think, probably have overrun that tradition, as in they assume that there are things that you can just work out the math. You'll figure out what happens, but it often isn't true. I mean, my sort of favorite phenomenon of computational irreducibility, if you know the rules for something, you know, sort of computational rules for a thing, you can always work out what happens with the thing, but by just following those rules step by step. But there's a question of whether you can jump ahead and say, oh, I know what the answer's going to be a billion steps in the future. I just have a formula for the Answer. The claim is that there are many things where they are irreducible in the sense that computationally irreducible in the sense there is no way to kind of jump ahead and just say, I've got a formula for what's going to happen forever. Instead, you have to follow the steps to see what happens. That's sort of at variance with the traditional sort of mathematically informed view of science, which just says there's a way to work out a formula for things. And that formula allows you to predict whatever you need to predict. This is saying there's this sort of irreducible computational effort required to do that prediction that is a necessary consequence of things being fundamentally computational. Now, a little bit of a footnote to that, there are some slices of reducibility, pockets of reducibility within that overall computational irreducibility. And it could be that the things that we care about, and this is to some extent true, the things that we care about live in those pockets of reducibility. And so while there is all that irreducibility underneath, we maybe don't care as much about that as we might. But it's important that there is that irreducibility underneath. And I think that getting experimental evidence for, for example, the discreteness of space kind of really finishes that story of is the universe ultimately computational or is it somehow sort of mathematical? And when we say mathematical, we kind of assume, not really correctly, that if it is described by equations and things that, oh, well, we can just get a formula for what happens. I think that knowing that space is discrete, as matter is known to be discrete, light is known to be discrete, and so on, if we know that space is discrete, we're kind of done the universe really is discrete. And that makes it look more computational and makes it sort of emphasize these computational phenomena like computational irreducibility and kind of is a great big sort of blaring thing. You've got to take this seriously. Now, in the case of mathematics, the idea that sort of math was important for understanding the world, which you might have sort of deduced a bit from Copernicus, still took another 150 years before that, really, really 130 years, whatever, before that really got serious with people like Newton and so on and the developments of calculus and the like, so it might not be, you know, one can't expect totally rapid progress. I have to say that I think sort of the computational view of understanding natural science, which I really started pushing at the beginning of the 80s, I think finally now we're 40 something years later, that is becoming quite widely accepted. So lucky for me, so to speak, that it sort of more or less happened within a part of a lifetime, so to speak, rather than, oh, whoops, you have to wait for 150 years. But these are things where I think sort of empirical experimental evidence for discreteness of space would be another serious wake up call. For it's all computational all the way down. And once you know that things like the ruliad, things about observer theory, things about computational irreducibility principle of computational equivalence, all these kinds of things then immediately become. You can't ignore them. That is just the way things work. There's no hiding from those things by saying, oh no, but really it's math. And even having said it's math, the idea that it's math that is reducible, that has the feature, you can just get a formula for the answer that doesn't even follow it is very difficult to assess how hard is it to do things when you have real numbers and things like this? Because the way of assessing computation is mostly in there's these discrete rules, there are these discrete steps of computation. It's like, how many steps of computation did that take? If you're doing things that involve kind of continuous equations and continuous variables and so on, it's less easy to tell sort of what was involved in doing this. There isn't really as definite, oh, you can count the number of steps type thing. So that's part of the reason why when people say, oh, well, maybe it's not discrete, it's not computational, it's mathy, they sort of imagine that as soon as it's mathy, got an equation, you can just solve it. That's not true. I mean, it is much messier to understand kind of the extent to which there's sort of computational limitations in what you can do with sort of continuous things where you sort of vary the time continuously and so on. But it's still the case that there are computational limitations, they're just harder to see. And you can kind of hide a lot of things in things like, oh, we've got an infinite number of digits and when we add two numbers together, we're dealing with an infinite number of digits when we do that addition. And so we're sort of hiding a lot of computational effort. And just like the idea that, oh, we can do an addition. Good example of this kind of thing. Sort of notable in this week when there's been a nice eclipse is the three body problem. The problem of the Three bodies, like idealized stars, interacting with each other according to gravity. What are the trajectories they follow? And often these trajectories are very complicated. And one of the questions that's completely unknown. It's known that there aren't easy formulas that can be given for those trajectories. It's known that certain kinds of formulas, algebraic formulas, simply don't exist for those motions in general. But the question is, can you sort of say, how computationally difficult is it to work out that motion? Is there some fundamental sense in which you, for example, can't know what the infinite time behavior will be? Will the thing keep squiggling around? Will eventually all the stars get ejected or whatever else? And that's the thing that I've certainly wondered about for a long time. Is there a way of using some computational ideas to sort of say, oh, well, actually the three body problem really is computationally complicated, but we don't know how to do that yet. And that's been a difficult problem to. To kind of merge the assumptions of real number mathematics, where you can just say we're just going to deal with an infinite number of digits. We just put a plus sign and we add these two numbers together, and they add even though they've got all these digits that have to be added to say that we sort of are more honest about what the computational effort to add those things is going to be. That's something that is not so easy to do. But now it also depends on is the universe a place where, oh, it can just take these real numbers and just add them together? The whole point is, if we find out that space is discrete, we know the universe doesn't work that way. We know the universe is stuck doing the same kinds of things that we have to do with computers and where we're looking at each digit and things like this, it doesn't just have some way of kind of jumping ahead and saying, we take these numbers, infinite numbers of digits, and just do something with them? There's a question here from Natter, Are all philosophers logicians? Well, no, there are different schools of philosophy. I mean, the Anglo American school of philosophy has tended to be more kind of formalistic and linguistic and sort of logic oriented. Well, the different schools of philosophy, I mean, there are other ones which are much more about sort of talking things through without the structure of logic as such, not as formalized as others. I think the concept that there is sort of a way to formalize things with something like logic, logic is sort of a Formalization of certain kinds of arguments, one might say. Does ethics have a logic associated with it? Well, people have sometimes claimed that, but the sort of the calculus of ethics is its own separate thing, independent of the pure idea of logic as a way of describing what's a valid argument. Let's see. Lola, will we ever find a cure for the common cold? I think absolutely yes. You know what's at issue? Well, our immune system has all these clever mechanisms for figuring out is that thing that has attacked you something that the immune system should attack or is it, for example, a part of you? The immune system has, the adaptive immune system, for example, in mammals and so on, has this feature. It's generating, well, altogether tens, billions, trillions, even of different sort of configurations of molecules which have different shapes. And it's sort of amplifying the ones whose shapes match the things that are coming in to attack you. And once those matches start to happen, it's saying, oh, if there's a match, kill that thing that's coming in to attack you. That's how our immune system works. Now what one has to do to, to sort of kill off the attackers. For example, when antibiotics were discovered about a hundred years ago, now the less than that. No, about, about 100 years ago, the. That what was at first being done there was that bacteria have different kinds of cell walls than us, eukaryotes, mammals, humans, whatever. And so the molecules which are antibiotics have this feature that they, the early ones, penicillin and so on, had this feature that they kind of explode the cells of bacteria and leave our cells, which have a different cell wall, intact. So it's always a question of can you find a way to distinguish the bad from the good, so to speak. And our immune system has a pretty well tuned way of doing that. Sometimes the tuning is off and then you get an autoimmune disease and so on. You start attacking your own cells. But the sort of, the idea is that the immune system, as it develops in a young human or whatever, will recognize what's self and what's non self. And then when it's non self, go attack it. And that's sort of the fundamental problem of kind of making sure that nothing infects you is if something is coming in and it's not, you kill it. So get rid of it in some way. And I think sort of the way one thinks about doing that is sort of can one imagine. Well, in one sense one can imagine a drug which in the past has always meant a molecule with a certain shape, certain binding to other molecules. And for example, when it's a certain shape with certain binding, it might just glom on to some particular thing that is part of the attacker and sort of prevent that from working or destroy it or blow up the cell wall or whatever else it is. And that, that's a feature that's just a. You got a molecule of the right shape to fit into that kind of lock and make this happen. That's been sort of the history of things. Now our immune system is a bit smarter than that because our immune system has this generator of diversity of possible shapes and then the selection for which shapes are the ones that are relevant for attacking these particular attackers. We can certainly imagine algorithmic drugs which do something a bit like the immune system, where instead of just being an, oh, I'm a molecule with a fixed shape, it's some more complicated thing that has the ability to, for example, change its shape, pick a shape that will go in and attack this particular kind of thing. You've kind of just got to find. It's so kind of in the end, it's very much like sort of computer security. It's like you've got this thing, it's a biological system, it's running in a certain way. Is it doing something bad? If it is, go and sort of block it in some way and be an algorithmic drug that's kind of smart enough to block what's happening. So my assumption is that what will eventually happen is sort of this kind of notion of algorithmic drugs where one will be able to sort of say, okay, if it has these and these characteristics, then it's a bad thing and go kill it. And that that will be true for both viruses, infecting cells, tumors, lots of other kinds of things like that, where it's like a thing you don't want. Can you tell what it is and can you have a molecule sized thing go in and zap it? Or even in some cases a cell sized thing like the T cells in our immune system, which is just circulating in the bloodstream and so on. But it's so my. I will be surprised if in the end there isn't a cure for the common cold. Now, interesting question is when you lock out all those different things and you have this kind of, you've locked out everything that can be zapped by noticing this or that feature of it, then evolution operates and the bacteria, the viruses, whatever else will evolve and try, and the ones that make it through that sort of firewall will be the ones that then expand and so then the question is, well, as we. As we. As. It's sort of this informational firewall. We've got these viruses that are developing, and we've got our blocking of those viruses, and it's all kind of informational, so to speak, can we always stay ahead? My guess is that it's actually, we will be able to stay ahead. Will be able to. I mean, that's an issue that's happened in the past when people say, can we make a flu vaccine that's going to be good for the next season? Oh, we're noticing, you know, H1N5, whatever it is, you know, the strain that's just a description of a part of the genome for the influenza virus. You know, that strain seems to be becoming more important. Let's put that into the thing that we have in our vaccine that will make people's immune systems kind of wake up and generate antibodies to that particular kind of sequence and so on and sort of predict that. But that's a very, very slow process. And what one's interested in is something more like our actual immune system that can respond quickly. But then there's sort of this question about if we have the superimmune system, do we have things that evolve to evade that? Well, just like we have things like HIV which evade, which block certain aspects of the immune system. Same with some tumors, block certain aspects of the immune system. And so when we have the super immune system that we've done with technology and engineering, what will the things that evolve to block it look like? My guess is we'll always stay ahead, because my guess is that the technological evolution is incredibly much faster than biological evolution and that it will be rather easy, in a sense, to be sort of, oh, there's a software patch, effectively, to your algorithmic drug that's easy to install relative to. Let's wait for a number of generations of this bacteria that's replicating every 30 minutes, and it's doing this and it's doing that, and it's got to have all these mutations build up that. That will be a much slower process, I think. Let's see. Well, DES is asking, could what we're saying here with algorithmic drugs mess up our immune system? Because they've always fought colds. You know, the immune system is not well understood, and certain aspects of it are understood, but others are not. For example, you know, a big sort of medical mystery of modern times is the rise of autoimmune diseases among, well, initially kids in sort of more developed countries and There are all these theories that say, oh, it's because, you know, kids used to be around more dirt and they sort of used to be introduced to more antigens early in life, and now they lead too clean a life and their immune systems are not developing the right way. I don't know if that's correct. But there are many mysteries and, you know, there are many mysteries about how vaccines work and, and, you know, whether, whether, you know, when you give this vaccine and you. And you sort of drill out this part of the, of the space of possible antibodies, you kind of supply lots of. Make somebody generate lots of antibodies right there, maybe that inhibits the generation of other sort of nearby antibodies. That's something that was actually observed in clinical trials of some vaccines. That, that happens for some particular choices of what's in the vaccine. And that's bad because then when the antigen, the virus, let's say, evolves and is kind of nearby, but not right on the nose of what the vaccine was for, then it can end up being. You can end up with a situation where your immune system generates fewer antibodies in that region around the place where the vaccine was sort of just saying, generate these antibodies, generate these particular antibodies, and so you actually have less response to these slightly mutated versions of the virus. And it could be that in the natural immune system that where we get kind of antibodies, anti antibodies, whatever, not completely clear how this works, that there's sort of a natural tendency for us to get sort of more and more variant antibodies to things and that that kind of matches the natural variation of some particular antigen. So, yes, there are many things not understood about the immune system. I mean, this is the story of medicine all the time, is we look at some particular feature of some symptoms, some particular characteristic of something going wrong medically, and we say we can fix that. And then knowing, oh, that thing has some sort of tentacle that connects to something else, and, oh, we fixed that, but we broke something else. It's really hard to know. And one of the things about sort of the intuition about medical kinds of things is in medicine, in life, I suppose, in living systems in general, sort of in some sense, everything is connected to everything. It's very different intuition from something like physics where you can say there are three effects and that's it. And we've got this closed system and those are the only three effects that are important. In medicine, for example, there's always another effect. There's always some particular subpopulation of people where they have this and a, that. And it always Works a bit differently and it's kind of a. So it's a really different intuition. And I think the part of the story is often when you have medical therapies is like, well, look, this thing is clearly going to be a problem, might kill you, let's fix that thing. Even if the result of fixing that thing is some side effect in quote somewhere else that maybe is not so desirable, but it's better than it killing you type thing. I mean, I think in the end this is a computational irreducibility story as much as anything. That sort of the system is kind of complicated enough that you can't readily foresee or what's going to happen if you make some change somewhere. In physics, a lot of what we study are computationally reducible phenomena. Things where we have simple equations of motion, things like that. In biology, we're stuck with us the way we are, so to speak. We don't get to say let's make a simple frictionless wheel or something like this. We are just, we are constructed out of the proteins and everything else that we're constructed out of. And so we're just stuck with this. And what we're stuck with has lots of computational irreducibility, which makes it really hard to say we're going to do this one thing and it's going to, we're going to know all its consequences. And also it's sort of inevitable that things get connected by virtue of this phenomenon of computational irreducibility. So I think that's some, yeah, few, few comments about that. Knight asks what about the possibility of injecting tiny computers into our bloodstream, into blood cells? Well, yes, I mean cellular sized computers is surely going to happen. There's one level is molecule sized computers of actually an individual molecule that can kind of sense its environment and effectively do a computation to figure out what to do. Another is cellular sized. In the immune system, there's plenty of cellular sized computation going on. We can clearly see the T cells, the immune system. You see little things where they attach to each other and they do something together and then they break apart. And that's undoubtedly we can think of that as a computational type process. I'm sure with the right exact setup, you could be computing prime numbers with an assembly of T cells or something. Now I don't think anybody's ever done that, but it's something I think one could imagine being possible. So yes, there are natural immune system cells, for example, that operate in a sort of computational way. And could we have a. And clearly the immune system as it surveils sort of what's in the blood and so on, and cells in general. It's kind of saying, does this match that? It's kind of a very computational type of operation. Now the question is, if you are making sort of cellular sized computers, are those cellular sized computers made of biological, like stuff like proteins, sequences of amino acids and so on, or are they made of things that are more traditionally sort of things used for, let's say, electronic technology? That's not clear. I think that the. It's, you know, are you going to miniaturize some, some sort of arrangement of carbon atoms that really isn't anything like a protein? Or are you going to have something which has the actual ordinary amino acids in it and you just arranging them in such a way as to make a protein that has the right shape and so on? I think there's sort of arguments in both directions. My guess is that in the end the molecules from scratch will win. But there may be an intermediate stage where it's. Because the best construction kits we have for molecules are the ones that biological life has provided for us of using proteins. And once you have that sequence of amino acids to make a protein, the protein folds itself up in some complicated way. You might say, well, all it has is its shape. But that's not quite true because, for example, one of the things a protein can do is to have a little cage inside it, a cage which fits just one kind of atom. So, for example, in our blood, hemoglobin is a protein that has a little cage piece in it that fits just an iron atom. And the iron atom binds oxygen and so on and does all the good things that hemoglobin does. But the point is that what's happening is you're just making those hemoglobin proteins and there's enough kind of random iron atoms hanging around that you'll always get one in that place in the cage. And it's sort of a way of dealing with. Even though iron is not something that appears directly in proteins, it has a way of dealing with that. And one can imagine similarly something where one sort of has a framework that's built with proteins, but the real action is in something not associated with proteins. Or one could imagine that the thing is built from scratch as just an arrangement of carbon atoms or something like that with its own little graphene wires or whatever else is going on. But I think in the end, the thing that, yes, it's definitely a thing to have tiny computers in one's bloodstream. I have to say I was in an event a couple of weeks ago now where a bunch of new technology things being shown and one of the dramatic moments was a company that makes kind of a bot the size of a large swallowable pillar. And the CEO of that company sort of swallowed one of these things and somebody else kind of drove it around with a little game controller actually, that looking at that chap's. The insides of that chap's stomach by being able to have this little submarine effectively that is the size of a pill and is navigatable and has a camera, is communicating by Bluetooth probably, and has little jets at the back that allow you to sort of move it around. Now that's a lot bigger than a blood cell, but it's a different. It's. Well, for the, for the digestive system, that's certainly a thing one can imagine doing there. And this particular bot was just a way of looking at things. One can imagine a more sophisticated version that has little pincers that can go and, and sort of chop off a polyp or something like this. Not here yet, but it does seem like that's a thing that will be coming. Let's see. Elsie says one might remember a boom in nanotech about 20 years ago or so ago, including many nanorobotics research labs and the subsequent bust of sorts of where is that nanotech boom and bust cycle now? And looking ahead, yeah, it was a sad story because the whole idea of nanotechnology making things the size of molecules, doing molecular scale manufacturing originally a lot of what was talked about there was take the things we know, work like machines that work with levers and cogs and things like this and just make them very, very small, make the molecular scale and then sort of operate them as molecular type machines that might be like a molecular factory even. That was kind of an early idea that was hard to do and it didn't really get done. And as it sort of was connected to people trying to make things typically out of carbon atoms that were very specific shapes. The sort of, the big thing that came out of that was buckyballs, which actually existed already in soot and things like this, but nobody recognized them. These kind of soccer ball like shapes, arrangements of carbon atoms and so on. And I think what happened was, well, it was kind of a complicated sociopolitical dynamic. But what happened is, particularly in the US it became nanotech. It's going to be critical. It's, you know, let's put a bunch of government money into Nanotech as a way to sort of stimulate this sort of us, particularly in the U.S. kind of sort of achievement in science and technology. Unfortunately, I think what really happened is that injection of resources had the wrong effect because there were incumbent fields like chemistry and material science that saw that infusion of resources and said, wait a minute, that's close to what we're doing. We have a whole infrastructure for absorbing resources. Let's go grab those resources. When the resources were put in, thinking they were oriented towards, let's make nanotech happen. The very fact that there were large resources caused the sort of the sleeping giants, I'm not sure how sleeping they were, to wake up and say, wait a minute, we're doing something a bit like that. We can call it nanotech too. Let's go grab those resources. And the result was that the place, the sort of innovative direction that was potentially the thing that would be going off in this new direction of nanotechnology, molecular manufacturing and so on, was completely starved of resources and basically nothing really happened there. Now, you know, and people sort of said, buckyballs are like nanotechnology, let's declare victory when we put them on nonstick frying pans and things like this. And it all kind of dissipated. I think that, you know, when will it be back and in what form? I think there's not a lot of activity yet in that area. Metamaterials, because being able to kind of make molecular scale structures in materials that aren't just standard crystal structures and things like this, that's a thing that is becoming more notable and starting to see sort of features of that in things like optics, electromagnetic kinds of things, other kinds of areas. So that's a sort of a thing a little bit like that molecular scale kind of technology idea. I think the sort of the big thing for nanotechnology, the way I see it, is we've done very well at getting electronics and miniaturizing electronics. And we have this mechanism of lithography where we're taking. Basically we're just sort of making a sort of reduced sized picture and where the ultraviolet light falls, we etch out that channel in the silicon. And the result is we kind of have this engraving of the circuit. That is how we make microprocessors with billions, hundreds of billions now of transistors in them. And that's. But that's all set up to move electrons around, to move electric currents around, and to do everything we do with computers with electric currents. Other areas of nanotechnology don't just want electrons Moved around. They want atoms, material atoms moved around. If you're trying to do something where you are sort of, particularly when you're interfacing to something biomedical that involves a bunch of atoms, you need atoms. It's no good to just have the electrons. I mean, maybe there are things where you can just sort of zap that cell electrically in a biological system and it will do interesting things. I think there's some evidence that that's an effective thing to do. But the sort of, the many. Most things in biology involve actual atoms, molecules and so on. And so I think one of the interesting directions is we have this mechanism for transporting electrons around in silicon and semiconductors, microprocessors, et cetera. We don't have a comparable mechanism for kind of having little tiny channels to transport different kinds of atoms around. I mean, it's always notable when you see things, I mean, that they're sort of molecular scale construction that to some extent happens. Like if you're synthesizing a DNA molecule with a certain sequence in it, it's always very charming to see these machines that are kind of like 3D, like color printers, where there's literally a thing for the, you know, in DNA there are these bases, you know, A, G, C, T, and literally the machine has a bottle of A's, a bottle of C's, et cetera. And you have to fill it up when you've. I'm sure if you were making. I think crabs have lots of as and Ts and very few Cs and GS in their genomes. So if you were synthesizing lots of crab genes, your machine might run out of A's and T's and you might have to refill that before you refill the other things. But the fact that those are sort of. You've got these kind of the source of these different, different bases that you're sort of flowing through in that case, a macroscopic technological device to go and sort of add on to the DNA. But you can imagine sort of a much more microscopic version of that where really instead of having an electric circuit, you have an atomic circuit, so to speak, where you're moving atoms around. I mean, the general picture is there's something you're doing things with to make some process, computational process happen. We've done it with electrons for the last hundred and something years. One can imagine doing with photons, with light. There's a whole field of photonics that involves trying to get kind of systems that do logic based on photons rather than based on electrons. And then the other version is doing it with atoms, doing it with pieces of molecules, actual molecules and so on, having those be the things that are sort of the working material for computation. I mean, it has to be said, in the case of photons and light, the real thing that hasn't isn't yet known how to do. There's just one thing. If that got solved, it would have sort of a dramatic effect. And all kinds of interesting things will be done just with light, not with electrons. And the thing is, it's hard to make a switch. So in the case of electrons in a microprocessor, there are so called field effect transistors, where essentially what happens is that if you have a voltage between, you have a kind of a wire that's going from one place to another, and sort of on the sides of that wire are things that come in from other wires. And when there's a voltage sort of on the sides of the wire, that wire doesn't, it depends on how you set it up, does, doesn't, whatever, let's say, doesn't allow the current to pass through it. And when there isn't a voltage, it does. And so that's how you make a switch. That's how you make something where if one thing happens, then another thing can happen. And if the first thing didn't happen, the second thing couldn't happen. It's also how you make an amplifier because you can say, oh, I'm putting a little, I'm having a little tiny effect from that tiny voltage. That's having a big effect on whether a current can flow here. But that's a thing that was figured out how to do with semiconductors and electrons and channels for electrons and semiconductors. No such similar thing is known for light. One reason I suppose that it's immediately difficult is when you have a stream of photons, a beam of light, for example, going one way and a beam of light going that crosses that beam of light, there's no sort of direct interaction between photons. There can be interference effects. That's a different thing. But photons don't scatter photons. Well, they do a tiny, tiny bit. There's a tiny thing in particle physics called light by light scattering, which is a very small effect, irrelevant for technology, relevant only for very high energy photons, gamma rays, things like this. But, but if you say, well, how do we make something where one stream of photons can affect another stream of photons? The answer is we have to have it in some material or Another, and we need it in some material where, for example, we have a very nonlinear material where. And it is the case that there are saturable absorbers of light. So there are materials where if you pump them with lots and lots of light, they will no longer absorb other photons. So if they. Without being pumped with lots of light, they will be transparent. Sorry, they will be opaque. They don't let photons through. But if you pump in enough photons, then you can make them transparent. The problem is that the only cases where that's known are cases that involve immense intensities of light and so on. And they're not the kinds of things you would want to make. Sort of a photonic computer, an optical computer, where there's sort of individual photons that are switching on and off different kinds of things. But sort of the issue is, can you find a material or an environment or whatever where you do get that kind of switching action for photons? If you can do that, then photons become a very useful working medium for computation. But in the case of nanotechnology, I think that the story has to be sort of atoms as the working medium for stuff going on. And the interesting thing is that in the case of electrons, we, if we want to interface them to us, well, yes, we can have something which directly stimulates nerves, for example, or directly stimulates neuromuscular junctions and stimulates muscles or something like that. But we're not. The electricity on its own has to be interpreted. We have to make a picture on a screen and we look at it with our eyes and things like this. If we wanted to make something which is directly interfaced to our biological nature, then being able to have something where the output is atoms or molecules would be really useful. Now, I mean, having said that, to some extent, just the output being electricity is relevant because some aspects of what we do as biological organisms is electrical, for example, in our brains, stimulation of our muscles, things like this. But if we could also do things where it is molecules going on, then that covers the awful lot of other subsystems that we have. You know, the operation of, I don't know, kidneys or whatever else you could imagine making something which is sort of doing the chemical things that. That does and so on, but with a piece of nanotechnology rather than a biological, rather than the kind of thing that's happening in biology. Let's see. Yeah. Lucas is commenting. For nanotech to really take off, we need new foundational building blocks from sort of convergence of biotechnology and electronics. I think one of the Issues is what are we going to piggyback on biology or are we going to just build from scratch? You know, are we going to make something which is like a computing with DNA, let's say, or are we going to make something that's molecular scale computing with molecules that were just created for the sake of making molecular scale computers? We know a lot about how DNA works. There's a lot of infrastructure for dealing with DNA that the last 3 billion years of biological evolution has provided for us. You know, whether it's ribosomes or whatever else for endoplasmic reticulum, whatever kinds of things there are for dealing with sort of handling proteins and such like. So we have a 3 billion year head start if we're using biology. But also biology has many other constraints that have to be kind of, they have to be realized. I mean, I was looking at some device that some people have where you're, where you're growing nerve cells on a. Essentially a silicon microprocessor type substrate. You've got this bundle of nerve cells. It's kind of a, kind of a weird sort of future ethics type issue because they're human nerve cells. You can grow there. You got a million human nerve cells and they're interfacing with this piece of. And the question is, can you talk to those nerve cells? Do you have a mechanism for kind of communicating? And it's not so great right now, but you can start to say, can you like do machine learning with actual nerve cells? But the thing that's notable in the actual device, at least I saw for doing this, it's like there's a little nerve cell compartment that has this stuff and then there's a big other piece that has a sort of a replication of many aspects of the human body. Because those nerve cells, they live at, you know, whatever it is, 98 point, you know, 99 degrees, whatever Fahrenheit, 37 degrees centigrade or whatever that is, what, you know, biological tissue likes, they have certain nutrients flowing through them. So you have to have in this device, it's like little nerve cell compartment and then big sort of body imitation to provide the sort of the wherewithal to get to keep those nerve cells alive. It really is a. And so, you know, whereas if you had other kinds of molecules that were just built not by biological evolution, not for the sake of being sort of biologically supported, but were just created for the sake of whatever purpose you have, you can imagine you don't need all of that support structure. It just becomes the Pure thing itself, the pure molecular computer. You know, I have to say it's really a strange kind of ethical conundrum as we have these organoids, these kind of clusters of things like nerve cells and where they're sort of potentially communicating with the outside world through electrical signals. And how should one think about this? I mean, a million nerve cells, it's not very big. It's bigger than an ant. Much bigger than an ant is 50,000 nerve cells. But it's a lot smaller than a human with 100 billion nerve cells. But, you know, imagine you scale that up, you've got tens of billions of nerve cells and you've got this whole clump of nerve cells. Maybe it doesn't have quite the architectural organization of the brain, but maybe that isn't all that important. Maybe just a big cluster of nerve cells is enough to do thinking. And it's. Then what on earth is going on at this point where you have, for example, human nerve cells? One could even imagine taking one's own, cloning one's own cells and having kind of the, the organoid of one's own kind of brain, like mini brain type thing. Maybe that would even be useful if you want to replace nerve cells in your own brain. You could imagine producing from stem cells, for example, nerve cells that would be sort of genetically matched with you, then injecting them into your brain and hope that they sort of go to the right place and become useful, which probably they would, because that's a typical thing that happens with cells in biological systems that they sort of migrate to the place where they can actually be used, but in any case. And they kind of knit themselves into the environment. But in any case, imagine you've taken clones of your nerve cells. You grow this organoid. It's a 10 billion cell organoid. It's sitting there. It really is a brain in a jar. And it's communicating in a sense by these electrical connections to the outside world. But gosh, what a strange ethical situation, particularly if you imagine it being, let's say, your own nerve cells and you're kind of. And then you say, well, gosh, I can see it's developing some eeg, it's developing some brain waves. It's like, what the heck is it thinking in there? And is that organoid like saying, let me out, you know, or whatever, or, you know, how do you tell what's going on? How should you think about what's going on in this sort of exotic, in this increasingly kind of disembodied brain? I Don't know. I think it's a very tricky sort of story because, you know, a lot of our ethical principles are based on the idea that there's people and they're all people at some level. And you know, at different times in history, people have said, oh, different, some people are different. But, you know, know, we have this notion that we're all people at some level and we have the idea that whatever we feel in our own minds and our own brains is somewhat similar to what other people are feeling. And so we can have empathy for them. We can sort of base our ethics on that idea. When it comes to sort of extending to how do we feel about animals, non human animals, then we have less empathy for those things because they don't seem like they're operating the same way that we're operating. But now, by the time you have the organoid that's made from human nerve cells and so on, what is our sort of ethical stance with respect to that? And to what extent do we say, well, doesn't have the body, the human kind of wrapping, so it's different, or do we imagine that the thoughts going on in that organoid are just like our thoughts and it's like, oh my gosh, what are we doing? Because we feel empathy for this. And if we said we've got a human brain, that we are human, where we've kind of sort of locked them up and put them in this thing where they cannot actuate in the world other than through this sort of, through this kind of connection to electronics, we would probably think that was really a shocking thing to do. But so how do we think about those kinds of things? Sort of interesting ethical issues. Let's see. Well, okay, couple more things here. Elsie is asking, what do you foresee in terms of substrates for the future of computation? Medium term, long term, you know, it's all going to get more molecular. I think electrons are still really convenient and. But those electrons. Right now we are, you know, we've got a picture of our. We've got something where we are literally carving out of silicon the etching out of silicon, the arrangement of sort of wires and the giant roadmap of the microprocessor. My guess is we're very proud of the fact that we got to the point where we've got 5 nanometer feature sizes for the wires and so on. But eventually that will be down to, well, not that much smaller actually, but factors of 10 or something smaller by the time we're down at, at individual atomic sizes. So you Know that will happen. Another thing that's been one of these coming technologies ever since I was involved in technology. So, you know, nearly half a century is 3D micro 3D chips. You know, at the present time, when you make a microprocessor, it's done in layers and there are a limited number of, I don't know how many there are now, a few tens of layers maybe, and the sort of wires are in one layer. But you can imagine having a fully volumetric kind of thing where you've got sort of this lump of silicon that's had things etched out of it and put in everywhere in sort of whole three dimensional structure. People have been talking about this as the way of the future for, oh, probably nearly 50 years now. It's just been technically difficult. It hasn't happened yet. Maybe that will happen. And there's a lot more room in 3D than there is in 2D. It's a lot easier to route wires in 3D. But still people have been able to do it in 2D with different layers and so on. I think my guess, as I say, is that non biological molecular computing will eventually win out over biological molecular computing. Although when you interface to biology, when you do biomedical applications and you're having to interface to us and our biology, then in the end there's got to be an output layer that is essentially proteins and things like this, because it's interfacing to our sort of protein type structure. I think that's as far as I can see, that's kind of the inventory. I mean, I don't think we're not, you know, exotic elementary particles. Not a thing in terms of things that are going to be useful to us sort of at our length scale and energy scale and things like this, I think that sort of made to order materials, which is not so dissimilar to the sort of 3D microprocessor story, but things where you're making something where the atoms are placed exactly where you want them to place. So in a crystal you have these regular arrays of atoms. In a glass you have something which is kind of like a solidified liquid where you've got just a whole arrangement of atoms just jammed in there. So instead of the fruit, the spherical fruit in the grocery store, packed in this nice packing, it's all just thrown into a ball pit type thing. That's the kind of thing that happens in a glass. It happens in a liquid too. Although in liquid things are moving around. With glass they're sort of frozen in place. And one can sort of imagine a made to order glass that has the molecules, has every atom placed in a particular way. In fact, something I've studied quite a bit is the possibility of making essentially random crystals where there is a definite algorithm for every piece about where it goes. But the only way to fit the jigsaw puzzle pieces together is something that looks kind of random. So that's another thing that if that's possible, which I'm sure it is, in fact, I know it is, but I just have. The examples I have are very complicated and not readily implementable with actual molecules. But by the time that's done with actual molecules, that gives one a very different bizarre kind of metamaterial that has. Well, I don't really know all the properties it has, but I suspect it will have some interesting and potentially useful properties. Maybe one or two more here, let's see. Mrudo asks, does the success of one field sometimes slow down other research fields? Absolutely. You know, I was going to say in terms of nanotechnology and the sort of the burst of interest in nanotechnology, there are at different times bursts of interest in different kinds of things. I think one of the big ones in modern times is quantum computing, which I have to say I'm something of a cynic about. I mean, I think that I first worked on quantum computing, I think in 1981, so a really long time ago. And there were sort of questions about what would you really be able to do there. And remembering that sort of the key idea of quantum mechanics is that whereas in classical mechanics sort of definite things happen, sort of know a ball moves in a definite trajectory, in quantum mechanics there are many paths that are followed and we just get to sort of experience the probabilistic aggregates of those different paths. And sort of the key idea of a quantum computer is sort of inside the quantumness of it all, there are many paths of computation that are being followed. And that's something you absolutely see in our models of physics that absolutely happens. The, the problem is, yes, there are many paths of computation happening, but our brains, our minds perceive definite things happening in the world. So somehow we have to knit together all those different paths of history to come out with a definite conclusion about what happened. And in the normal formalism of quantum mechanics, it's kind of a cheat because one just says, oh. And then that just happens. All those threads get sort of knitted together and we get definite results. There's no description of the mechanism by which that, that operates and what people have found experimentally Is that there are all sorts of things that kind of limit their decoherence effects, noise effects, things like this, which are kind of pushing back on the idea. Oh, it isn't actually an instantaneous, sort of optimally efficient process. And I think the thing to realize is that if our brains were sort of quantum all the way down, we wouldn't have a definite thread of experience. We would always be experiencing all these different histories and we would say, oh yeah, you know, everything's quantum and we can operate in this quantum way. What's the answer? What did we actually do? Oh, we did all these things. We did the superposition of different quantum paths, so to speak. And I think that that is. But the fact, the limitation is that we are classical. Our perception of the world is something where there's a single thread of experience. And sort of corralling all those quantum threads into the single thread of experience is not easy to do. And I think that the sort of the challenge there is can you do it in practice? And I think the answer is no. Is it worth studying quantum computing? Well, I think the answer is yes, because I think a lot of what's going on is, oh, how do you make a computational system out of ion traps or out of superconducting junctions or out of some other kind of atomic scale system or something, or something based on photons. That's a very worthwhile thing to do to explore other substrates for computation. Meanwhile, the formalism of thinking about how these threads of history go in parallel, that's very interesting too. Very related through our physics project to these multi way graphs that we study. Very related to even things like game graphs of all the different paths of, of playing of a particular game. Things like this. There's a huge sort of thing of theoretical interest in studying multi way graphs which are equivalent to quantum processes. And that's something worthwhile in its own right. But in the middle there's kind of like, oh, and will we have magic quantum computers which, for example, can break the integer factoring that's involved in lots of encryption that's used in the world. Will there suddenly be some moment where somebody announces, we made a quantum computer and we can take all these things that take you exponentially long to do on a classical computer for working out, for trying to break these codes and boom, we just broke them all. I don't think that's going to happen. I think that's not, that's kind of like. And you're also going to make warp drive and go and you know, fly in faster than light spaceships. I don't think. I don't think it will happen. Now, what's happened in the world is that people say, well, look, there's quantum mechanics. There's a theoretical way of saying you can do this. Factoring that theoretical way of saying you can do that doesn't take account of the mechanism of measuring the result. If you do take account of the mechanism of measuring the result, so far as we can tell, and we can start to see some things about this in our models, there is no advantage. And when you see experimentally people trying to do things, they keep on getting kind of ensnared in these various forms of sort of noise and failure of the system and so on. I kind of think the analogy is this. It's like, let's make a better telescope. Okay? We do this. We make it out of copper. We make it out of this thing. We have quartz lenses, we have glass lenses. We have this, we have that, the other. And we keep on seeing. Oh, my gosh, whatever we do with this telescope, we never manage to get perfect resolution. There's always some fuzziness. What's going on? Well, in the case of telescopes, we know what's going on is diffraction in light. The fact that light is not just a geometrical Just going in geometrical rays, it's diffracting as a result of wave phenomena. But what we're seeing in quantum mechanics or quantum computing, I think, is the experiments that people have done, I think are kind of giving us a big. They're waving a big flag saying there's a new physics thing that's sort of not diffraction, but something like that that is associated with large number of degrees of freedom, quantum systems. I think it's something might be something that we can kind of untangle in our physics project from our new understanding, quantum mechanics. But my guess is that's what's going on there. And so it's an interesting thing that, you know, billions, tens of billions of dollars going into kind of quantum computing. And I think mostly out of fear that, oh, somebody's going to, you know, if we don't make a quantum computer first, if it's possible to make a quantum computer, somebody's going to make one and they're going to break all the encryption in the world. It's a strange dynamic. I think that there's sort of a, you know, whenever there's a big burst of funding in one area, it starves nearby areas. I mean, and what happens often in the sciences I've been sort of out of the world of academic science for a long time now. But, you know, people would have this, oh, are you going to build a bigger particle accelerator? No. If you build that, it will starve money from other parts of physics. You know, don't build the big particle accelerator. Because it is true that by the time you put tens of billions of dollars into building the particle accelerator, people are going to say, let's concentrate on the particle accelerator. Let's not do these other kinds of things which are sort of other directions in science. So, yes, absolutely. It's a, you know, insofar as there's a fixed amount of money allocated for science, it's like the thing which is the biggest fad wins and sometimes slowing down progress in other areas that would have been able to flourish had it not been the case that sort of all the resources were sucked out for the latest sort of big fad. All right, well, I think I should stop there, but thanks for asking a bunch of interesting questions. And I see there are. There are yet more that I didn't cover here, so I look forward to doing that another time. Thanks for joining me now. Bye for now.