SIO431: The Em Drive Is Supposed To Be Impossible. So Why Did Experiments Say It Worked?

Transcript

This transcript was generated automatically and may contain errors — the audio is the record.

0:00:16 Thomas: Hello and welcome to Serious Inquiries Only. This is episode 431. I'm your host, Thomas Smith, and today we've got some fun science for you. Have you ever heard of the M-drive, as in E-M? It was something that was vaguely familiar to me. It's a thruster for spacecraft that got a lot of buzz and seemed to show promising results. But there's just one problem, and that problem is that it should be impossible. It seems to violate the law of conservation of momentum and, you know, other laws of physics, too. I feel like when it comes to laws of physics, it's usually a one strike, you're out. Like, you violate one, you done. But, yeah, the point is, what's going on here? What happened there? What should we think about this? Is it real? Is it not?

0:01:02 Thomas: Well, who better to take us through those questions than Dr. Bryan Gillis. He's going to tell us all about it. And in the process, we're going to learn a lot about not just science, but kind of how science is done. That I think is really fascinating stuff. Before we get to that, I'd like to ask you to please support the show on patreon.com/seriouspod. If you like this work and you want to see it continue, please do so. Not to mention you get to avoid the commercial break that is coming right now for all those non-patrons. Thanks so much to all those who make the show happen. And with that, let's get on over to Dr. Bryan Gillis and talk about the M-Drive. And welcome back to the show, Dr. Bryan Gillis, astronomer. Everyone's favorite astrophysicist, how are you doing, Brian?

0:01:42 Bryan: I'm good. Don't tell that to the other astrophysicists you had on recently.

0:01:48 Thomas: No, I stand by it. You know what? He can write an email. I stand by it, it's so good to have you again. I'm excited, uh, I'm not sure if we're gonna hear like some science is some BS-y thing or an amazing thing. I mean, the thing you sent me, we're talking about the M-Drive, I assume it's called, but in the URL, I know I'm like, oh, this looks like, okay, could be legit, but it says the URL is, slash, just the end of it, slash m drive and cold fusion. So I don't know if that should color my opinion, but I'm very curious to learn if we got some science here or if we have some BS or what.

0:02:23 Bryan: Well, Yeah, I'm not sure what's going on with the URL myself. My best guess is that they changed the title on this relative to their draft. To set the stage, I sent Thomas an article from Wired from back in 2013 when this was kind of blowing up in the news. It's about a type of space thruster for propelling spacecraft in deep space called the M-Drive, which was proposed to solve a lot of the major problems that we have with long-distance space travel. So if you listen to our previous episode on could we get in contact with aliens, one of the big problems we face is that the cost of fuel goes up exponentially with any linear increase in the velocity you want to achieve.

0:03:14 Bryan: Because you have to carry along the fuel before you use it, and any fuel you carry along you have to accelerate up to the same speed as the spacecraft. You do the maths on it, and the amount of fuel you need goes up literally exponentially with a linear increase in speed, it gets to a ridiculous amount. So there have been a lot of thoughts about how to get around this. And the M-Drive, if it worked, would completely obviate this problem, because it doesn't require you to carry around any fuel at all.

0:03:49 Thomas: Interesting.

0:03:50 Bryan: And unlike The one proposed way to do this with a light gun, it actually would be relatively efficient. If compare it to the best solution for deep space travel we currently have, called ion thrusters, which have the most efficient ratio of thrust we get to the mass expended, the M-Drive is supposed to be able to get about seven times more efficient, in force per energy, and wouldn't use up any mass at all as propellant. So for the sake of deep space travel, this would basically solve the one biggest problem we have, that of the immense fuel costs.

0:03:50 Thomas: Well, Wow. Yeah.

0:04:32 Bryan: You still do need energy for it, but the energy doesn't explode exponentially. It just goes up linearly with the linear increase in velocity, which is a lot more manageable.

0:04:45 Thomas: Is So why isn't this on all our space vehicles? It sounds like maybe something went wrong or is it just taking longer to develop?

0:04:48 Bryan: Well, It kind of fizzled out. And if you look into the article I sent, which I'll put in the show notes, you can kind of get clues as to why. Probably the big biggest clue comes in the second paragraph, which I'll read, it says, and It all started with a British engineer whose invention was ignored and ridiculed in his home country.

0:05:07 Thomas: Hmm. Yeah yeah. I'll admit, right when I saw that, I was like, okay, this is, how often is that that real? I don't know, I feel like not, not very often.

0:05:18 Bryan: Yeah. There are a lot of people who come up with some sort of pseudoscience who say, well, they laughed at Galileo, and he was right. And I think the best retort I heard ever for that was, was someone who said, well, yeah, but they also laughed at Bozo the Clown.

0:05:18 Thomas: Yeah. Yeah.

0:05:34 Thomas: For me, it has the same vibe of, it's often in like Christian stuff where they're like, the doctors told me I would never walk again, they said you're an idiot if you try to walk again, you never get, I'm like, I don't think they say that, don't they usually, aren't they usually like, yeah, you might not walk again, we'll see, you know, like, they're not they don't usually ridicule you for stuff like this, they're probably like, yeah, I don't think this is working, like, if it were legit, you know, but if it's a guy with like a perpetual motion machine, then yeah, okay, maybe he's ridiculed, but that's because he was wrong.

0:06:03 Bryan: Yeah, funny you should bring up perpetual motion machines, Thomas, because that was one of the things scientists that pointed out about this, that if this worked as he said it would, it could actually be used as a perpetual motion machine.

0:06:04 Thomas: Oh.

0:06:14 Thomas: Well, That's a good sign that it's not real, because, uh, I feel like that's one of the most, there's not a lot I'm certain of in the world, but that, I think, is pretty a inviolable rule. It's

0:06:17 Bryan: Yeah, Yeah. So that's one of the problems with it. The other problem with it is, so perpetual motion basically means it violates either conservation of energy or entropy. There are two types of perpetual motion machines. One violates conservation of Energy, the other entropy. This one would strictly violate conservation of Energy, which from a scientific standpoint is worse. Entropy is a bit softer of a law in that there's a statistical chance you can get around it. Energy, not really. And even worse than violating conservation of energy. It also violates conservation of momentum, which is, as we've talked about in the past, an even stronger law.

0:07:05 Thomas: Wanted in seven countries for violating all these laws, this is not okay, huh?

0:07:09 Bryan: Yeah, as for how it violates conservation of Momentum, that's actually pretty simple. What it's proposed it does is you would have one of these out there in space, you would turn it on, and without it doing anything to the the outer environment, it would just start moving. So you turn it on, it gains momentum and keeps gaining momentum in one direction. And there's nothing else coming out to balance that.

0:07:38 Thomas: That Seems hard to believe, but I don't know, I guess science is pretty sciencey sometimes. This is magnets or something?

0:07:42 Bryan: Yeah. Funny you should mention that, Thomas. We'll come back to that. So this is why it was ridiculed, because, you know, as we talked about in previous episodes, if you violate conservation momentum, something pretty, it severe is going wrong in our understanding of science. It means we have something, like there's some new force in existence that we don't understand, which doesn't act like normal forces, or space isn't the same everywhere, or something else absolutely insane, saying like it's transferring momentum to something else that, that we've never before detected, like, like maybe it's transforming all the momentum to dark matter or something, it would have to be something weird like that, which, but even then, have we, we don's any way that easy to interact with dark matter, so you can kind of see why people were, were suspicious of it. Is, and yet, despite all of this, there are a surprising number of experiments that seem to indicate that this actually did work.

0:08:48 Thomas: Interesting. Well, I also have to point out, I don't like this, ignored and ridiculed in his home country, like the internet existed, I'm sure that like you put up a paper or a video or whatever explaining it, I don't know why it would be like, well, you know, those British astrophysicists, they, they were full of crap, but once I got into a different country, it was fine, that just seems weird to me, so anyway, that's just rhetorically, feels a little, a little sus, as the children say, yeah, everywhere, that is what I mean, all over the world, it wasn't just In his home country, but okay, you are saying there is experiments that, you know, made it seem like this, would this works?

0:08:57 Bryan: Oh, I'm sure plenty of people were ridiculing it. Yeah. Oh yeah, definitely. That's the interesting thing. So let's get into the details of it. Think I first, let me tell you what an M drive is. Imagine you have, you can picture a stereotypical metal bucket. You've got, open on top, kind of angled sides, and a bottom which is a bit smaller than the top. So, let's imagine you take that bucket, put a light bulb inside, and then close up the top with the same metal as the rest, polish the inside a lot so it's nice and shiny and reflective, and turn on the light bulb. And that's your M-drive, basically. It's ridiculously simple. And I'm going to send you another link to the patent filed on this, which shows this simple diagram. This will also be in the show notes. So this shows you in figure one, diagram of an engine concept. So do you want to describe what this looks like to you, Thomas?

0:09:27 Thomas: Okay. Okay. Okay.

0:10:30 Thomas: Like a coffee maker, it's you put the grounds in the top there, and then there's some, a magnetron for some reason, I don't know what that does with the coffee, but maybe it pushes the water through and then the coffee comes out the bottom. Think it's Gonna work, but I don't, what you're talking about, this, this looks like science to me.

0:10:43 Bryan: Exactly. And you get magic thrust. Yeah.

0:10:50 Thomas: I've seen this. So you're using like a frigging flashlight to move something like I get, you know. Okay. Well, for one, I don't know if you already said this, but like in space, first off, no one can hear you scream. And secondly, you only need like a tiny, tiny, tiny bit of like net kind acceleration, right? Like if you can do any amount, then you just turn it on for, you know, a year and then it gets going. Cause nothing's slowing you down.

0:11:07 Bryan: Yeah, yeah. And in fact, you can get tiny amounts of acceleration a lot of ways. The problem is getting it efficiently. We have one known way where you can turn pure energy into acceleration, which is just a light gun. You turn on the lights and light has momentum and it will propel you. The problem is it's ridiculously inefficient. If you go through the math, would take forever. It would technically work. You don't fall into the problems of exponential growth, but it's so inefficient that even with the just linear growth in energy requirements. It's just not practical.

0:11:21 Thomas: Hmm. Hmm. Um. Yeah. Okay, how is this different from this? Because this kind of just looks like that, like there's a light and that's it.

0:12:00 Bryan: Yeah, and that's the thing that isn't really clear. So this page explains how the inventor, Shawyer, I think Roger Shawyer says that it works. So let me read here. The principle of operation is based on the well-known phenomenon of radiation pressure. This relies on Newton's second law, where force is defined as the rate of change of momentum. Thus, an electromagnetic wave traveling at the speed of light has a certain momentum which will transfer to a reflector, resulting in a tiny force. So, so far, nothing wrong here. This makes sense. You can imagine a big sail in space where sunlight light hits it, bounces off, transfers a bit of momentum to the sail.

0:12:32 Thomas: Sure.

0:12:44 Bryan: Nothing wrong with that.

0:12:44 Thomas: But but this is sounding breath of the wild-esque. This is something where you can, where I don't know if they fix that in the in Tears of the Kingdom, but where you could put the, like, you're, you're standing on the little boat and then you can wave the Korok, you know, leaf, like, while you're on the boat and that fills the sail and then you move shockingly fast.

0:12:46 Bryan: Oh, oh. Oh, Thomas, I would love to do an episode on Tears of the Kingdom. I had so much fun doing physics experiments in that game.

0:13:09 Thomas: Okay, I'm not... Look, here's the thing, I've played a shitload of it. Now I'm now I haven't done... I'm I'm the kind of gamer, here's here's the kind of gamer I am, I don't know if you could guess this, Brian. You know, there's a classic comic about the Breath of the Wild where there's two types of gamers, and one is like, you know, you've got, you're dressed to the nines, you've got every single piece of gear, you know, and you're maxed out and you're going against Ganon, and then the other, like, it's a picture and he has like the tree branch and three hearts, you know, he's like, I'm ready to fight Ganon. I'm the first one, I have to do everything in the entire game. And then I do finish the story, I don't know why, I'm just an idiot. So haven't, that's why I didn't want to like mention, because like someone will say whatever the story request is, and I haven't done that, but I've done like a million other things, so maybe someday. Okay.

0:13:32 Bryan: Yeah. Yeah, I'm that type too, but apparently I'm also the type of gamer where when the game has good enough physics, I will do enough physics experiments in the game to reverse engineer how it works and write up a 20-page document explaining it.

0:14:04 Thomas: Wait, did you do that for Tears of the Kingdom? Oh my god, oh, I can't wait, I am so, so excited for this. Okay, we'll pause, on that, but this is more of the Breath of the Wild branch of physics, uh, than the Tears of the Kingdom one, I

0:14:06 Bryan: Yeah. Yes. Seriously.

0:14:17 Bryan: Yeah, So this is something like actually Mythbusters did this once. And they base this off of like a common problem you see given to high school physics students, where you have a sailboat and on the sailboat you have fan. If you blow the fan into the sail, will you generate any net thrust forward? And it actually turns out to be more complicated than you think in that example, because what actually happens in reality when you do that is is the fan blows out air, and in doing so, it pushes the fan back a bit. The air hits the sail and bounces off. So when it bounces off, you actually get double the effect, because it's not going from positive to zero, like from the fan it was going from zero to positive.

0:15:07 Bryan: From the sail, it goes from positive. To negative. So you actually do get a bit of net thrust blowing into your own sail. There's an episode of Mythbusters that tested this out, and yeah, you do actually get a bit of forward thrust.

0:15:13 Thomas: Interesting. Huh, well, there you go, yeah. Now that I think about it, I, I guess I could imagine, interesting, it's like you're setting up a bigger kind... the the sail is a bigger receptacle for that kind of force, I don't know if that has anything to do with it, but I, as I said that, I was thinking about it, it's like the amount of thrust backwards from just turning on the fan, I guess, wouldn't necessarily, yeah, it doesn't necessarily have to equal the amount you generate by the air hitting the sail, that wow, so that seems to violate things, but I guess, wooden if if it worked?

0:15:43 Bryan: Yeah. Well, in that case, the extra momentum in whichever direction goes into the air around you. So it does all balance out that way. Now, the M-Drive, you can kind of think of it like this principle, that you have an enclosed container and you're shining light inside of it. This light bounces around and hits the walls at either side. And if it doesn't balance out, then there will be a net force in one direction or the other. And the claim of this device is, well, let me get into the next paragraph So what he says is, if the same electromagnetic wave is traveling at a fraction of the speed of light, the rate of change of momentum and hence force is reduced by that fraction.

0:16:34 Bryan: "The propagation of velocity of an electromagnetic wave and the resulting force it exerts can be varied depending on the geometry of a waveguide within which it travels." So what he's saying is, let's say you set something up so that it hits one wall at the full speed of light and another wall at a fraction of the speed of light. So the same photon is going to give more force when it hits at the full speed of like than it does when it hit that a fraction out of the speed of life.

0:16:56 Thomas: Okay I guess, I guess that makes sense.

0:17:09 Bryan: Now this doesn't really resolve the Newton's third law problem, as you know momentum overall still does have to be conserved. So So what's happening to the excess momentum? So if you look at this, the only things you really have in this problem are the two sides of the container and the photons inside. Let's just say magically you have a way that photons hit one side at the full speed of light, the other side, I don't know, half the speed of light. And momentum gets transferred to the photons whenever the container accelerates. So you've got two things involved, container and photons. Container starts to move one direction. In order for momentum to balance out, the same amount of momentum in the opposite direction has to go somewhere, and the only somewhere it can go is the photons inside. Inside. So the photons inside are going to start moving more than not in the opposite direction of the container.

0:17:17 Thomas: Mm-hmm. Huh. Okay.

0:18:16 Bryan: So are you following me with this?

0:18:18 Thomas: I'm not sure, when you say the photons, I don't know if, do you mean, are just the ones, the same, the very same, or you, or they're like photons just hanging out, like, like background photons? Okay.

0:18:21 Bryan: Yeah. Okay, let's not think about photons. Let's say we've got an empty container with a big bouncy basketball inside of it. The basketball goes through, it hits one wall, starts it moving in that direction because it transfers velocity.

0:18:36 Thomas: Mm-hm. Well, okay, this is where I think this is already where I'm confused, what about the part where you threw the basketball and that set you back a little? Where's that? Oh,

0:18:50 Bryan: So what's happening in the real case here is that a lot are coming out of the light source. You've got multiple ones. There's an initial burst of energy that spreads them out evenly in multiple directions. So if you want a basketball example, it could be that you've got two, yeah.

0:19:09 Thomas: Oh, I see, you're shooting basketballs in every direction, and so there's no net kind of, you're, yeah, it nets out to zero.

0:19:16 Bryan: So let's say you just start out with two taped together, you put a little explosive in between, you set it off, and now you've got two basketballs flying in opposite direction.

0:19:16 Thomas: Okay. Okay. I like this. Okay.

0:19:26 Bryan: And so we can do it with two but let's just focus on one basketball. So you've a big container with the basketball going around inside. Side. You can imagine the basketball slams over to one side, hits the edge of it, and transfers some momentum to the container itself. Container starts moving in one direction. Basketball bounces back in the other, hitsthe other wall, and it transfers some momentum to the other wall. Now, what he's saying happens here is that over time, he set it up so that the basketball is always going faster when it hits one of the walls than when it hits the other wall. And so over time, the whole thing will accelerate in one direction.

0:19:31 Thomas: Huh. Yeah. Okay, okay.

0:20:13 Bryan: Now, the problem is from conservation momentum, that momentum would have to go somewhere, and the only possible somewhere is into the basketball. So if this possibly happens, let's say he set it up so that it works at least temporarily, then once the container starts moving a decent amount in one direction, the basketball, on average, is going to be going in the opposite direction.

0:20:39 Thomas: Interesting, let me, this may be really stupid, but is there some way that, like, if you're trying to, you're try and conserve momentum, is there any way that, like, some momentum could go out perpendicular to both directions and, like, cancel out that way or something? Yeah.

0:20:46 Bryan: Yeah. Well, perpendicular directions won't cancel out things perpendicular to them. It's like, Like, yeah, yeah.

0:21:00 Thomas: If it goes out, so, like, let's pretend we're just using four directions, whatever, north, south, whatever, the other ones are, uh, if if you're saying like, okay, he's got it so that when the basketball goes north it's on average, it's going faster than when it comes back and hits south, right? This is like the basic concept. I'm wondering if, if, like, based on the the geometry, if a way that that is done is like some equal amount east and west momentum happens, and that, that stuff cancels out, and then the difference applied that way to the north, or is that not possible?

0:21:16 Bryan: That's what he claims.

0:21:33 Bryan: Well, the thing about momentum is momentum is a vector, which means it has a directional component to it. And in Euclidean space, which we're more or less in, perpendicular components don't really interfere with each other much when it comes to vectors. It's like, you know, imagine the Earth was flat. If you wanted to get east of here, there's no amount north you can go that will change how far east you need to go. Now, obviously, in reality, the Earth is actually round. But if,

0:21:37 Thomas: Yeah. Well, but I I guess I was looking at the angled

0:22:08 Bryan: Yeah, I

0:22:09 Thomas: Sides, like that won't convert a certain amount of it, you know, different directions.

0:22:12 Bryan: Think you've hit on the key point here, Thomas, is that's where the trick actually comes in. Is he wants there to be some way that it hits one side faster than it hits the other side. And how do you arrange that to happen? And the answer is the angled sides. This is what he calls the wave guide in his patent, is he want to set something up. So basically, if I'm interpreting what he says as charitably as possible, filling in the gaps. Light always travels at the speed of light. The only way it could impact at a lower speed is, well, this is in a vacuum. If it weren't in a vacuum, you could put it in a refractive medium, which slows it down, but that's not what's going on here. The only other way possible is if it hits at an angle. So, if it hits at a very shallow angle, then the velocity it impacts with is going to be much less. So that's like, instead of the basketball slamming into the wall, it's like skipping off of it at an angle. Now, yeah.

0:22:13 Thomas: Yeah. Okay. Yeah. Yeah. I think that's kind of what I was trying to get at, because I was thinking, like, if it hits that angle, and maybe this wasn't exactly right, but if it hit that angle, some amount of that is going to diffuse left and right, you know, and that would account for difference, the that's kind of of what I was trying to say.

0:23:33 Bryan: Yeah. Yeah, so you're onto something here, it isn't quite what diffuses left and right, but what happens before it hits that wall. Because at some point between hitting one side straight on and hitting the other side at an angle, it's going to have to change direction. And the only way it can change direction is if it transfers its momentum to something else. And this is kind of where the design of it comes into play is the angled walls act, as he says, as a waveguide to cause it to change direction. Now, if you're careful about the physics here, you can say, okay, I'll allow the sides to change its direction, but that operation is also the ball bouncing off the sides of the container. So essentially what's happening here is instead of the ball transferring all of its momentum in one bounce when it hits the bottom, it does it in maybe three bounces. It hits one side, hits the button at a shallow angle, hits it the other side, and then goes straight back up to hit the top.

0:23:35 Thomas: Oh, okay, yeah, how does it go straight, though? How does it always end up straight? It doesn't go, it back at the top at a weird angle?

0:24:53 Bryan: It depends on the specific geometry of it. It's not too hard to set something up. I think if you do a 120 degree angle between the bottom and each side, I think, if you'd do that, the mats will work out so that it'll do three bounces and always come straight back. There is a bit of quantum mechanic weirdness that it will always end up a bit off, but the specifics of it hitting straight on don't matter. The issue is that that in any of these simplifications, you have to account for the pressure of the light against the sides of the container. And that's the big thing he leaves out.

0:25:31 Thomas: Yeah. But okay, that's actually what I'm saying, though, but if it was going at each side in an equal manner, you would think just like the structure of the container would, would kind of absorb it in a way, or is that dumb? Yeah.

0:25:34 Bryan: Yeah.

0:25:43 Bryan: Yeah, that's kind of what's going on. So you can see in the diagram here, he shows two force arrows, one pointing at the top and one at the bottom. But he should also show one against each wall of the the container. And so whenever you have a pressure force, it acts perpendicular to the container. Like if the pressure against the flat ground is pointing directly down, the pressure against the wall is pointing correctly into the wall. So if you have two slanted sides of a container, they're pointed into that side, which means it's somewhat out and also somewhat down.

0:25:56 Thomas: Yeah.

0:26:20 Bryan: Now, once we do that, we're changing it from having two arrows, one up, one down, to one up one down one right and a bit down and one left and a bit down. And if you add up those bits down, you see where all of the forces balance out in the end. And this is the thing that he left out of his patent. And yeah,

0:26:20 Thomas: That's why, all the British scientists ridiculed them, specifically British ones.

0:26:48 Bryan: Yeah, because it's a very simple mistake, but it worked. That's the weird thing. Or at least he says it works. His initial thought process around it is provably wrong. He's added some wrinkles to it, saying things like, oh, it's not the individual velocity of photons, it'S the group velocity, which not even going to get into it. It's nonsense, it doesn't work that way. When asked why it can violate the third law of motion, he gives some muddled statement about the special law of relativity, which doesn't help. Basically, his proposed science is nonsense. So we just have to grapple with why do experiments show that it worked.

0:26:53 Thomas: Yeah. Okay.

0:27:34 Thomas: They're fake.

0:27:35 Bryan: Well, that's the thing. It were just him doing this, that would be plausible, but he convinced enough other people to give it a try, and they claimed that they found the same results. So one possibility looking at this is, okay, he was wrong. He tripped over the finish line, however, and found something else that actually did work. And, you know, one interesting thing is while this was going on, another inventor came up with a very similar device. This is Guido P. Fetta, which he called the Cannae Drive,

0:28:14 Thomas: Italian scientists ridicule them for? Oh, hmm.

0:28:17 Bryan: Oh, no, it was a Scottish scientist who ridiculed him the most, I'll tell you why in a moment, which, which is basically the same thing, except it had lots of extra slots inside, and and he said, no, these slots are where the magic happens.

0:28:27 Thomas: It's for the power crystals.

0:28:30 Bryan: Yeah. Now, eventually this got tested. Um, when they did the test, they tried one version without slots, and found that it did did the exact same thing. So slots were completely irrelevant. And once you take out the slots, it's exactly the same as the M drive. So you've got another case of somebody basically tripping over the finish line into whatever is going on here. But now let me go back and ask you, do you know what the word cannae means in Scottish English?

0:28:31 Thomas: Yeah. Oh, My god very Interesting. Canna? Oh,

0:29:02 Bryan: Yeah.

0:29:03 Thomas: Cannot.

0:29:03 Bryan: Yeah, cannot. So, basically, he's calling it the Cannae Drive. Yeah. I don't know how he chose that name, but, well, it's like the time I think Chevy released the Nova in Hispanic parts of the world, where Nova in Spanish means doesn't go.

0:29:04 Thomas: Yeah. I right, yeah, Nova, yeah.

0:29:24 Bryan: Do like that name just because of how telling it is about the whole thing around it. And I wonder if he might have known that. I just can't find any information on how he got to that name. But okay,

0:29:28 Thomas: Uh, so the speed holes in the car don't do anything. Yeah.

0:29:39 Bryan: Okay, so we're get into the question of why is this showing results? So I think first, maybe we should go into what does it mean to show results in an experiment? So let's say they put it in a lab. The lab does their work, and they publish a paper, and let's say they tell you, well, we found out that this actually produces 50 newtons of force, plus or minus five, in the forward direction. So what does that mean?

0:29:51 Thomas: Probably.

0:30:10 Thomas: That it produces 50 newtons of force, plus or minus five, in the direct... I don't know, I'm not sure a better way to say that than what you just said. Yes.

0:30:16 Bryan: Well, what does it mean when you hear it, or how do you interpret it when you see a scientific result that gives like a plus or minus something.

0:30:23 Thomas: The margin of error, right, it could be that's how much the measurements might be potentially off.

0:30:25 Bryan: Yeah. Yeah. That's the trick, is it's not how much they might be potentially off, it's the amount the known factors influencing it could cause it to be off. The big thing about any of those margins of error is that they only ever quantify the known sources of error. They don't quantify the unknown things that could have influenced it.

0:30:50 Thomas: Sure. But if you reproduce it a bunch of times and you get this error in the same direction, you wouldn't think it was error, right?

0:31:01 Bryan: Well, that's the thing is that's why replication is so important. Now, you have to be careful to check that all of these things.

0:31:09 Thomas: Are people getting a bunch of different results and not reporting it? Uh,

0:31:12 Bryan: Yeah, that is one of the things. What if somebody somebody said, could this possibly work? They're doubtful of it. They it's easy to make. It's a light bulb in a bucket, basically. And it produces a surprisingly high amount of force.

0:31:25 Thomas: Oh, are we kind of p-hacking in a way? Yeah.

0:31:27 Bryan: Yeah, that's one thing that could be going on, is the people that see this positive force due to whatever influence going on, we don't know what's going wrong, but maybe it was windy that day, there was a strong magnetism going on that pulled it forward, they got positive results and like, oh wow, this works, whereas somebody else was testing, is like, nothing, oh, this doesn't work, this is junk, don't bother with it.

0:31:37 Thomas: Some measurement, something, problem.

0:31:53 Thomas: Yeah. Exactly, somebody else was like, oh, it's going the other way, it's actually, it's actually accelerating backwards. Okay, oh.

0:31:55 Bryan: Yeah, oh, Surprising you should say that, Thomas, let me send you a link that this is also in the show notes, to to everyone else, I apologize, this link is in Chinese, but from what I understand of what it's saying, sometimes it did go backwards. So this paper, go down to page eight, and on the right panel, you should see figure 18. So yeah, it's a way back machine.

0:32:04 Thomas: Okay that doesn't yeah loading

0:32:29 Bryan: The original's gone now, but this But is, I think, a preprint they posted. 8, figure 18. So on the right side, from looking at what other people have said is going on, this is from the lab in Xi'an, China, that tested this many, many times. The serial number is the x-axis of the plot. It's basically which test number they did.

0:32:27 Thomas: Loading, oh, okay, huh.

0:32:55 Bryan: And the y-axis is the amount of force it provides where positive is forwards, negative is backwards. And you see a line zero right in the center. And just kind of as you joked, roughly half the time, if not more than half the time, yeah, it's actually producing force backwards more often. So what happened early on, apparently, is the experimental setups just by chance were ones that showed some force forward and those got reported. Oh, wow, this really works. But once people started testing it in earnest, they said, well, half the time this goes backwards. It's a large amount of

0:33:10 Thomas: But yeah. I was gonna say, it looks like more often it's backwards. Yeah.

0:33:42 Thomas: Because yeah.

0:33:42 Bryan: Force.

0:33:43 Thomas: You're trying to, you're trying to measure something really minuscule, right? Like it's just barely producing any net force, right.

0:33:46 Bryan: Yeah. Yeah. Yeah. Most of these tests so far have been low power tests. And this is a pretty massive drive. You're looking at the force of light on the inside of a big container. So yeah, there's going to be error. Now, when they went and calculated, okay, what's the statistical error on this putting in everything they knew? It was a very statistically significant result. Like the The error bars on any of those individual points were small, but when they did multiple tests, and unfortunately, I don't read Chinese, so I don't understand the details here of how exactly they reset things for different tests, but however they did it, they found that between different tests it got a very statistically significant something different every time.

0:34:37 Bryan: Like, the first time it was positive 0.3, the second time it was negative 0.5, then the third time it was positive0.5. what this is really saying is the error bars are really only the known statistical errors. They weren't the true errors on it, and This is actually something big problem with a lot of science reporting and how things get published is that the errors you quote are only ever going to quantify the errors, you know, about.

0:34:54 Thomas: Yeah, yeah. Right Right. It's Not like God is coming down being like, "Hey, by the way, like, uh, you, that experiment you're doing, you're possibly off by as much as 10 joules," or whatever. It's like, no, you're trying to figure out what you could be off by. But it's, uh, it's the old known knowns, known unknowns, and Unknown unknowns.

0:35:29 Thomas: Yeah. Why Did that asshole have to say something That was pretty Applicable. And remember somebody else say that in a Different context? Please, ah.

0:35:28 Bryan: Yeah. The Donald Rumsfeld thing is you have your. Yeah, yeah, OK. Well, Thomas, you see, you have these known knowns. You have your known unknowns, your unknown knowns, and your unknown unknowns.

0:35:41 Thomas: Okay. Yeah. Very Interesting New concept, okay, yeah. I get it.

0:35:46 Bryan: Yeah. And so what's quantified in the statistical error are the known unknowns. But the unknown unknowns will still cause error. You just have no way to quantify them. And this is a problem throughout anything reported in science. And it's not something I think there's even an easy solution for, aside from independent replication, ideally through as different an experiment as possible. You can find some completely different measuring apparatus, there are multiple different ways to measure the force. One of the ways here, they used a a pendulum type.

0:36:27 Thomas: Metachlorians? Oh.

0:36:29 Bryan: Yeah.

0:36:29 Thomas: Yeah, sorry.

0:36:30 Bryan: Another way you measure the midichlorians, or you could put a spring, or something like that, oryou could apply it for a while, see what the change in velocity is. You could try all of those, and then you also try it in different directions, do all of that, and over time, you can come to a better idea of, you know, at least, include more more unknowns in it, even if it's not necessarily everything.

0:36:39 Thomas: Mm-hmm. Yeah. You don't want to use this same, like, apparatus every, every time and be like, "See, I did a bunch of times." And it's like, well, if you, if you are still producing an error that you don't know about, you know, there's no way to, so you're saying like, if, if your varying it a lot, like different ways of measuring different things, then you can kind of isolate what is actually in effect and what might be an error of a given, you set up.

0:37:08 Bryan: Yeah.

0:37:17 Bryan: Yeah, and the more different setups you throw at something, the more confident you can be that what you see is actually real. Now, the way science works, if you dig down into the philosophy of science, is you can never be 100% sure of anything. It's all what they call inductive reasoning. Can't, you You can can never prove that something is the case. You can just make it more and more likely.

0:37:42 Thomas: Yeah. But we won't worry about that. That, that's true of All everything, all the time, so that cancels out, right?

0:37:45 Bryan: Yeah, yeah, yeah. But I think the important point here is that that becomes more important when you're talking about things that seem initially implausible. Imagine you're NASA and you need to invest money in a new type of thruster. And you see two proposals. One is this M-drive. It shoots out nothing but starts to go for seemingly no reason. And the other is an ion drive. It shoots our ions. It isn't as efficient as the M-Drive claims to be, but it works on understandable physics. Now, you can look at the experimental results for both of them. See, okay, they both claim statistical certainty that they work, but are you going to give equal credence to them both working when one seems much less plausible than the other?

0:38:41 Thomas: Yeah. And when you have an explanation for one and you don't For the other, really, or one that you accept anyway.

0:38:46 Bryan: Yeah. Yeah. So this actually goes back to something you brought up in the final episode, the one you did alone on simulations. You were talking about Bayesian priors, Where in the case of simulations, like say, okay, either there are no simulations or we can but don't, or we cannot do and it's one third for each. And we won't get into that, but you kind of highlighted how that's a ridiculous way to assign priors.

0:38:58 Thomas: Yeah. Right. Yeah. Well, just To clarify, I think it Can be okay, but then if you're not ever collecting any data, like if you set, okay, we got three options, we'll set priors at 33, it is a way that it can work, but you would need data to be able to, like, hone those into something more interpretable that but if you're just setting those priors at a third, a third, a third, and you never really get any data, they're kind of meaningless, like, you, they're just arbitrary, okay?

0:39:17 Bryan: Yeah, yeah.

0:39:43 Bryan: That's a good point there. Now, we're at the stage where we do have data on each of those. And we should think about, like, let's say for point of argument, both of these have the same statistical likelihood of working. So let's take the standard threshold they use in science is two sigma, which is two standard deviations. It's a mathematical term. You can just think of standard deviations as the typical error you might expect in any measurement.

And statistically, if something is measured to be at least two standard deviations away from whatever threshold you set, in this case, the threshold would be zero, then there's only about about a 5% chance that something which actually had a value of 0 would produce a result this extreme. Have this and you look at either of them. Okay, well, that one doesn't seem likely in the first place, but produced an extreme result versus that one, which does seem likely and produce this result. What a lot of people are initially taught about science is to just take, okay, there's a 5% chance that it could be this extreme if it were nothing. So there's the 5% chance that this is real.

0:41:02 Thomas: Right, yeah.

0:41:03 Bryan: But that's not actually how statistics work. Doing that is what's called frequentist statistics, which seems to work at first glance, but doesn't really. So You there's a two sigma measurement means that there's 5% chance that something which actually is no effect would, would, by statistical chance, produce a result as extreme or more extreme than what we actually observed.

0:41:03 Thomas: So Yeah. Uh-huh. So that, okay, so I have taken that to mean that's that, that, like, 0.95 thing, right? That, that's a, like, standard, uh, p-value, whatever, thingy. So I'm Very glad you're seeing this, because I think I have, as a layperson, just thought, oh, okay, so there's like a 95% chance this is real and a five percent chance it's fake, and you're saying no.

0:41:41 Bryan: Yeah. But, it's actually, a bit it different from that. And we can go into some examples.

0:41:53 Thomas: This Is Important, yeah. Oh.

0:41:55 Bryan: So what I'm saying is, if it were fake and we measured it, there's a 5% chance we would get something this extreme, but that's not the same as the chance that it's real. So let's take an example. Let's do a real world example, which actually impacts people screening for breast cancer. Now, yeah, so most people don't have breast cancer, it's thankfully a minority. I'm not sure what the actual numbers are, but let's say if you screen a certain sample of people, only 10% of them will actually have breast cancer. Now, let's say you have a test for breast cancer, which is pretty accurate, but it's not great. If they actually do have breast cancer, there's an 80% chance it'll show positive, 20% chance it'll show negative. If they don't have breast cancer, it's an 80% chance that'll show negative and 20% it'll show positive.

0:42:08 Thomas: Okay. Yeah. Yeah. Yeah.

0:43:00 Bryan: So the problem we get is, okay, 10% actually have breast cancer. Of those, 8% will be correctly told that they have breast cancer. Then 90% don't have breast cancer, and of those, 20% of 90%, which ends up being 18%, will be told, yeah,

0:43:01 Thomas: Right.

0:43:20 Thomas: Basically You're Going to get way more false positives than real positives, and so if somebody has a positive result, it's more likely that they don't have it than they do have it, which is crazy. I thought that was a different Thing than what we're talking about. What's the name for that? I remember that, that thing, okay.

0:43:33 Bryan: Yeah. From the same statistics, that you have to look at the Base rate, I think, is the term.

0:43:39 Thomas: Related Yeah. Yeah, okay, hmm.

0:43:43 Bryan: So this actually happened to me recently. I was going in for a regular checkup with my GP. They took my blood, screened me for all sorts of things. And they came back a week and said, yeah, something looks like it's up with your liver. Don't worry yet. Just book an appointment in a couple of months. Take the test again. Now, I don't drink any alcohol. And I looked up medically what could be wrong. And they say, well, very occasionally, somebody who doesn't drink will develop something with their liver, but it's pretty rare. So I was in this myself. So So I came back a couple of months later, took the test. They called back, said, no, it's negative that time. We still want to be sure, so we'll test you again. But it looks like, you know, that just happened to me. I was most likely a false positive.

0:44:31 Thomas: That's Interesting, you know, I've always wondered, I swear, I, like, okay, this is gonna sound weird, but I've had so many friends and, uh, you know, a partner or two, who went in for a pap smear and they're like, "Hey, there's something we gotta check out, you know, there's a, a there's thing we got a biopsy," like to the point, like, it's, it's like half a dozen times I've heard of this, and I was always like, it never is anything, that seems weird that it's that much, I wonder if it's same kind of thing where it's like It's More False positives You're screening a lot of people, and so you're gonna have a lot of false positives, and then they go in and do the check, and then it's, well, because at first I thought, is this an insurance thing?

You know, because you gotta remember, I'm, unlike you, I'm living in a country where it is just for profit, so it's like, do they just do it to get the insurance money, you know, just, oh, something looks weird, you know, all the time, is it going to be over-reported, but maybe it is just this, where it's, that initial screening is going to just have a high false positive rate.

0:44:54 Bryan: Yeah, it is. It's exactly this.

0:45:21 Bryan: Yeah, and that's pretty much what's going on. So what they're doing here is let's go back to the breast cancer. We've done the first screening, and we've got a number of people who tested positive for it. So of the initial, let's say it's just exactly 100 people, eight of them do have breast cancer and tested positive, 18 don't have breast cancer and tested positive.

0:45:36 Thomas: 100 people, okay. Right.

0:45:46 Bryan: So of that sample, it's about one-third actually do have Breast Cancer. Compared So to the initial sample of 100 people only 10% had it, Now, with this smaller sample, about a third have it. So anybody in this sample, it's more likely they have breast cancer than the initial larger sample, but it's still not on the whole. So what you want to do in this case is, well, you know, if these tests are random, if they aren't correlated by person, if we just test each of these people again, we can get a bit more certain of it. So once they're in this sample, they give them another test. Like for me, they ask you, okay, wait a couple months. So like your hormones and stuff are re-randomized over, you know, what they might be any given day and test you again. And each time you do this, the purity of the sample gets more. Now, one thing you do have to worry about is that you do miss some people who are actually positive each time. Like, remember, 10 people had it, but two of them were missed.

0:46:07 Thomas: Yeah. Yeah. Yeah. Yeah. Mm-hmm. Yeah. Mm-hm.

0:46:51 Bryan: So I think that's why they want me back a third time. I was positive once, negative once. And they're like, well, you're already a bit at risk because you were positive once. We don't want to go overboard and throw you out. So we're going to test you a third times just to kind of be sure.

0:47:10 Thomas: Yeah. And with medical stuff, I know that very often the false positive rate might be one thing, but the false negative rate is like zero. Or, you know, like one of them will usually be way less. And that kind of helps compensate for that. Because I think they they don't want people to have it and not get detected. So I think they probably err on the side of detecting more. And so usually it's like there's not going to be too many people who have false negatives, but the false positives might be more kind of thing.

0:47:19 Bryan: Yeah. Yeah.

0:47:35 Bryan: Yeah. And that's something that goes into test design is like often these tests, they can choose like what threshold do they use to determine positive and negative. And depending on what you choose will alter what the false positive rate is versus the false negative rate. And so there's a balancing act there.

0:47:38 Thomas: Yeah. Yeah. Right. So how did we get from drives to that? With

0:47:57 Bryan: OK, so we're talking about somebody proposes a thruster and says this works. Now, let's imagine instead of a hundred people, maybe with breast cancer, we have a hundred prototype thrusters with different designs. People are throwing physics at the wall to see what sticks.

0:48:12 Thomas: Light, cancer, yeah, yeah, Okay.

0:48:18 Bryan: And let's say that some of these do work, but a lot of them don't. And you run a test on 100 thrusters, let's say we have a sample, 10 of them actually work, will always give a positive result, they're that good, and 90 of them don't work. But remember what the statistical significance is. Something that doesn't work will hit two sigma significance, 5% of the time. So 5% of 90 ends up being about 4.5. So of those 90, you'll have four or five that just through statistical chance looked like they might've worked. And so we end up with a problem that, okay, we measure them all. We've got, let's say four ones that don't work, but looked like they did, and 10 that do work and looked like they did.

0:48:47 Thomas: Well, in this case But if the the bare reality could be zero of them actually work because it's not possible, then you're just looking at some false positives, right? It's as simple as that.

0:49:14 Bryan: Yeah. Yeah, exactly. And so the way we handle this mathematically to try to put some numbers on it is what's called Bayesian statistics, which can be applied to either of the problems we're looking at here. And in the case where the initial probability is zero, you put it through a famous equation called Bayes' equation to calculate, okay, we've got initial probability, these are our priors, then we've got some new information, and we basically multiply it together, normalize it, and we've Got our updated probability. And if the initial probability is zero, the updated probability will always be zero.

0:50:05 Thomas: Yeah. No.

0:50:07 Bryan: If the initial probability is small, the smaller the initial prior is, the smaller, the upgraded probability will be. So the prior does matter. Now, in the case of this, I wouldn't say the chance of it working is zero because, you as extraordinary as it would be for so much of physics to be wrong, we can't rule out that it's possible.

0:50:29 Thomas: I just meant in a true sense, yeah, like Yeah, no, not that we could assign that probability, but from a like, you God's eye point of view, if we're talking about something that's physically impossible, then the the true probability is zero and we if we do enough of the right test, we should get closer to that. You know, we're never going to be a hundred percent accurate, but like you should get closer to the true probability and if it's zero, it's here. I just, I just brought that up because like you're not going to have the false negatives, really, if something is physically impossible. Yeah.

0:50:31 Bryan: Yeah. Yeah, yeah, yeah. That's right. So And what you can do for something like this is do repeated tests and you will approach the result. If you look at the math, the initial prior does stick around and no matter how many tests you do, the The initial prior will still influence the final result. And

0:51:18 Thomas: But the more you do, you know, vanishingly small, right? I mean, if I, I was going to bring that up before. Like if I said, if, pretend we're we're an alien who knows nothing, or just a somehow a sentient creature that has no information, and you're like, okay, I got a coin, what are the odds that it'll be either side? And let's assume it's true coin and for some reason I'm like, well, that side looks better than the other, so I'm going say it's like 25 on one side and 75 on the other side, you can assign that as your prior. And if you never flip the coin, like that's nothing, just like doesn't it, like useless data, but if you flip it a gazillion times, it will approach 50-50, um, because just mathematically, like you say, I'll never truly reach whatever, but it's that, it's the the principle what was that called? Whatever that is, it will approach the probability over time.

0:51:49 Bryan: Central limit theorem, I think. That might be what you're thinking of. Yeah, and that is basically what happens with enough. That's why they want to do like repeated blood tests on me is they said, okay, well, you've got, yeah.

0:52:03 Thomas: Yeah yeah Central. Thank you. I think so. See if you got some thrust in you, you know, see if you maybe, yeah.

0:52:16 Bryan: Yeah. Yeah. The initial probability is low, but we want to hone in on the real probability to see for sure. That's one of the things you have to take into account. And I think that's probably the biggest takeaway when you hear claims of positive results is that the criteria used for reporting things is just this simplistic, pure frequentist, does it hit two sigma a significance. But something that doesn't work will hit that one in every 20 times.

0:52:44 Thomas: Yeah And you got the file drawer problem? A bunch of light bulbs are just sitting in a file drawer that didn't work somewhere. So is that real? Simple is that, for for this drive, I mean this M drive, I do remember seeing this back in the day. Now that, now, now that you've uh explained it more, and now we're going on a decade, right, or more, it did this just kind of turn out to be, well, nothing, or was there something there?

0:53:02 Bryan: Yeah. It did turn out, as far as we can tell, to be nothing. Now, I think there are a few more subtleties to why there was initial hopes for this. And the statistical things-based theorem were part of it, certainly. I think another part of it was a bit of human bias creeping in to how people did the experiment. Guy behind this truly believed in it.

0:53:11 Thomas: Hmm.

0:53:31 Thomas: The guy was like holding it, he's like, I'll hold it steady, and then he's like, oh, it's like, no, I just

0:53:36 Bryan: I don't think it's that malicious. Think he, yeah, from everything I can tell, he He honestly believes in it. He's not a fraudster. He's just wrong. But the fact that he believes in it. So let me give you an example. Let's say you want to measure something like you play hockey. Let's see. You want to make sure the strength of your slap shot. How fast does it go? So you go out, you set up your experiment, you put up like a video camera, check how the puck is moving frame by frame, do your your calculations, it comes back and says 200 miles per hour. What do you do?

0:53:42 Thomas: Okay. Yeah, well, I know that, that's insane. Uh, the fastest slap shot ever I think is 108 miles per hour. It might be 109 I know, dat off the top of my head.

0:54:23 Bryan: Yeah, yeah, yeah. So you get a result that seems obviously wrong. You say, okay, well, I made a mistake somewhere and you go back to it. So let's say you do it and you say, minus two miles per per hour. Well, okay, I know the puck is going forward.

0:54:32 Thomas: Yeah yeah I know I can hit a slap shot like I'm not that bad. Okay, yeah.

0:54:43 Bryan: Yeah, yeah. So now let's say instead of hitting a slap shot, you're measuring your thruster. You've done the math. You know this should work, but you're measure zero. So obviously something is wrong in the experiment. So you go, you look over your math, you find something. Well maybe that was a mistake. You fix it.

0:54:50 Thomas: Yeah right yeah Sure, or yeah, like a little instrument that you adjust or some sort of something, yeah.

0:55:04 Bryan: Yeah. You you look at the experiment, you find something that could be better, and you keep fiddling with it until the result is within the range you think might be right. So like you're measuring your slapshot, you know, if you calculated 50 miles per hour, you'd say, oh, that seems reasonable.

0:55:07 Thomas: That's interesting. Wow. Yeah that is probably right, yeah. Yeah

0:55:22 Bryan: If you calculate 200, no. You calculate 100, you're like, I don't think I'm that strong. Wrong, but 50, 60, you know, you could still have mistakes in there in what you're measuring.

0:55:29 Thomas: Right but if the mistake is plausible, you are not really going to catch it. That is interesting, but it is your version of plausible, so you're saying the bias leaks in because, like, what if, uh, just for pretend, what if it turns out I just happened to like grow a robot arm I didn't know about and and I actually was capable of like an 150 mile per hour slap shot, I keep getting that like result, I'm like, well, this isn't right, and so I throw that thing away and I get another one, and it's like, it would be hard to know. I mean, theoretically, if I did it enough, I would believe it, but like it'd be hard to know and it'd be easy to miss. Hmm.

0:55:37 Bryan: Yeah. Yeah. Yeah, yeah, yeah. Yeah yeah. Like you could be stronger than you think, but it just seems so implausible that you ignore it. And this is a serious problem with this type of thing that it's not an explicit file drawer effect, but it is something that can creep in. And a lot of scientists who are careful about this know that this is an issue and try to correct for it. So to give an example from some work tangential to mine, I talked a bit before how I was working on the Euclid Space Telescope to measure masses of distant galaxies for various cosmological parameters. Now, this is one of many studies in this. A previous one, which I'd worked on some people with, had gotten the data before Euclid.

0:56:49 Bryan: It wasn't as big a survey, but they got enough data to calculate results. And they were very cognizant of the possibility that we really want there to be some certain results, but we don't want that belief to influence whether or or not we think the results are wrong. We don't want to say, oh, this is exactly what we want and stop versus if it wasn't what we wanted, we would try to find an error in it. So what they did, one solution to this is, okay, let's deliberately blind ourselves to the actual data. We'll split the group. One group is going to be charged with creating plausible mock data for us to measure, and the other group will create the best method we have to measure data.

0:57:37 Bryan: So then we'll put the measurement and we'll apply it to, we'll start with just doing purely mock data, see if we get out what we put in. And then once we're ready, we think we've solved everything to make the final measurement. We'll look at three sets of mock data and one set of real data, ourselves not knowing which is real, apply it to all of them, and assuming the whole thing looks good and we don't see anything obviously wrong throughout all of the data, we'll unblind ourselves and publish whichever is the real data, whether we like it or not.

0:58:13 Thomas: Interesting.

0:58:13 Bryan: And so you can go through all of that work to make sure that that you aren't just fiddling with the methodology until you get what you expect. But a lot of people and a lot of what you see published doesn't go through that.

0:58:14 Thomas: Yeah yeah It takes a lot of resources and a, you lot have probably incentives to not do that stuff sometimes, you know.

0:58:34 Bryan: Yeah, the People in Charge of this were tenured so they didn't need to worry too much about publishing at any cost. So they had that luxury, but a lot of people working in science don't, unfortunately.

0:58:41 Thomas: Yeah Really interesting. So it turned out to you be less of a physics, physics debunk and more of a like, yeah, I mean, this is the methodology of science and this is how it works. And this is how messy it is. I really liked that because it's something that I think doesn't get covered enough. Well, that's fascinating. Well, I'm sad. The M drive is, still space travel is very impossible. I mean, not impossible, but very difficult.

0:59:08 Bryan: Yeah. Difficult.

0:59:08 Thomas: And, Uh, it's always, I'll never, I'l never forgive the universe for that. There's a lot of things I'm mad at the universe about. That's on the list, you know. As a kid, you watch Star Star Trek, you're like, that's freaking awesome, and they're like, yeah, actually, none of that's possible. Like, gosh dang it, some of it, surely give me some of that. No, pretty much none of it.

0:59:09 Bryan: Yeah.

0:59:25 Bryan: We got cell phones because of Star Trek.

0:59:27 Thomas: We because of Star Trek? Oh, I didn't know that.

0:59:28 Bryan: Yeah, well, we got them when we did because of star Trek, because they came up with them for the show. Somebody was watching the original series, saw the communicators they were using. I like, you know, I think we actually have the technology to do that. And they put it, put in the work and that's where the first cell phones came from was inspired by those on Star Trek. So we did get some things out of it.

0:59:41 Thomas: Interesting. Well, there you go. And we have interracial kisses too, so there you got. Invented by Star Trek? Exactly, I know history. All right, Dr. Dr. Bryan Gillis, that was a ton of fun.

0:59:54 Bryan: Yeah. Yeah. Those were invented by Star Trek? Yes.

1:00:01 Thomas: Look forward to our, Tears the Kingdom, a mega series. It'll be a 10-part series. Look for it in stores near you because I'm going to 100% that game before I talk about it, but it's been a long, slow process because I'll have like a day to play.

1:00:01 Bryan: Thank you.

1:00:14 Bryan: Oh, geez. That is a painful game to 100%. Oh,

1:00:20 Thomas: I'll only a little bit of time here and there to play, but it was so fun. Anyway, um, that was awesome. Anything you want to plug? You got your YouTube channel, uh,

1:00:27 Bryan: I haven't been able to do much on that, but I'll give a plug to another good YouTube channel I've found. It's called Up and Atom, that's A-T-O-M, by a woman named Jade Tan Holmes. She does a great job at explaining a lot of math, science, and computer science things. So you can search for the channel Up and Atom, that's ATOM, and I'll send Thomas a link to put in the show notes.

1:00:32 Thomas: Oh, Oh, Is That any personal connection or just a fan?

1:00:54 Bryan: No, just something I found. I think it's good to share.

1:00:54 Thomas: Oh, That's so nice, yeah, I love it. All right, Dr. Bryan Gillis, thank you so much, excellent.

1:01:00 Bryan: Thank you for having me on, Thomas.