SciShow Tangents - Randomness

Episode Date: January 8, 2025

Is all life predestined, or is existence blown about by the whims of entropy? If we restarted the universe with the same world generating seed a la Minecraft, would it all play out the same??? These are ginormous questions that we do not get into in this episode! Instead, we're examining slightly more manageable examples of randomness in nature, in computing, and even in ourselves. SciShow Tangents is on YouTube! Go to www.youtube.com/scishowtangents to check out this episode with the added bonus of seeing our faces! Head to www.patreon.com/SciShowTangents to find out how you can help support SciShow Tangents, and see all the cool perks you’ll get in return, like bonus episodes and a monthly newsletter! A big thank you to Patreon subscriber Garth Riley for helping to make the show possible!And go to https://store.dftba.com/collections/scishow-tangents to buy some great Tangents merch!Follow us on Twitter @SciShowTangents, where we’ll tweet out topics for upcoming episodes and you can ask the science couch questions! While you're at it, check out the Tangents crew on Twitter: Ceri: @ceriley Sam: @im_sam_schultz Hank: @hankgreen[The Scientific Definition]The Lost Boarding Pass Problemhttps://www.varsity.co.uk/science/23162The Secretary Problemhttps://www2.math.upenn.edu/~ted/210F10/References/Secretary.pdfhttps://slate.com/technology/2014/12/the-secretary-problem-use-this-algorithm-to-determine-exactly-how-many-people-you-should-assess-before-making-a-new-hire-or-choosing-a-life-partner.htmlThe Sleeping Beauty Problemhttps://www.youtube.com/watch?v=cW27QJYNXtUhttps://www.scientificamerican.com/article/why-the-sleeping-beauty-problem-is-keeping-mathematicians-awake/[Trivia Question]Sum of numbers on a D120 vertex where 10 faces meethttps://www.wired.com/2016/05/mathematical-challenge-of-designing-the-worlds-most-complex-120-sided-dice/[Fact Off]Fastest computer worm in history was the random-scanning worm called Slammer (aka the Sapphire Worm) https://cseweb.ucsd.edu/~savage/papers/IEEESP03.pdfhttps://www.science.org/content/article/fastest-worm-everhttps://ieeexplore.ieee.org/abstract/document/1523881Effectiveness of randomly promoting employees in a business instead of promoting by merit[Ask the Science Couch]Humans are (maybe) bad at intuitively understanding or predicting randomnesshttps://cocosci.princeton.edu/tom/papers/hard.pdfhttps://pmc.ncbi.nlm.nih.gov/articles/PMC5215234/https://pmc.ncbi.nlm.nih.gov/articles/PMC5933241/https://www.sciencedirect.com/science/article/abs/pii/S0010027717302895https://zhaolab.psych.ubc.ca/pdfs/Zhao_2014_JEPHPP.pdfPatreon bonus: pseudorandom number generators and randomization seedshttps://www.random.org/randomness/https://csrc.nist.gov/glossary/term/pseudorandom_number_generatorhttps://www.acsac.org/2003/papers/79.pdfhttps://www.fourmilab.ch/hotbits/[Butt One More Thing]Dung beetles that attach themselves to butts to find poop non-randomlyhttps://www.sfzoo.org/dung-beetle/https://pubmed.ncbi.nlm.nih.gov/19169550/

You're listening to a Complexly Podcast. INTRO Hello and welcome to SciShow Tangents, the lightly competitive science knowledge showcase. I'm your host Hank Green and joining me this week as always is science expert and Forbes 30 under 30, though she is over 30 now, Education Luminary, Sari Reilly. Hello. They can never take it away from you. You're gonna die.

30 under 30 Education Luminary. Yes, but hopefully many years after getting the title, fingers crossed. Yeah, hundreds of years from now. And also joining us is our resident everyman, Sam Schultz. What's up? Are you still permanently the new Sam who killed the old Sam? Is that where the lore is at? Yeah, I think that guy is gone.

He's gone. Yeah, he's a ghost. He's just gone somewhere. Yeah, he went ghost protocol. Okay, I don't know what that means, but I have a question for you. That is unrelated. Okay, I don't know what that means. But I have a question for you. That is unrelated. And it is so we're going to launch a new line. I don't know if you've going to look like. And you can do anything as long as it's a really cool looking monkey. No, anything. Really timely question.

Yeah. I mean, I heard it finally getting into this back. And then, yeah, it sort of took some time to get it together. Yeah. Sari, what do you think? Well, since I've got to go first, then the expectations are rock bottom. Non-fungible, I think I'm going to just go straightforward and have it be like a yeast cell or other types of fungi. A bunch of single cell fungi? Yeah, a bunch of single cell fungi. And then it's a non fungible, but it is indeed a fungi. I think that's a good idea.

You get all this good because it is it is you can make a billion of them, you know, just like change some of the organelles around, have the cell wall be a little bit different or whatever they have. They have cell walls. There's an infinite number of them out there. They can mutate. Maybe there could be some kind of game where you like shmoo with another yeast. And then what's that word now that you said?

I think that's the word. I think it is some sort of asexual reproduction. So you can't really shmoo with- I can't shmoo, but one could. Let me see. How yeast cells use shmoos to shmoos. So it is a sexual reproduction. They, I forget, that's the only thing that stuck with me

from some cell biology class and I don't remember what it is even. I honestly thought that was like your grandma's word for kissing. Oh, no, my family did not, I was too repressed. They did not teach me anything about smooching or anything like that. Schmooze were like a comic book character in the like the 40s,

Lil Abner and they were hot. They were big. They were the skivvy. They were hot. No, they were the skivvy toilet of their day. They were a big deal. They were a big deal. They were not attractive to look at. No, they were kind of grisly guys with mustaches. I think the easy choice for me is to just do a variety of upsetting furbies. Oh, that's a great one.

Yeah, so like every furby is upsetting in a different way. Some of them are made of beans, some of them are made of tendons. I think that's really good. They have different color hair, you can swap out their guitars. I didn't mention that they had guitars. Oh, they all have a guitar, a different kind of guitar. Each one has a guitar, but they hold it in different places. Acoustic. Some of the guitar actually, sometimes it's like inside of their body. Yeah, oh

Double necked guitar, maybe yeah, a lot of them play bass actually like a weirdly high percentage of them are bass players Okay, I made a great joke about bass guitars everybody but just I'm getting new internet as soon and I hope that it fixes all of my problems bass guitars, everybody. But just. I'm getting new Internet as soon, and I hope that it fixes all of my problems. Yeah, my NFT will be all the faces that you make when you're frozen, telling us a really funny joke. Just different Hanks going like, oh, and like, oh, all of those fun pictures of Hank. And I think there's one right now. It's happening as we're speaking. And that was a great one. You had a beautiful smile on your pictures of Hank. And I think there's one right now. It's happening as we're speaking.

And that was a great one. Oh my God. You had a beautiful smile on your face, Hank. Screenshot right away. And I put a clip art guitar over him too. So he's always playing a guitar. I figured out what a shmoo is. It is a nodule that they can use to join together

and sexually reproduce. And the process of shmooing takes around two hours. A sex handle. That's a long time. So it's a sex handle, yes. On-demand sex handle. Yeah, I think as on-demand as any sort of sex can be, I don't know what the E4 play is.

You don't know. We have a very sort of on-demand sex system. It doesn't work until it's time, and then it's very worky. That's pretty fortunate too, for the most part. There's also like some frogs grow a special thumb pads for holding on during sex season. They're called nupidal pads, also known as thumb pads.

I'm probably, probably not. Oh, I don't know how to say that. Nutschel. You know, that's one of the words that I've read a lot of times. Can you tell that I don't talk to people? Not sure. I'm a little sad.

So every week here on SciShow Tangents, we get together to try to one-up a maze and delight each other with science facts while also trying to stay on topic. Our panelists are playing for glory, but also for Hank bucks, which I will award as we play. And one of these people is going to be the winner at the end. But first, as always, we must introduce this week's topic with the traditional science poem. This week is from me. An entropic force, it rules unseen.

A cosmic dice, a chance decree. From Adam's spin to Gene's routine, it shapes the way the world can be. No pattern binds, no fate contrives, just chaos breathes and life derives. Through randomness, this arrives. You scatter stars, de-sculpt the clay, the threads unweave, they always fray, the flesh decays and falls away.

Order comes to visit, but order never stays. The topic for the day is randomness. That was cosmic. There's something wizardy about that, I think. Yeah, I was sort of, I did that on a plane. I was in a very sleepy mood. In an odd state. Was one of the lines a chance to cream?

Decree, like the decree of the. Oh, okay. Okay. You never know! a chance to cream? Decree. Oh, okay. Yeah. Yeah. Yeah. Yeah. I mean. You never know. Sure, sure.

You always have a chance. Yeah. Yeah. Yeah. All right. That makes way more sense. Sometimes you're making eclairs. So the topic this week is randomness.

Before we dive in, we're gonna take a short break and then be back week is randomness. Before we dive in, we're going to take a short break and then be back to define randomness. The topic for the day is randomness, which it turns out like it seems like very straightforward, but it turns out is kind of a very important thing for the nature of the universe. And also that like there's this thing about like information and the randomness is kind of the opposite of I don't know exactly. But maybe Sari can tell us what randomness is kind of the opposite of I don't know exactly, but maybe Sari can tell us what randomness is. Yeah, I think I can get around the edges of it maybe. I don't know if it's a lack of information, but it is a lack of a pattern

or predictability in information. Okay, yeah, yeah, yeah. So you have the information as part of randomness, but the idea is that you can't predict the next thing in a sequence of events or symbols or steps or what's going to happen next if it is random. Right. And there are lots of ways to define randomness in mathematics, either whether it's probability

of something happening or statistically something happening, which is like related, all interconnected. Oh, wait, is probability different from statistically? Because that's going to mess up my whole day. I don't know. They're probably subsets. It's probably like rectangle square. One's inside of the other one. Yeah. Yeah.

And all of them kind of encourage people or attract people to think about the idea of like how frequent is an option of something going to happen. And if that thing is like predictable based on the things that came before it, then it is non-random in some way. And if you have no idea what's going to happen next or can't predict that, then that is random. And if somehow you can calculate what is coming up next, then that is non-random. This is very strange to me because I feel like the idea of something being random was a very sort of intuitive, normal, like human thing, kind of, where it's like, ah, so random. But then it also like, it turned out to be sort of very important mathematically and cosmologically

and like the entropy of it all. And so we got stuck with this like fun little word that actually represents this really complicated and important set of ideas. And it started out as a fun little word. It meant it came from old French, randon, randon, which then translated into the same word in English, but we like to change the Ns at the end of the word to Ms. So like seldom, ransom, random, all started with an N

and then we decided to, in English, mush an M on the end. But those words in old French, middle French, et cetera, meant very quickly or like violently or one after another. So something random in old French just meant like, oh, you're so random. Like you're so frantic. You're just happening really fast.

Uh-huh. Yeah. You got too much tea. Yeah. And then we kind of adopted it as, in the 1600s, as far as I can tell, to mean having no definite aim or purpose or haphazard. And then we've shoehorned it into this idea of mathematics of. Yeah. We don't know what's going to happen.

But it doesn't seem fast anymore. Yeah. At some point, I want like a list of words like this, where we just like we needed a word for a big new idea. And we just took one. Right. That's close enough. It happens so much in science. And I like term I I'm kind of, I'm like obsessed with terms being bad and good.

And I don't know, I think the random is pretty good actually. Like it's actually a pretty good, like what it actually, like it feels like what it actually is. Yeah. It's just that the concepts are kind of complicated. And even more than like stealing it, it feels like we came up with it and then we were like,

we looked into it more and we were like, uh-oh, uh-oh, uh-oh. And then the whole universe, it turned out, was like... Yeah, I was like, at first it was like, this is the right word to use. And it was like, oh, I did not realize that this was now going to be an entire subset of physics. And then I think a lot about how like a lot of stuff isn't actually random.

Because I feel like people like to argue about that on the internet. It's hard to actually. The actually only random about how like a lot of stuff isn't actually random. So I feel like people like to argue about that on the internet. It's hard actually. The actually only random thing is like a microphone in the middle of the woods. That's like picking up ambient air and then you have to like get some kind of data from that. That's like some kind of random thing, right? Like security. Yeah, like that, like to get true random numbers, if you want to use it for like cryptography or something.

Right. And protect yourself from getting cracked. You have to listen to the background radiation of the universe and convert that into ones and zeros and have that be your starting point rather than using any kind of math thing because if you're using a math thing, no matter how many like weird operations you add on, there's always a way to backtrack to what the equation is. I do think it is maybe a divide between computers and not computers, because a lot of the random things or the things

that we don't know how to model yet, which is, to our knowledge, random. We've just gotten good at modeling certain kinds of things. But Brownian motion is how particles move microscopically, or genetic mutations we think are random. And maybe they're constrained to certain parts of the genome, but like they happen randomly. Do they happen randomly or do they sort of happen like, like, this is the well now see

where we are, we're arguing whether or not the universe is actually sort of like a thing where if you're in a simulation again, it would happen exactly the same way. I think that the quantum effects appear to actually make it so that it wouldn't all happen exactly the same. But I'm not sure of that. I'm just a guy. There's beams shooting all over the place, hitting all kinds of stuff, right? There's so much stuff bumping. Yeah, I put my hand in that lot. I'd rather someone come away from this episode and be like, Ceri believes that our universe is not predetermined.

That other than Ceri believes our universe is like Minecraft, where if you start with the same big bank seed, then you could regenerate everything exactly as it is. But we'll never know. Well, probably with. I mean, honestly, we probably will. We could.

Like. We might not know. I mean, we're not gonna get to run the simulation again, but we keep finding out new stuff. It's amazing to be 44 where you're like, oh, I've seen like a bunch of new stuff get found out. So you guys have that to look forward to. They better hurry up.

Yeah, haven't seen it yet. I'm getting really close. All right, you guys, You guys have that to look forward to. They better hurry up. I haven't seen it yet. I'm getting really close. All right, you guys, are you ready to play our first game? Yeah. Yes. Well, that's good because otherwise this episode would just end. It's done.

We're going to play a game called the scientific definition. Our lives are shaped by randomness all the time, and it can be a lot to think about. But luckily, mathematicians know how to have fun. They make up probability puzzles. For example, one well-known probability puzzle is called the birthday problem. I love the birthday problem. It asks if you are in a room

full of randomly chosen people, what is the probability that two people will have the same birthday? One of the interesting results from the math is that if you want the probability of finding two people with the same birthday to be more than 50%, all you need is a room full of 23 randomly chosen people. It seems like it would be way more than that,

but there's reasons why it's not. So today, we're gonna play a game of puzzles. It's a version of the scientific definition. It's all these games that mathematicians play with themselves. And I will name a probability puzzle name and you will describe a hypothetical situation the puzzle might be asking about and whoever comes closest to the solution will get the point. And which of you snickered when I said play with themselves? Sarah did.

That was my fault this time. Sam heard cream and I heard this. Let me snicker in peace and let the audience figure it out. All right. The first puzzle is called the lost boarding pass problem. The lost boarding pass problem. What are the odds you could find a boarding pass on the ground and, I don't know, figure out.

Yeah, keep going. You're almost there. I don't know where, but you're almost somewhere. Get on the right train and be on the right train and find the, I don't even know what a boarding pass is for crying out loud. Sam, you must! It's a pass for boarding. Like for a plane, like a ticket for a plane.

Yeah, it's the name that they use for a ticket for a plane. You know what, Sam? You're right. We didn't need a separate word for that. That's a ticket. Yeah. That's a ticket.

Who do they think they are with a whole new word with a bunch of syllables when ticket was right there and excellent? Yeah, it sounds like a business lingo to me. You find a thing on the ground and what are the odds? You can match it up to which, that it would be something that would be leaving in time for you to get on the plane before it left

or something like that. Yeah, something like that. I think the lost boarding pass problem is something to do with seat arrangements. So if you lose your boarding pass and you don't have any proof that you know where you're sitting, you forgot. All you know is your preference.

How likely is it that you will sit in the correct seat and not be asked to move? Oh, that's gotta be it. Sari, that is almost exactly it. That is fantastic. So the actual, it's like one more step away from that. So the lost boarding pass problem, if you have a plane, this is like a simplified boarding pass problem, if you have a plane,

this is like a simplified version of it. If you have a plane with 100 seats and 100 passengers, and one of those passengers gets on the plane, but doesn't have their boarding pass and so it doesn't know where their seat is, they just sit somewhere and then someone asks them to move and the rest of the passengers board the plane, they look at their boarding passes,

they sit in their seat unless someone's already sitting there and if the seat's been taken, they pick a seat at random. So what is the probability that the last passenger will actually get their assigned seat? And the answer is 50%, which is surprising. The interesting thing is that it doesn't actually depend on the number of passengers on the plane. If you want, you can simplify this down to a small plane and write out different circumstances

and you'll find that the passenger to the last passenger to board will have a 50-50 shot of picking their assigned seat or the first passenger's assigned seat. So it's like the last person there. Yeah. Who cares? Why they do this? You know, Sam, I have to tell you something about mathematicians. OK. They care. I don't know why, but they do. They care so much.

They're like, did somebody come up with a weird game? This is very exciting for me. What would really happen is everybody would get in trouble. Right. You know, like, yeah. What I feel like you should just get on last. If you don't have your board. That's a bold. It's a bold move. Is right. Yeah. Have you ever bold, it's a bold move is right.

Have you ever been the cause of a seating cascade? I have. I have too. It was very embarrassing for me. And I had my boarding pass, I just read it wrong. And it was like a five person, cause I didn't notice, but then other people shifted to accommodate me of like, oh, I guess it's really the middle.

It's really that second person's fault. Like it's not your bad. It's that next person who should say, but they should be able to talk to you. You're in my seat. Sometimes people do go shut the fuck up and get out of my face when you say that. But I, yeah, I like to get into fistfights on airplanes

because that way you get to find out a whole new way the airport works. You got to go into fist fights on airplanes because that way you get to find out a whole new way the airport works. You got to go into the secret rooms? Yeah. It would be a really easy way, Hank, to get the most watched TikTok video of all time if you got in a fist fight. Hank Green, get the fist fight on. There's a cheat code for you. Yeah.

All right. Our next round. The puzzle is called the secretary problem. Damn, I heard of this one. I don't know anything about any of these. So you did great, though. I was great. Yeah. Well, now I've set my expectations too high. I did pretty good, too. I think the secretary problem is if.

A. problem is if a sec, so like hypothetically you have a secretary. Off to a really good start. Yep. Okay. I didn't know hazing was part of the game. And a bunch of people call you. So it's like, it's like Ron Swanson and April in Parks and Rec where you have someone scheduling out your calendar and then someone random walks in. And what is the likelihood that this person actually has a meeting with you on the correct day and time

because your secretary's notes are all over the place. I mean, women, right? That's right. I didn't mean to add that, but I don't want to ruin myself with a sound bite. It is sort of asking to be probably sexist in that they called it the secretary. I'm not too confident about mathematicians. I know some very nice mathematicians, but also... Yeah, but this could be from the 50s, you know. I'm digging the hole so deep.

I think it's almost exactly the same as the previous one, but about files instead. If you've misfiled something, what are the odds you're ever going to find it again? Something like that. Now I have to, now I have to do some hard work here because neither of you are close at all. Dang. I think I'm gonna give it to Sam. So let me tell you what it is. So the secretary problem, you're hiring a secretary and you've been presented with a pool of candidates

to interview, you interview them in a random order. And after each interview, you have to decide right then whether you're gonna reject that person or hire them. You don't get to change your mind on this decision but to help, you do get to rank the applicants you've interviewed so far, though you won't know anything yet about the applicants you have yet to see. And everybody you've hired so far is out, is that right?

Everyone's you've hired, yes, correct. All the previous interviews. The secretary problem asks, how do you optimize your ability to get the the best applicant? It's called this is a type of problem called a stopping problem And there are other versions with names like the marriage problem and there's a solution you take the total number of candidates You have interview the first 37% of them and then pick the candidate after that point who is better than those rejected candidates.

So if you so like not like the next one after the 37% but the next one who's better than all 37% of the ones that you've rejected so far. So wow that's that's how you solve the secretary problem which is a problem that you would never encounter in the real world. Right. Unless you're some kind of psycho, evil person. You got me a real jerk. Yeah. And I gave that to Sam and I can't really tell you why. It just seemed more like. I mean, I think Sarah was closer, frankly, but.

Well, no, I like the files are like people, I don't know. Yeah, okay. Well, thank you anyway. All right, our last round, round three. It's called the sleeping beauty problem. I don't know enough about sleeping beauty. Okay, I'll just go to get it out of the way. I don't know enough about the sleeping beauty story.

I don't remember. Rose, prick, something. Rose prick something. Oh, yeah. That's all in a sleep. Getting woken up by a princess kiss, I think. So I think this is just about. The mean number of hours of sleep you get under or it's a probability question. I'll shoot. Oh, shoot. I'll tell you, you're not going to get it, but you could get close.

It's so weird. OK, well, then it is the likelihood of you being woken up by another person after you have slept more than, what's the obscene amount? The most I've ever slept is like probably like 18 hours after you slept more than 18 hours, the likelihood that another person is gonna be the thing that wakes you up as opposed to, I don't know, a loud noise or naturalness. Yes, to your own self.

Okay. Sure, sure, sure, sure. That's okay. What are the odds? Or is that what it is? Whatever. That, you know, you kiss somebody, then they wake up and they see you love at first sight. What's the love at first sight?

Chances. What's the chances of love at first sight. What's the love at first sight? Chances, what's the chances of love at first sight? I'm a complete stranger. After you kiss, I think that probably if you kiss somebody who's sleeping, that the chances of love at first sight are hopefully zero. That's why we have to run the numbers. That's why we have to run the numbers, Hank.

Yeah, but we'll run the numbers and see. Well, you guys, we're gonna figure out who wins this after I tell you about the Sleeping Beauty problem, which I hope that you will approach understanding. So this is a strange experiment that is not, it's just a thought experiment, but imagine that you, the subject, are put to sleep on Sunday,

and then researchers flip a coin, and if they flip a tail, you will be woken up on Monday, the next day, and then put back to sleep on Sunday. And then researchers flip a coin, and if they flip a tail, you will be woken up on Monday, the next day, and then put back to sleep. But you will have no memory of ever having woken up. The same thing will happen on Tuesday. You will be woken up and then put back to sleep and have no memory of that happening. But if they flip heads, you will only be woken up on Monday. So each time you are woken up during the experiment, you will be asked, what is the probability that the researchers flipped ahead?

But you won't be given any information about the day or anything else when you've woken up. So there are two schools of thought for how people think about the probability that researchers flipped ahead. The first is because you have no information when you have woken up. It should be one half because getting heads or tails

is equally likely. But the second approach says that the probability they flipped a head is one third because there are three situations that could describe the situation you have woken up. It's Monday and they flipped tails. It's Tuesday and they flipped tails. Or it's Monday and they flipped heads. It's Tuesday and they flipped tails or it's Monday and they flipped heads and each of those

Events are equally likely. So the probability of a heads is one-third and the basic idea here is to create a kind of paradox Because there isn't a way to know the actual answer to the question and I guess I'll give it to Sari because it was more about sleep They're just being silly. They are. They are. They are. Like a lot of like sometimes the funnest thing in math is to be like, what if we can use math to break math?

That feels like what they're doing here. That is kind of fun. How do we break math with math? I think that's a nice thing. Yeah. Yeah. How do you break this thing that you've poured so much hours into? Yeah.

Feels like speed running a video game, you know? Like you find the right pixel to jump on and then Mario is at the end of the game. Yeah, you glitch it. You glitch the whole idea of math. So, that means that we're heading into the break with Sam at one and Sari at two. Next, it will be time for the Fact Off. [♪ INTRO & MUSIC PLAYING THE FACT OFF THE FACT OFF

THE FACT OFF THE FACT OFF Welcome back, everybody! Now get ready for the Fact Off! Our panelists have brought science facts to present to me in an attempt to blow my mind. And after they've all presented their facts, I will judge them and I will award Hank Bucks any way I see fit. But to decide who goes first, I have a trivia question for you.

The D120 is a 120-sided die made by DiceLab. The die is a Dysdeacus Triacontrahedron, probably. And it has 120 triangular faces and 62 vertices. One of the major challenges in designing the D120 was figuring out how to position the numbers. The designers counted the number of faces around each vertex and decided that those faces should add up to a certain number Based on how many there are for example at any vertex where there were 10 faces Meeting the faces should add up to a certain number, but what is that number? Why does it matter if the numbers on the dice are?

I think that you have to have a way to order the numbers and they're not the kind of people that would just do it at random. I don't know. I think also it's good to have the numbers not be like two numbers that are close to each other be close by each other. Oh, okay. Well, I won't think about it too hard and I'll say that they should add up to 45. Great. There's nothing wrong with there's nothing wrong with that. There's a little something wrong with that.

OK, well. Sarah, do you have a pencil out right now? I am. If you are cheating. No! Am I allowed to do mental math? No, that's fine. I just like you're such a nerd. No, I'm allowed to do that. I just like you're such a nerd. So I was like, I don't want to do a realistic answer.

Are you are you are you did you take your pencil out and you started adding 10 plus 60 or 60 times 10 or something? I started I did. So the math that I did was 120 plus 119, 118, 117, 116. So like the five biggest numbers plus five, four, three, two, one. So the five biggest numbers, five smallest numbers to get 605. The answer is 605. It's not really. Oh, my God. It is. I haven't lost it.

I went to math school. I went to math school. Wow. I've never competed in a math competition in my life. What's happening? I can't believe it. The rush. I can't believe it.

You gotta live on that for the next week. Truly. There's so few moments outside of the school system where you get rewarded for a completely useless thing. And I don't want to go back into academia. I don't want to solve problems for my my living. Yeah, so beautiful. You don't get a feeling like this. A beautiful thing.

That was that was one of the best parts of my day. I didn't even do it. I just feel good for you. What I would have done is I would have I would have just multiplied 660, like divided 120 by two and multiplied that by 10, which would have got me close. But you did something in there I don't even understand. I don't know.

I don't know why that would have worked. Sometimes my brain just does things. This is how I've operated my entire life and how I've become the science communicator I am. We're just like, that makes sense. I was like, surely that's so stupid what you're doing. Oh, that's intense. Well, Sari wins the episode, but do your fact off.

I'm sweaty. So I rely on all kinds of computer programs every day, including to prep and record tangents, of course, but also, of course, not every chunk of code is for good. Malware, as it's called, is created by programmers who want to cause a little bit of a ruckus, like causing disruption to a network of computers or accessing private information. But one type of malware, which is actually called a worm, is designed to spread as fast and as far as possible, infecting computer after computer.

And in my understanding, these worms don't often, but sometimes, change the computer systems themselves, but they cause harm because they take up bandwidth as they're jumping around and blasting past security and therefore slow down any other kinds of data transfer or how the computers are operating because there's this malware running on it. And in January 2003, the fastest computer worm

in recorded history spread throughout the internet. It overwhelmed networks and disabled database servers by taking up way too much bandwidth, leading to canceled flights, Bank of America ATM failures, 911 service failures in Washington state, and lost internet service in countries across the globe. And this worm was called slammer or sapphire. I think slammer is kind of cooler. I think slammer fits. Yeah, that's a good one. That's what I use throughout the rest of the fact.

And it's what's known as a random scanning worm. So this fact is indeed about randomness. Basically, the way that slammer operated is it randomly picked IP addresses, checked whether that they were insecure, and then infected any hosts that were vulnerable. And in three ish minutes, slammer was completing more than 55 million scans of possible hosts per second. And it infected at least 75,000, which is a much smaller number within 30 minutes, but likely much faster than that. We just don't have an estimate of because it happened so fast. We don't know how quickly it infected those machines. Mostly in the US and South Korea, but sort of all over.

And specifically, it targeted computers that were running Microsoft's SQL servers and hadn't been patched for this vulnerability yet. And it basically scanned as fast as a computer could transmit data packets using that random IP address picking. And it made more copies of itself and clogged up the internet bandwidth. But then it slowed down how many computers it could infect as it slowed down the internet as a whole across the- It was a whole other thing. Yeah, exactly. And it ran on what's known as a pseudo random number generation algorithm, which is where there is, you have an initial number that starts the random number generator,

and then there's an equation that calculates the sequence from that. And the programmer, to my understanding, this is where my understanding falls apart, made two choices that make it hard to accurately measure how many IP addresses were infected and which ones because of the way they set up that algorithm. Some IP addresses were likely never scanned and some networks may have been more disrupted than average or less disrupted than average because of the way those numbers were generated, those IP addresses were generated. So it happened really fast and really randomly, but because it was a pseudo random generator,

it wasn't like all IP addresses across the world. It was still a subset of them. So we still have some estimates. And in any case, it is interesting as a case study to think through these kind of random scanning worm disruptions and especially high speed worms like slammer that use this technique and how to prevent the internet going down in such a spectacular and global way, which it hasn't in this manner since 2003. So I assume we've learned something.

How though? How is this not always happening? Like, how's the internet even keeping itself together? The virus got up. That is it. Virus protection software. I think there's like a bunch of network security experts saying, how do you make computers check

what data they're receiving to make sure it is good and won't replicate and then pass itself on in the case of a worm? Or take over software and then why you need people to be also smart and check what they're downloading from the internet. But I like to just download whatever though. Sometimes you can't click. Yeah, sometimes I want an NFT of a little Furby. That's really upsetting. Then like a digital skull pops up on your screen goes. I wish that kind of thing happened

more often. Never does though. They're never very fun. Yeah, they don't want you to know what happened. Yeah. Well, that sucks though. Sam, can you put me in the slammer? I wish I'm working on it. I haven't got a file I'm building. All of my grinds. Yep.

Running a business is the most difficult and noble thing that one can do, at least according to people who run businesses. Managing people is the trickiest part of all of that. Workers must always be striving to improve. Managers must nurture their talents. And it must be assumed that the cream will rise to the top of an organization through hard work and intelligence. Then once all the people are in the best spots, you can really start raking in that cash.

It's the American way. But reality does not always align to the lofty dreams of capitalism. People are complicated, different jobs have different unforeseen challenges, and skills are not always transferable. Thus was born Lawrence J. Peters' Peter Principle, which states that in time every post tends to be occupied by an employee who is Incompetent to carry out its duties Put another way people who are good at their jobs get promoted until they get to a position where they aren't good at their job Anymore that's dragging down the efficiency of the company that they work for

Put another way. Maybe the way that we promote people is kind of fucked up Put another way, maybe the way that we promote people is kind of fucked up. So in 2010, a team of researchers set out to learn more about the Peter Principle and explore different, possibly more effective methods of promotion. And what they discovered was totally rando. The team created a simulated business with 160 employees divided into six levels of responsibility. So basically like the managers on down. Each of those employees was given a competency value and an age. Then they simulated time passing and at each particular time interval, I think it was a year, they'd remove people

who hit retirement age and they would fire the employees that sunk below a certain competency level. Then they'd fill those positions by promoting up from the level below where those vacancies were made. And this is where they started getting weird with it. So they tested four different methods of promotion. They promoted the best worker, they promoted the worst worker, they alternated promoting the best and the worst worker,

and they promoted randomly. So they had a few different methods of assigning new competency with each promotion. So like you'd get a new job and then to reflect how you would do in your new job, they did a couple things. One was the common sense method in which competency carried over. So the same competency you had in the previous position you brought to your new one, and the Peter hypothesis in which a new competency was randomly generated for the promoted person to kind of reflect the idea of the Peter principle that not all job competencies carry over and they also had a version where both of these competency transmission methods were

Used so then they did some math to figure out the overall efficiency of the whole company Based on the new structure each time this happened And here's what they found when they ran the model promoting the best person only improved efficiency consistently in models that use that common sense competency transmission that like one to one in my new job, I'm just as competent. And the Peter hypothesis model, promoting the lowest competency person improved efficiency. So the worst employee getting promoted improved efficiency.

In the version where both of them were used interchangeably, the best method they found was to promote people at random. And since in the real world, we don't really know how a person is going to do in their new job, the researchers proposed that the safest bet is just to randomly promote people in your company. So they published this paper, and business news websites start to write articles about it wondering if like meritocracy is an illusion and HR is a sham.

But the team seemed a little uncomfortable with their results being reported on so definitively because it was as they called it a toy model. So then a year later, they set up a new model with a more robust, like realistic, hierarchical system and a more granular passage of time. They did it every month instead of every year. And they found the exact same thing, promoting at random, consistently generated the highest efficiency in a company. So is the importance that our culture places on management level jobs, promotions, and meritocracy, a fairy tale that C-suite executives tell

themselves to make their own success make more sense. Could basically anyone have any job and it would make no difference in the grand scheme of things? Is the very foundation of the American success story built on sand? I don't know, but I think this is kind of funny research. And also just like takes a huge amount of pressure off me. I could just be like whatever. Whatever. See how that goes. That's what I want to hear. Just give me research where I don't have to

do as much work and that everything, anything that goes wrong wasn't my fault. Yeah, that's what I'm looking for. The universe wouldn't have mattered anyway because of the universe, man. And ultimately most of the animals are nematodes. So're really in charge and they don't even have freaking jobs So I have to choose between the fastest computer woman in history being Slammer and well or random promotions to management are the most efficient and meritocracy is a lie and HR is useless and meritocracy is a lie and HR is useless. They put tuna in the tuna. I think I'm gonna go with Slammer.

That seems awesome. Both are very scary to me in their own little way. Yeah. But also I feel like when you get a trivia question exactly on, that's just another point in your favor. Yeah, that's true. So she has to win no matter what. Congratulations, Sari, on being whatever it is

that you got all the Hank bucks and you're the king of the episode. matter what. Congratulations, Sari, on being whatever it is that we did. You got all the Hank bucks and you're the king of the episode. Wow. Congrats. Yeah, not just Halloween. I'm the king of the whole episode, baby. And now it's time to ask the science couch where we ask listener questions to our couch of finely honed scientific minds. Avin DeSora and Whiswine5898 on YouTube asked,

why is there an apparent abundance of society that associates random with evenly spaced? Can you talk a little bit about the double-edged sword of our fantastic pattern recognition when it comes to how we perceive and react to truly random events that just don't feel random to us at all, AKA, why are humans so bad at detecting randomness? Yeah.

So I think that what's being discussed here is that if you have a blank square of space, and then you say, OK, put random dots on this, people will put dots that are equally spaced out between each other. They will not clump them together, which is actually likely to happen if you get a random program.

But it looks so messy. Like a remit dot generator to actually place them. And then also like we have this problem where, like statistics teachers, this is a great statistics teacher trick, where you have people write down strings of heads and tails as if they wanted to make them look random. And then you have another piece of the class

actually flip heads and tails. And then the statistics teacher can usually pick out between who actually flipped coins and who pretended to be random because the random people won't do long strings of heads or long strings of tails, but that will happen with random flips. Okay. And why are people bad at this? Because we don't understand what's going on. Like we don't. We're not random number generators. We're

people is I that's how that's how I feel. What do you think, Sari? Yeah, I mean, that's basically it. Psychologists study this a lot of how we understand randomness and what knowledge about the universe or expectations about the universe carry over into what we think of as random. One of the misconceptions that I think was popularized in psychological literature is called the gambler's fallacy, which is what you were describing, Hank, with the coin flips, which is basically this idea that like an increasingly long sequence of heads on a coin flip, like if you flip a coin, 50% chance of heads, 50% chance of tails. The longer the sequence of heads is, like you got to be seeing a tails soon.

When really in any individual flip, it's a 50-50 chance. And seeing a long string of heads doesn't mean that it's more likely you'll get a tails. It's weird, because if you actually like group them statistically and say, you know, this is this is part of like, like all of these tales are in a probability experiment, you are more likely to see it approach 5050. But if you just look at the next one, it's not how would it ever be affected by what has happened before? Yeah, and it's tricky because, like you were saying, it's like, if you look at the situation

now of what is the one roulette spin, it you know, the probabilities. And if you look at the situation now of what is the one roulette spin, you know the probabilities. And if you look at the situation of like an infinite number of roulette spins, you know the probabilities. But it's those like in between sequences that really trip us up. And there are researchers, what is interesting, especially these two psychologists Warren and Han, I saw their names a lot on these papers that are kind of arguing that humans are dumb, but randomness is hard. And so we may not be as dumb as we say we are like it is if you teach people how to identify

what a random sequence of coin flips look like. So in the example that you talked about about the statistics teacher saying, write down what you think is a sequence, another half of the class saying flipping the coins, if you then show the entire class a coin flip sequence, then they will get better at identifying what randomness is. And so saying that essentially, this is a very complex problem because if you bring people into it blind, then they might try and create some pattern or order because part of how we survive in this world is creating pattern or order. But once you teach someone what mathematical principles underlie randomness, even if they're counterintuitive, then they'll get better at emulating that or picking it out from a sequence like that.

So are a gambler's fallacy really a fallacy? Yes, because it shows up over and over again. And it's how casinos make money. And you can psychologically train yourself out of it as well by like thinking about it in the way that we're thinking and talking about it. I feel like the dots thing also has an aesthetic element to it though, where it just feels nasty to draw them too close together. You know? We can't help ourselves but make it look nice.

Yeah, that one feels weirder to me, where it's like, I do feel weird about putting them close together. And it does not feel like... It feels like there's something, I do feel weird about putting them close together. And it is it does not feel like it feels like there's something like some violation I'm doing when I put two dots close together and trying to assign random dots. I couldn't find any papers about that. I didn't look super, super hard, but I assume that's probably what plays into it, too, of if you were like looking out at a field full of flowers,

if the flowers were more scattered, then you'd think, oh, that looks more random than a cluster planted together. Because you can reason and say, well, maybe the flower sprinkled its seeds there, whatever. And so that is something that they talk about in these papers as well, is you can't inspect a snapshot of a situation and describe it as random or non-random like you can't walk out to a field of Flowers and be like oh, that's so random or oh this obviously happened non-randomly randomness is a quality of a system over time So if you watch a lot of different events or if you like a bunch of different coin flips a bunch of generations of flowers Can you predict where the next one will be right? But if you walk into any sort of situation,

it is kind of a misleading question to say, is this random? Is it randomly generated is a question. Right, you can't ascribe randomness to a static system. Yeah, exactly, because a random number generator might produce the same number thousands of times in a row, but the system as a whole could go on infinitely long.

And so that snapshot would look non-random or something non-random could generate what looks like a random sequence, but if you kept going, it would indeed be. But it's more than that. It's literally that randomness is about things happening, not the way things are. That's a good way of putting it, yeah.

Alright, well, we have another question, but this one is for our listeners on Patreon. We're answering a bonus Science Couch question. Sam, what is it? Shelley on Patreon asked, if I, as a regular person, want to generate a random number, what's the best way to do that? For example, is there a publicly available random number generator online that's more random than a dice roll? If you want to hear the answer to that question, as well as enjoy all new episodes totally ad-free, head over to our Patreon.

That's patreon.com slash SciShow Tangents at our $8 a month tier. You get new episodes ad- free and extended shenanigans as we answer a bonus Science Couch question every episode. Our patrons are the best. We're so grateful for their support of the show. If you wanna ask the Science Couch your question, follow us on Twitter at SciShow Tangents or check out our YouTube community tab

where we will send out topics for upcoming episodes every week, or you can join the SciShow Tangents Patreon. Thank you to Violet on Patreon, at Frostbite on YouTube, and everybody else who asked us your question for this episode. If you like this show and you want to help us out, it's super easy to do that. First, you can go to patreon.com slash SciShow Tangents to become a patron. Get all that stuff I just talked about.

Shout out to patron, Les Aker for their support. Second, you can leave us a review. That's very helpful. I love reading them and it helps us know what you like about the show. And finally, if you want to show your love for SciShow Tangents, just tell people about us. Thank you for joining us. I have been Hank Green. I've been Sari Reilly. And I've been Sam Schultz. SciShow Tangents is created by all of us and produced by Jess Stempert. Our associate producer

is Eve Schmidt. Our editor is Seth Glicksman. Our social media organizer is Julia Buzz-Bazio. Our editorial assistant is Devoki Chakravarti. Our sound design is by Joseph Toon-Amedish. Our executive producers are Nicole Swingy and me, Hank Green. And we couldn't make any of this without our patrons on Patreon.

Thank you, and remember, the mind is not a vessel to be filled, but a fire to be lighted. But one more thing. Dung beetles are known for rolling around balls of poop, which they also eat and feed to their young. And many species find lumps of poop sort of randomly, using scent to track where an animal did its business. But a few kinds of dung beetles, like the Canthon quadrigutatus, don't leave it up to chance. They latch onto the anal region of certain monkey species so they can get the dung straight from the source, falling with the poop to the ground as the monkeys excrete it. I don't like this, but boy, I guess poop is in high demand.

I love it. They're so smart. They go right to the source. I think a butt's probably one of the least random things in the whole universe. Right, because it was one of the first things that life did. It was like, we need to get stuff out of here. We got to get it in.

We got to get it out. So we got to get those two things settled first. I guess cells also kind of can, the single-celled organisms can kind of have a butt. They could have like a specialized area that for getting stuff out. The universal constant, the butt. You're not getting surprised by the fact that there's a butt.

Yeah, I mean, some single- celled organisms don't have butts. They have they have like vacuoles that like pop, but they it doesn't like happen in a specific area of the cell. Well, so take that. If it looks like a butt, smells like a butt. It's a butt. If it looks like a button smells like a butt, it is probably a butt. Yeah, that is true. Fair chance. Yeah.