(gentle electronic music) - Hi, and welcome to "Conversations at the Perimeter." Today we're excited to share with you this conversation with Ray Laflamme. Ray is a researcher at the Institute for Quantum Computing and the Perimeter Institute, and he's an expert on everything related to quantum information. - I was so excited to have this conversation with Ray. I've been looking forward to it for a long time 'cause I've known Ray for about 12 years 'cause he hired me.
He was my boss at the Institute for Quantum Computing when I worked in communications, and he's really responsible for teaching me the things I first learned about quantum computing by showing me around and taking me to the labs. And his passion and his dedication to the science is really infectious. And as you'll hear, he's a wonderful storyteller.
And he told us stories about his early days studying under Stephen Hawking at Cambridge and then working at Los Alamos National Laboratory and then more recently, his struggle with lung cancer and how that's shaped his perspectives on life in the future. - I really loved hearing his stories about how quantum computing and quantum technology has evolved over the years and also what we can expect for the future. So let's step inside the perimeter with Ray Laflamme.
- Ray, Laflamme, thank you so much for being here. It's great to see you. - You're welcome. - I wanna start with a big question. How's life? - Life is really good. Definitely after a pandemic of two years, life seems to pick up again of seeing people.
And then I think that pandemic has made us enjoy the kind of precious moment even more than we did before, realizing that there are things that goes in different ways as life goes, and then you adapt to them, and then suddenly you realize kind of what are the diamonds that kind of before, you were kind of neglecting. - This is part of the reason I was so excited to talk to you.
I've known you now, I did the math, I think it's been about 12 years since we first met because you hired me to work at the Institute for Quantum Computing in Waterloo, where you were the director. And I still remember the first thing you said to me on the first day of the job. I bet you don't remember this. I remember it clearly. I walked into your office. You said, "Hello," and you said, "Lose the tie," because I was wearing a tie. And I thought, "Oh, okay, I like this guy."
And then you were such a mentor to me. I had come out of a journalism career, and I knew practically nothing about quantum computing and quantum information. So you showed me around this Institute for Quantum Computing. Could you tell our listeners what the Institute for Quantum Computing is, how you got involved, and what it's for? - The Institute for Quantum Computing is an Institute at the University of Waterloo whose aim is to develop the science of quantum information.
And that includes quantum computing, quantum communication, quantum metrology, or quantum sensors, and some materials that are essential to build devices that use quantum mechanics. So it's an institute whose first goal is to do the research, the basic research related to quantum information, the second one to train a generation of students and scientists and engineers who think quantum. We all kind of grow up, and you are learning Newton's mechanics or classical mechanics.
Even if most people don't call it that way, that's the way we understand how to control our car, using our bicycle or whatever, flying in an airplane. But really, the world at its very fundamental part behaves differently, like atoms and molecules and electrons and protons has a different set of rules. And then we want to use these rules to manipulate information.
- Could you give us maybe an example of one or two of those uniquely quantum rules that you're trying to exploit and why harness these properties? - I'm happy you say why harness these instead of why do they behave that way because we just don't know why they behave that way. The world is built in some way, and maybe there's a fundamental reason that one day we will discover that it cannot be otherwise. It had to be this way.
But right now, we just kind of explore the world and try to understand how it works and not necessarily why it works that way. So one of these properties of quantum mechanics is called the superposition principle. In the physics of Newton or what I call classical physics, we think of objects made of particles, and particles are little things that are at a given position in time and at a given velocity.
This works very well to describe most of the world that we interact with in day to day, except people like me and my students and my colleagues who suddenly go in labs and isolate these particles very well and try to see how they work. And what we find is there's something called a superposition principle. These fundamental particle of nature can be at more than one place at a given time. So a single system can be here and there at the same time.
We are trying to use this property, the superposition principle, and use it to compute. And so we kind of discovered that if we use these properties, we can build computers, or we are attempting to build these computers. We're still kind of early in that stage. Although we have some really good prototypes that shows that the science is okay and kind of things are working the way that we're thinking they should work.
And so we found that by using the rules of quantum mechanics, the theory that described this kind of very small part of the world, if we use the rules of quantum mechanics to compute, we can solve problems which seems to be intractable with classical computers.
And suddenly it tells us that if we can do that, there's a wild world of information that will open to us that we haven't, which is totally surprising because it is mind-boggling what we've seen in the last 50 years with the computing or information revolution. Before people were having to jump on a horse to tell the story to somebody else in the 16th, 17th, 18th century. And suddenly people invented the telegraph where we can send things, short waves to send messages.
This turned into computers in the 1970s and '80s and to cellular phone that we have today. When we look at our kids, if you want to give them a hard time, you take their cell phone off. It's like the end of the world, like they cannot connect. You and me have grown up where we had neighbors, and our friends were neighbors. Our kids grow up where their friends can be in France or in Japan or in South America.
Instead of having a little local village, the earth is a global village, all putting this together. That changed the way that people think, how we behave, and what we think about the future, so this incredible change of how we conceive the world because of this information revolution. And suddenly quantum mechanics comes in and tells us things can be very, very different. We can evaluate even so much more information that these classical computers cannot do. So, there it is.
And this is what the Institute for Quantum Computing, I would say the Perimeter also, are investigating these pieces and trying to put all of this together. - So as you've alluded to, this field and the related technology, it's really very quickly growing and changing. So I would assume that would mean the goals of a place like the Institute for Quantum Computing would also have to evolve. Can you tell us a little bit about how those goals have changed throughout the time that you've been there?
- Yes, I'm kind of thinking of this and kind of putting myself back 20 years when I first came to Waterloo. At the time, the goal was mostly to convince people around that this idea of quantum computing was not totally crazy, and I say not totally crazy because we don't have them yet. And me as a scientist, as a scientist, you shouldn't kind of believe something until you see all the goods.
We don't have a full-fledged quantum computer today, and until we have one, you should have a little skepticism. Although I really believe we will have one, but this is a belief and not the scientific data, and quite a distinction between the two. In the 2000s, a lot of the work of the director of the institute was to convince people that this was really an important field.
And we seem to have done a very good job because people now are, there's investment from government, industry, many universities across Canada and around the world have group of quantum information. Now this is different.
Now a lot of the work is really to develop these ideas to in part better understand where the power of quantum computing comes in, find new class of algorithms that kind of quantum computing could help and kind of make more efficient and then turn into how do we rebuild these devices and building them. So maybe 20 years ago, we were really asking how can we really build these things? Now we have a bunch of blueprints, and people are in lab trying to show them, and industry.
Building quantum computers has become complex enough that it is hard to make this in a university, in part because it takes a long time to go from the first steps to the device that we have today. A generation of grad students are three, four, five years, and that's all too short to kind of keep things going.
So we can make proof of principle of certain mechanism or certain things, but it is really the IBM, Googles, and Xanadu and those that are really kind of putting all the engineering together and kind of developing these ideas to get devices. And indeed, they are producing devices, not the one what we finally want, but enough to give us confidence that we are on the right track. - Takes a lot of pieces and a lot of collaboration, I guess.
- Yes, a lot of pieces, a lot of collaboration, a lot of stamina, lot also of understanding where the problems and the challenges are and get over them and kind of moving forward. - You said that your primary job as director of the institute for the first 10, 15 years or so was convincing people that this wasn't a crazy idea. Was there a time when you needed to be convinced that it wasn't a crazy idea? Were you a skeptic about quantum computing before you were a preacher about it?
- Yes, my first piece of work on quantum computing was to try to prove that they would never work, and I failed. - You failed and succeeded, I'd say, in an equal measure. - So after my PhD, I went to Vancouver, and I worked with a physicist called Bill Unruh. And Bill is an incredibly good physicist, and he has this tendency of, he really likes to argue.
And as a post-doctoral fellow, it turns out that sometimes it was very hard to work with him because every time I had a new idea, I would tell him black, and as a person who really likes to argue and sharpen your ideas, he would say white. We'd would argue for black and then black, and then he would kind of poke holes at my arguments, which, after a while, it gets really tiring, every time you have a new idea that you kind of get poked.
It is really good scientifically to do this, but as a human being trying to do research and trying to make your name with kind of being poked. But I learned that it was important. Probably 10 years after, I started to work on quantum computing. I went, in fact, to a conference in San Fe. My mentor at Los Alamos National Lab, Wojciech Zurek, told me there was this conference on the physics of information.
And I initially said, "I don't want to go there because I don't know anything about the physics of information." And Wojciech told me there's really neat people going there. Like, what is it? It'll take two days. And it is kind of 45 minutes away from Los Alamos. Says, "Just go," and so I said, "Okay, I'll go." And it turns out that was the first time I heard about this algorithm called the Shor's algorithm, which is crucial for quantum computing.
It's related to factoring numbers which are product of primes using a quantum computer, which turns out to be an algorithm on which cryptography, in fact, most of today's world cryptography is based on the difficulty of factoring numbers which are products of primes. - Cryptography being the stuff that keeps our information safe?
- Absolutely. When you use your cell or your computer to log into your bank, the cryptography that is set up so that it is private is based, like breaking the cryptography is equivalent of finding the factors of a number which is the product of prime. So I went there, and so this computer scientist, Umesh Vazirani, explained this algorithm, and in fact, he started with a very funny story. Umesh is a really smart guy. He always has great ideas, all this.
And he started this talk by saying, "I haven't done anything interesting in the last two years." What? And usually scientists are very, all of them are not humble. - That was put politely. - So that was a little surprising to hear. And he said, "But I've heard about this algorithm," which was going to be called Shor's algorithm. And he says, "From this guy called Peter Shor to factor numbers which are product of prime."
And there was a buzz in the conference that this was really important, and people were talking about it. At the time, I didn't know, I knew very little about cryptography. So it was very hard for me to really assess everything. But there was really a coherence in that conference. For those who know physics, it was like Bose condensation of human beings' thoughts of kind of suddenly, wow, something's happening here. I came back to the lab, and I started the thing.
I was working on something which was called quantum decoherence. And I said, "Oh, this quantum decoherence is gonna be an obstacle to quantum computers." And I started to use little kind of simple models to show that if there would be quantum decoherence, quantum computers would not work. And so I kind of put things together. Not everything was tight and clean, but I was pushing the idea that quantum computers would never work.
And one day, there is this thing called the archive where we get pre-prints for everybody around the world. I look at the archive, and there's a paper by Bill Unruh on quantum decoherence and quantum computers giving exactly my method. So I was pretty miffed. (all laugh) And then I said, "Oh, it's pretty much the idea that I had." So, okay, my last couple of months of work, it kind of goes in the garbage.
But then I said, "Oh, Bill always asked me to argue white when somebody says black, and the other way around." So I started to work to demolish his paper, and I tried to poke holes at it. - Which was essentially poking holes in your own ideas as well, right? - Absolutely. - Well, as a scientist, - Idea? - You have to look at both sides. You don't know where the truth is.
We have ideas, and you never know if these ideas are right or wrong until you go through the whole details of the mathematical models and all of this. So I was poking the other way around to try to kind of demolish his idea. Then I didn't have to say to other people that I had the same idea. I can say, oh, this guy was wrong. So by doing this, I stumbled into quantum error correction, which shows that not all errors will be kind of deadly for quantum computers.
There's family of models of errors that if they happen, there's ways to take care of them. At that time, many physicists thought that this was impossible. - Because decoherence causes too many errors and makes your computation worthless? - Yes, and quantum mechanics has this property, it's called unitary. That is, if you make a computation forward, if it is quantum mechanical, it should go backward also. If noise comes in naively, it seems that you cannot go backward again.
So they would say it's just not going to be possible to do this. The idea at first level seemed to be okay, but if you start to think about it very carefully, it is not really correct. And this is what quantum error correction is really about, is to find a way to be able to go forward and backward in your quantum computation, even if noise comes in. - So you essentially demonstrated the opposite of what you thought, that quantum computing is possible.
- I think we should have to, I should be a little bit more precise. It didn't show that quantum computation was possible because we don't have them yet. So we still don't know. It shows that noise and quantum decoherence are not a fundamental objection to get quantum computers. - And we also have to think about error correction in classical computers, right? So can you tell us a little bit about the difference, really, the fundamental differences between classical and quantum error correction?
- Now that becomes a little bit more, could become a little bit more technical. So I'll try not to be too technical. The idea is first related to the type of noise that we have. In classical computers, all the information is encoded in bits of information. Bits in information is the smallest unit of information that we have typically encoded in a system with two levels.
And we call them zero or one, like something which is either pointing up or pointing down, that little kind of magnetic moment, or a pulse of light which is there or not there, or a switch on or off. So all the information is encoded in this. And the type of noise that we have is called a bit flip. You have one bit. Let's say you want to send it to me. If it is zero, we'd get zero, but suddenly there's noise between you and me. And then it gets flipped to one, and I get the wrong answer.
The idea of classical error correction is not send your bit one by one, but to encode them so that instead of sending zero or one, you'll send me zero, zero, zero, or one, one, one, three bit for the one. And if one of them flips, you can still recover the information just by taking the majority. If there's two errors, then it's gonna fail, that process. But the process here will take care of the one-bit error that comes in, which would not have been taken care of if you sent it single bit.
So now when you try to translate this for quantum computing, there were fundamental objections that this would happen. First, the noise in quantum mechanics is not discrete like a zero, one, but it could look like continuous. The second one is that it seems that when we have taken a bit, zero, and encoded it in zero, zero, zero, we've copied it twice. And quantum mechanics tells us that we cannot copy quantum information.
And the last thing is that when we try to make this majority voting, then we have to measure the bits. Another property of quantum mechanics that I could have mentioned at the beginning of this podcast, when you measure it, you kill the superposition of zeros and one. By doing this, then you kill the quantum information. So the question was how to get over this and these three objections. And the three ways now seems obvious once you know how it works, but it wasn't around the 1990s.
And I'm not gonna go into all details how it happens, but maybe I'll mention one thing. So, type of noise, it turns out that although the noise can be thought to be continuous, there's a way also of thinking it as being discrete. And the type of quantum noise can be simplified to have either bit flips, the classical noise, or what is called a phase flip. So when we have superposition, there's something called a phase, and this phase get changed from plus to minus.
So we certainly have two types of discrete noise. And the combination of the two makes the type of noise that we have. So the continuous noise that we have can be thought of as a discrete piece, and then we can get over that first objection. And the last two are a little bit more complex, so I'm not gonna mention exactly how it works, but there's a way to go through.
So there is a theory of quantum error correction, and it turns out that classical error correction is like a subset of quantum error correction. It's quantum error correction when we don't have phase errors. We have only bit flip errors. And then in that case, things are lot, lot simpler.
- And these considerations about noise and copying, they are challenges that you need to overcome when you're doing quantum error correction, but are they also kind of advantages when we're trying to encrypt data? - Well, the advantage is that you can keep these superpositions that we don't care in the classical world because your bits are either zero or one.
In quantum mechanics, and maybe I should have said this at the beginning of this podcast, that the bits in quantum information are called qubits or quantum bits, a name coined by Ben Schumacher. And these quantum bits can be in a superposition of being in zero and one. There's one way of kind of having a picture of this. You can take the surface of the sphere as the kind of quantum state that one qubit can have. The classical bits are the north and the south pole. We'll call them zero and one.
But if you are anywhere else on the sphere, then you are in zero and one at the same time. And so these different kind of states allows you to do something different. In fact, the transformation to go from zero to a superposition of zeros and ones is something that classical computers cannot do. And so by having a quantum computer, suddenly you have different types of transformation you can do with your information.
And the hope initially was to find shortcuts, that there would be shortcuts that if you had different transformation, if you can do some things that your peer cannot do, maybe you can kind of find a shortcut to go somewhere else. And indeed, quantum mechanics and quantum algorithms are exactly this. They are shortcut to get to the answer.
- When you showed me around the Institute for Quantum Computing for the first time and showed me what the labs were, and then I would give tours to visitors around the labs, and you would see in one lab, it would be all dark with lasers and mirrors. And you'd see these lasers bouncing off of things. And you go to the next lab, and there's a big, the nuclear magnetic resonance can. I don't know what you call it, this super cool, your quantum computer prototype in one lab.
And then another lab has ion traps. There's all these different approaches. Are they sort of different attempts to find the right way to do quantum computing, or are they all sort of part of the same effort at harnessing quantum information? - They're all different blueprints for quantum computers, and it's not clear yet which one is the winner in all of this. People are making bets of which one will be the best one. Different companies have different ideas of which one will end.
What they have in common is that they all want to manipulate quantum information, and they have different physical implementation of how to do this. In some sense, we can think of, in classical computers today who have all chips will all look the same. But if you do classical computing, you could have an abacus. An abacus is a way of manipulating information, and you can have a slide rule which tells you how to calculate, also. There are different ways of doing this.
Now, I'm not gonna compare the different implementation to the slide rule and today's kind of modern, which one is a slide rule, which one is the modern computer. I'm gonna kind of say nothing about this exactly, but the different implementations are aiming to be the quantum computer. And maybe it's possible that there could be more than one that kind of works. Maybe one will work better for certain applications. Another one works better for some other type of things.
And so investigating those right now are all kind of worthwhile endeavor to do. And there's also, I would say, a spinoff of having these different implementation, which relates to not necessarily quantum computing, but to quantum sensors. So quantum sensors are sensors which use, again, rules of quantum mechanics to better sense certain phenomena. It could be better sense the electric field. It could be a better to sense the gravitational field.
It could be better to field some different properties that we have around. And by studying either atoms or ions or light, then, in that case, they can be more appropriate to certain places. An example of this is light. Fundamental difference between using light as a qubit and an atom as a qubit is that light goes at the speed of light. It doesn't stop. So you cannot take a photon and keep it here. While you're doing something else on your qubit, they are gonna move around.
And so you have to find a way that if you want this photon to interact with that one, that even if this one goes around, that it has come back in the right state to kind of interact. With ion, it's easier because they are in a trap, and they are there next to each other. Now, the difference is if you want to send information which is in an ion trap, quantum information to your partner who's somewhere else, then you have to use light to be able to do this.
So you can transfer information from the other. from one implementation to another one. - I remember you describing that in the early days of quantum computing's being somewhat like the early days of classical computing, you had to try different techniques. And there were vacuum tubes, and that was a step, and that we don't use vacuum tubes now. Do you imagine the future quantum computers will perhaps use elements of what we've seen before, but perhaps things we haven't even investigated yet?
- Yes, certainly we might stumble into a better physical implementation of quantum information which is more robust to noise. In fact, I would say the biggest challenge that we see today trying to build quantum computers is the noise and quantum decoherence. That's why quantum error correction is really important. That's the main technique that we have right now to have the idea of scaling up. But I can see this changing.
There's this beautiful quote from "Popular Mechanics," 1949, saying the ENIAC, which was one of the first classical computer, the ENIAC had 15,000 vacuum tube and weighed 30 tons. And we could imagine the future having computers which would weigh only a ton and have 1,000 vacuum tubes. (all laugh) - Dare to dream. - Yeah, that's it. And if this would've been a scientific paper, you would say, okay, they tried to be careful, but that was "Popular Mechanics."
You would give them a license to kind of dream and kind of having a wide imagination. So it does show that suddenly when transistors were just appearing in labs, totally disconnected, and people didn't think they would be used for computers at the time, appeared. And this changed things completely. Maybe the implementation that we have today of quantum computers are the ENIAC type. Suddenly we could find some form of artificial particle in material science.
And there are suggestion about this called topological quantum computers with anions. Maybe these things, if we can make them in the lab, and they're able to control them, they would become naturally robust to noise and give us a chance to scale up. This is part of the dream, and we hope that we see these things. In fact, it would be very neat to discover something that kind of suddenly makes it a lot easier because today building quantum commuters is very hard, very, very hard.
- On the topic of dreaming in the future, I think this would be a good place to play for you a question that was sent in by a student. So this one is from Mohamed Hibat-Allah, and he's a PhD student at the University of Waterloo and the Vector Institute in Toronto. - Thank you for taking my question. So I'm a physics student at the University of Waterloo. My question is related to quantum computing.
So as we all know, there is a lot of research all over the world for the purpose of building a useful quantum computer. So my question is, what do we need to build a quantum computer that is useful to real-world applications? And what do we need to do to reach the point where quantum computers can outperform classical computers? Thank you.
- That's a very good question and in bunch of different parts, so maybe I'll start by the last part to, like, what do we need to have a quantum computer which is more powerful than classical computers that we have around. And it turns out we're pretty much there. We have quantum computer prototypes around the world, one at Google, one at IBM, that are big enough to do a computation which classical computers have incredible difficulty to solve. So we are just on the border.
And a little bit discussion of if we're there, but to me it doesn't really matter. And so the challenge there is that these problems that these quantum computers are solving are not that interesting for day-to-day application, but I think it's quite a milestone. If I compare this to 10 years or 20 years ago, then to arrive there, 15 years ago, there were people who were saying that we will never be able to build a quantum computer. And here we have a prototype today.
We have controlled these quantum bits well enough to do a computation that the classical computer can barely do. To turn this into a device which is very useful for practical application, then we have to scale the number of these quantum bits. And as we scale the number of quantum bits, it's very hard to make them more and more precise.
If you have an error rate pair operation, which is P, then as you have N of these qubits, if the error pair qubit is P, if you have them, then the error rate goes like N times P. So if you have 100 qubits, it's 100 times higher, and if you have 10,000 qubits, it's 10,000 higher. This tells us that we won't be able to compute in a way which is fault-tolerant or to have confidence in the result if we don't have a mechanism to take care of these errors.
And again, that's what quantum error correction tells us, that we can bring this NP to some constant value and compute. We need to be able to have a device with quantum error correction. At least the focus is there. We don't know how to do this without quantum error correction. And so we need to do this. And that will probably take another 10 years. Hopefully I'm wrong, and it's in three years, or I hope that I'm not wrong that they'll be in 50 years.
But the consensus and some people in industry claim that probably by the end of this decade, roughly 10 years, we should have these devices. And we'll need many thousands of qubits to do this. So the noise has to be thought carefully and how do we control these qubits also. Right now, we do this brute force. We send one little wire for every of the qubits that we have, but if we have a million, how do we kind of link all of this and make all these wires that goes into all of the qubits?
It's not totally clear right now. There's different architectures. I saw something from IBM Open Day last week about making a sandwich of qubits and having wires to come to them in kind of different ways, which I thought fascinating. And although I've seen kind of people talking about (indistinct) architecture, they had very concrete plans to do this. So there will be progress that will happen the years to come. So that's why the field is incredibly exciting.
There's new things every day in this field. - This field wasn't the original field that you got into when you started studying science. You were more interested in the universe in its largest scales, right? You were more into cosmology? - Yes, I was in cosmology, but a small branch of cosmology called quantum cosmology. What is quantum cosmology? So the universe is very, very big, and I've told you a few minutes ago that quantum mechanics is what described the world when it is very, very small.
So these things seem to be, at first sight, contradictory in terms, but the university is very large, but it is expanding. Instead of thinking about the future, if you look at the past, means that the universe was a little smaller yesterday, even smaller the day before, even smaller before. And then we can trace back using Einstein's theory of relativity. We can trace back and ask the question, how long does it take before the universe is kind of small to a point?
And it's roughly about 13 billion years. And at that point, quantum effects should come out. What I was studying is how do we use quantum mechanics to describe the beginning of the universe? So I worked in Cambridge with Professor Stephen Hawking on a proposal that he had called the Hartle-Hawking or the no-boundary proposal. So he was trying to understand how this proposal was kind of fitting what we observe in the universe and does it make sense and try to interpret this wave function.
A wave function is a mathematical tool which describe everything we can learn from the quantum system that it represent. I was trying to understand how it interpret. In usual quantum mechanics, quantum mechanics in the lab, we interpret the wave function as it gives us the probability for something to happen. And we show that we have the right wave function by repeating the experiments many, many times. And then you get the probability distribution of different events.
And this probability kind of maps with the wave function. The problem with this, the universe, we cannot kind of having many of these experiments. We have only one of them. So how do we use this wave function to make prediction? And it turns out that decoherence is a tool, or quantum decoherence is a tool to turn this wave function into probability of classical events. I knew this quite well, and this is part that I learned while I went to Vancouver with Bill Unruh.
And it turns out that when I was at Los Alamos, my mentor, Wojciech Zurek, was probably kind of the best known person in the world working in quantum decoherence. And when I went to this talk about quantum computers, then I could put the two things together. Quantum decoherence was an asset to interpret the wave function of the universe but an impediment to build quantum computers. But again, it's the same mathematics.
I jumped one to the other, and at that time I thought, oh, I'll work a little bit on quantum computers for a few weeks. And then I'll come back to the fundamental issues of quantum cosmology and work with the universe. But I got stuck on quantum computers for a little while. - And actually, one of your current students sent us a question about this time in your career. So maybe we can play that one. - Hi, Raymond. This is Matt Duschenes, one of your students at IQC and Perimeter.
Ray, how did your advisor, Stephen Hawking, react to your career pivot? Did you two still discuss science and quantum gravity topics after you transitioned to quantum computing? - Again, a very good question. So I had the chance, after finishing my PhD, I would bump into Stephen or kind of go to Cambridge from time to time. And Stephen has always been driven by curiosity. This is something which has kind of always puzzled me. When I was a student, he was already incredibly disabled.
So he couldn't do pretty much anything by himself. When I started as a grad student, he could speak then, and he could move a joystick on his wheelchair to move around, but he would not be able to put his leg, or he was barely able to put his leg back on the little stalls of his wheelchair by himself. But he couldn't feed himself, couldn't go to the bathroom by himself. And he couldn't kind of lift himself in his wheelchair, so, incredibly disabled.
But Stephen was curious, and he knew an incredible amount of things. I always wonder, how did he learn all of this if he had to read a book. Today, we read on the internet. We just kind of move from pages to pages or read a book. He couldn't turn the page by himself, of a book. So he had to have somebody all the time doing this. Despite all of this, he had an incredible knowledge on a broad level and curious about so many things.
So certainly when he came to Waterloo, and Stephen did come to Waterloo many times in the last 10 years before he passed away, he was always curious to learn different things. And I remember at some point I asked him, "Are you interested to come and visit the labs?" He's a theoretical physicist, so I was not totally sure if that would interest him. And he was really keen. He said, "Oh yes, absolutely." And then I learned that while he was here, he had gone to SNOLAB in Sudbury.
This is a lab where people have measured the mass of the neutrino. Professor McDonald got a Nobel Prize for the work that they have done this. And then it turns out that the lab to measure this in a mine, and the mine is still active, but you can go and visit it. You go in this kind of elevator. You go down, I think it's two miles down the ground, and then becomes really hot. You go down there with the miners.
And as you get out of the elevator, the miners goes to the right, and scientist goes to the left. And then you walk for about half a kilometer down there. And then you arrive to a place which is a clean room. So incredibly, incredibly clean part. So it's all closed off, sealed off. You have to get a shower before going to the other side. The men and women get on two different parts. They're all stripped off, go through the showers, go and get dressed, and then go and observe things.
By the way, if you have never seen this lab, and you have a chance in Sudbury, just go. It's totally amazing place. But I hear that Stephen was interested, that he went. (laughs) And I say, "How was it?" And he said, "The elevator was great. It was like free fall." So for somebody who was the master of gravity to be in elevator for that long and felt like free fall, he thought it was absolutely fantastic.
So he went to visit there, came back here, and during the week, he came and visited the lab. And at every lab, he had some really interesting questions. And you know that he knew some pieces of all the different parts that we were talking about, which totally, I mean it, I said, "Where does he get all that knowledge?" But it was very interesting. - I remember when he visited IQC a number of years ago, we had a gift made for him that you gave to him, and it was a wooden boomerang.
Can you explain why we chose a boomerang as a gift to Stephen Hawking? - The first project when I was a graduate student of Stephen, which was to try to prove that the wave function that bear his name, the Hartle-Hawking wave function, showed that the arrow of time would reverse at the time of maximum expansion. So the universe got started very small. We have the big bang, which is like an explosion.
At that time, the people thought that the universe would reach a time of maximum expansion and re-collapse. At the beginning of the universe and at the end of these universes, there are something called singularities, places where physical quantities would go to infinity, which essentially tells you that the theory by itself breaks down, the place where something different will happen.
Classical relativity, Einstein's theory of gravity, would break down there and should be replaced by something else. And Stephen had been working that quantum gravity would be what would replace and smooth out these singularities in some ways because he want his wave function that he had picked up, his quantum wave function, was smoothing out the singularity at the beginning. He thought that it would smooth it out also at the end. But today we see entropy or disorder to increase as we go.
So at some point, this would have to reverse and come back so that it goes to a smooth thing. That was my first project. I had to show this. So he had the idea. And then he had to show that the math agrees with the idea. So we've talked about this a little bit before. It's great to have ideas. Some are right. Some are wrong. And usually really smart people, clever people like Stephen, get them all right straight away. And I started to work on the math of it, and I couldn't make it work.
And I would go and see Stephen once a week and show him my progress. And he would always pick apart some of my argument or my comments, or ask me, "Have you checked this carefully?" Well, I was a grad student just starting, so I had not checked everything. And I'm sure other grad students would understand very well that kind of, you go and see your supervisor.
You think you have everything neatly done, but something, oh, there are little pieces under the carpet there that kind of you had assumed. And that's why Stephen would pick at me all the time. Fortunately, after a couple of months, one of his ex-post docs, Don Page, came, and he asked me what I was working on. And I told him, and I said, "I just cannot make it work." And he said to me, "Oh, it's interesting because I've been thinking about this. And also I believe that it's not gonna work."
So then I felt reassured because for the last kind of six months, I thought that I was the one who was wrong. And so Don, who was a bit older than me, said, "Stephen will never agree if we just go brute force and tell him he is wrong. What we have to do is slowly putting the pieces together and build our arguments and add one more piece to the other, which each of these little piece don't say that the arrow of time will not reverse, but kind of all together will kind of bring."
And so we did this, and after about two months of arguing, that both of us kind of were putting, Stephen came and said, "It's never gonna work." (all laugh) And then, so you say, "Ah, yeah." When I finished my PhD, Stephen gave me a copy of his book, "A Brief History of Time" and, well, a quotation at the beginning: "To Raymond, who showed me that the arrow of time was not a boomerang. Best, Stephen Hawking." So when he came to visit here, it was kind of 20 years later.
Then I gave him a boomerang of this. And then he had a big smile on his face. And interestingly enough, in the following year, there was a documentary on him that I was watching at some point, and it was at his house. And I could see in the background, the boomerang was there on the wall, put there. So this is the story of the boomerang. - Well, this actually leads into one more question that we got, and this one was from Nayeli Rodriguez Briones.
She's a postdoc at the University of California, Berkeley, but she did her PhD doing some work with you. And so she actually wanted to know what your fondest memory is with Stephen Hawking, and you've shared a few now, but I'm curious, what's the fondest when you look back on all those times? - Maybe one piece that surprised me about Stephen is scientists, some of them are very stubborn. Some of them kind of have ideas, and they think that all their ideas are right.
One thing which really amazed me with Stephen is when suddenly he realized that he had made a mistake on this proposal on the arrow of time, he totally turned around and said explicitly in conference and talked in his book that he had made a mistake and realized that things were going differently. And he gave me a lot of credit for it. I thought it was incredibly generous. And so that was a part of Stephen that I didn't know that much before, that he was my boss and incredibly smart person.
I was a small graduate student kind of stumbling on this problem. - I can't imagine the intimidation factor of telling Stephen Hawking, "I think you're wrong about that." - Yeah, I must admit that I never said it explicitly that way. I would go to the Blackboard and say, "I don't see how this can work," or "I don't see this working." And then you would go, I would tell him I've done this and this and do this calculation. And if I look at this, I get this result.
And so I'm not getting what you're thinking. So a little bit different than saying, "You're totally wrong, Stephen." Another thing of Stephen which is amazing is his sense of humor. He liked to tease people, and this happened often between him and me and the staff around kind of where we'd play tricks on each other and throw things to each other at different place. He would have a good laugh, and I was not very shy.
And this, I don't know why is that, but many people who arrived next to Stephen totally freeze, and he was very, very disabled. And his voice was distorted when you started to talk with him. When I started my PhD after that, he had a computer. So people would kind of, often if I would be helping Stephen, they would ask question to me instead of Stephen because they don't know what kind of reaction.
Maybe I can thank my mother, who was very hands on and not be shy and always telling us, the kids, to help whenever help was needed. So I would help Stephen, and sometimes I would say to Stephen, "I just don't know what to do now," kind of trying to help him when things were not going right. Stephen's sense of humor comes out in different ways that quite often people don't notice. And once you notice it, he likes to joke. He likes to tease people. He likes to kind of have fun.
That's one part that I really was impressed of him, of somebody who's incredibly disabled. I've rarely seen people as disabled in my life who can still function and do extraordinary things of being a worldwide scientist, traveler, nearly a rock star in many ways but at the same time being incredibly, incredibly disabled. This is something that really impacted me, saying, if you really want to do something, then things are possible.
Like don't let yourself kind of pity yourself and stumble on through things because of hurdles. - That actually leads into my next question. One of the reasons we're so delighted to have you here today is that a doctor told you some years ago there was no guarantee that you would be anywhere today. So you had a prognosis of cancer that came out of the blue. Can you tell us what that was like for you? - Yes, 5 1/2 years ago, I was diagnosed with lung cancer, stage three also.
People who know about this know that prognostics were not that great at the time. So I was told that I had about 20% chance of surviving five years. Except for the lung cancer, perfectly healthy. I like to do outdoor stuff. I like to go on bicycle. In fact, while I was being diagnosed, I went on my bike to a little cottage that we have on the way to Owen Sound very far from me, not very far from here, about 125 kilometers. I could do this, so didn't feel really that that impacted me that much.
But the doctors told me the chances are very slim to go through. And that only made me change a little bit my view of the future. It turns out that I had been director of the Institute for Quantum Computing for pretty much 15 years. I had told the institute that I was not going to ask for another mandate. So I could just kind of quickly wrap up. And I thanked my colleague and partner Kevin Resch, who have taken over at the institute at that time.
And then I had to go to chemo, surgery, radiation, and all that stuff, and more chemo. So many things seems to have kind of picked up, and things went okay. I had a recurrence about three years ago, but it turns out that the particular cancer that I have is, one particular piece of my DNA has changed, and there's targeted therapies. It's a mutation of the DNA, which is known, and there's a drug that kind of came to the market five or six years ago.
And that particular drug kind of stops the cancer to go in, and it's been incredibly good. So, thank you for medical researchers, people who design drugs. And thank you for the company who makes this drug that I'm still alive today. And here I am, and things look good. - You mentioned last time we spoke that there's actually a note on your file at the regional cancer center saying, "Note, this guy's a physicist, and he'll ask a lot of questions."
Can you tell, were you asking questions about the machinery they were using for their diagnostics? - Yeah, in fact, when I had treatment, there was a kind of three piece of it. Surgery, that's kind of biology stuff. And with this, I don't feel that much attraction to that kind of stuff. Then there's chemotherapy, chemistry, all these stinking stuff and kind of liquid things that again, I'm not too keen on this. I tried to understand a little bit how it works, but it's not natural to me.
Radiotherapy, that's radiation, That's electromagnetism. Ah, this, I can read about this. I can try to understand kind of how it works. So I started my treatment of radiotherapy here at the hospital in Waterloo. And I started to ask questions about the machine, and I was really curious to see kind of how it works. So I started to say, "Well, what is the frequency of the radiation that we have?" and the staff who were going kind of, "Oh, I'm not really sure. We'll kind of ask."
And they would come back and tell me this. So some of the question I would ask, they would come back and know the answer, kind of how long does that thing process work and all this, how do you calibrate the machines? One day I came in, I said, "Well, do you have the instruction manual for the machine? I'd like to double check a few things." (all laugh) So they burst laughing. And he said, "We have a medical physicist on staff at the hospital. Maybe we can set up an appointment with him."
So his name is Ernest Jose. And so the next time I came after my appointment, he was there waiting, and he said, "Apparently you're a curious man." And then he told me this. He said, "Apparently in your file, there's a note saying, 'This guy is a physicist and asks a lot of questions.'" - Amazing. Did they ever give you that operations manual for the machine? - I didn't get it through the hospital. I got it through the internet somewhere. I kind of figured it out.
And then interestingly, the medical physicist was giving course on oncology and therapies at the university, which I didn't know about. I asked him if I could get in this course, just be a listener, to which he laughed, and he says, "If you want to, that's fine." So I joined the course, and I spent the rest of the term kind of at the back of the lecture room and asking question from time to time, trying not to intimidate the students who are at the front.
But quite often, they had questions about what do patients feel when certain things happen and kind of questions that patients would have. And so there, I could help them a little bit. So now I know about radiation theory, and I'm thinking that maybe there are ways that quantum technologies could help them kind of controlling, after all of this, kind of comes down to controlling radiation. In the NMR quantum computing that I do, we do control the radiation in certain ways to reach certain goals.
So I think some of these techniques can be applied there. - So you've sort of become interested in new fields a few times, right? You started out working in quantum cosmology, and then you became interested in quantum computing. And now you've acquired some knowledge about medical physics. But I think it must be very difficult when you try to start learning about a whole new field. There's so much vocabulary and maybe assumptions that people take for granted.
Was it difficult to move into those new fields a couple of time? - Yes, there's definitely some hurdles, but I would say this is the price I'm paying for being curious. It's a wonderful thing to be curious because it tells you that you're never bored. There's always some neat things that you can learn about or kind of, I like to understand how things work, let it be quantum computing, qubits, or quantum cosmology.
Or I have a Volkswagen van 1979, and it doesn't always work perfectly so that there are piece I need to understand there. And while you do this, indeed, there's lingo that people are using. And that's probably most of the challenge, is to understand exactly what is the lingo that people do. And all different fields have different kind of assumption that people absorb without saying very explicitly. And these are the hard parts to kind of get over.
But once you know that these are two hurdles, then you can pick the brains of other people. So knowing this and not be shy about asking questions is something which is critical. I try to get my students not be shy. Personally, I'm a person who's shy in kind of lecture room and all of this. But now that I know enough things, then that gives me less shyness to be able to do this. And if my question is a little too naive, then I'll say, "Okay, I'm just new in that field.
So what?" And maybe being a little bit older, I've become less shy of being seen. So quite often, I see students worry about looking like fool of asking naive question. But what I have learned is that sometimes the naive questions are the hardest one to answer like the basic assumptions in fields of why do we do this this way? It's kind of much harder to answer than a really specific technical tricks that people are using to get a solution of something.
So asking things about the assumption is, I think, very important because once you know the assumption, and then you learn a bit of the mathematics, then it's straightforward to kind of move on. - You've mentioned in this conversation several times the power of curiosity. I remember you gave a TEDx talk about 10 years ago, and the very first word of it was curiosity. That was a standalone sentence, curiosity. Is that something that's you've had innately your whole life?
Have you always been that way, or is that something you've developed over time? - That's a really good question. I think there's an innate part that everybody has. I think it comes down to Darwin's principle. Humans who are really curious can harness this curiosity to turn their life into something better. And I think that's the story of developing technologies.
So you start with curiosity, and suddenly people like me will face this kind of trying to understand the world around us and try to ask question about how does it work and kind of why do we see what we see and trying to build what we call theories. Theories at first step try to put together a bunch of data of observations and kind of link them in a consistent way. But a second step of the theory is something that allows us to make prediction of what's going to happen in the future.
And then that's one way that in a scientific matter that we prove that our theories are good, is that they have a predictive power, and they agree with the future experiments that we're gonna do. But also once you know how to predict things, you can learn how to control these physical phenomena that we have around. And once you can control these physical phenomena, you can make them do things that helps you, that helps you to survive. And that's what we call technologies.
And I think the cycle of going from curiosity, making theories, controlling, and then kind of developing these technologies have come again and again. And once you develop new technologies, interestingly, you can push your curiosity steps ahead or further down or further up, depending on the scale that you're looking at in trying to understand more physical phenomena and take that cycle going again and again and again. And so I think curiosity is something that everybody has.
People use them in different degree, I would say, but I think everybody's curious, and that's what makes us drive and kind of ask interesting question. - Ray, I could talk to you all day, and there have been times when we have chatted for hours, but we won't keep you any longer. Can you just share with us what your outlook is for the future and your research and your curiosity? - Well, I think you want to talk for a few more hours. (Ray and Colin laugh) I'm curious about different things.
Now I have the luxury of not being director of any institute. So I have plenty of time to do things which, when I was director, I had to focus on developing this Institute for Quantum Computing, but now a lot of time to read things. And I get lost from day to day on kind of reading too many things, but definitely on the scientific part, I want to better understand how to control these quantum systems.
Also having been a director of an institute, I want to understand how do we really go from ideas and this curiosity to really develop technology, and how do we do this in a Canadian context, for example. What are the piece that we're doing very well in Canada, and what are the piece which are missing? That's another part. And if we go to kind of larger and larger scales, one day, I'll come back to quantum cosmology and trying to understand how the universe works.
And between the small world and the large scale of the universe, there's plenty of things to push my curiosity in different direction. - Wonderful. Well, thank you so much for spending the time with us today. - You're welcome. Great pleasure. (soft electronic music) - Thanks so much for listening.
Perimeter Institute is a not-for-profit charitable organization that shares cutting-edge ideas with the world thanks to the ongoing support of the governments of Ontario and Canada and thanks to donors like you. Thanks for being part of the equation.