[RADIOLAB INTRO]

LATIF NASSER: Lulu?

LULU MILLER: Hello?

LATIF: Hey. Should we start?

LULU: Let's do it.

LATIF: Okay. All right. So today, we're gonna start with a guy, a very sweet, very tall guy, named Sunil Nakrani.

SUNIL NAKRANI: Yeah, hi. I'm Sunil Nakrani.

LATIF: Where did you grow up, and were you just a computer kid, like, you just loved computers? Or how did that—how did this all start?

SUNIL NAKRANI: No. So do you want to start from there?

LATIF: Yeah. I mean, a little bit.

SUNIL NAKRANI: Yeah, so, I—I was actually born in Kenya.

LATIF: But he grew up between India and the UK.

SUNIL NAKRANI: Right.

LATIF: He studies hard.

SUNIL NAKRANI: Bachelor's and a master's degree in electrical engineering.

LATIF: And in 1989, he lands a job at IBM just as the world is encountering this new thing called the internet.

[ARCHIVE CLIP: Welcome.]

[ARCHIVE CLIP: What about this internet thing? Do you—do you know anything about that?]

[ARCHIVE CLIP: There's loads of useful information in here. You can get news, recipes.]

LATIF: So Sunil is like ...

SUNIL NAKRANI: Okay, why not go and study communication engineering?

LATIF: He goes to Oxford to get his PhD. And one day, near the beginning of the semester, while he's on one of the desktop computers in the computer lab, his whole department, including him, gets an email from one of his professors, who's an American guy, and the email just says ...

SUNIL NAKRANI: "Hey guys, America is under attack. Come down to the common room and then watch," right?

[NEWS CLIP: It came out of the clear blue sky on a mild fall morning in Manhattan.]

LULU: Oh, it's 2001.

LATIF: Yeah, it's 2001. Sunil's standing there in horror.

[NEWS CLIP: Smoke appeared everywhere, as though a mist had suddenly settled on ...]

LATIF: And I mean, he has a million questions, like who did this? Why? Is America at war? So he goes online.

SUNIL NAKRANI: To get any sort of new information.

LATIF: But he just can't.

SUNIL NAKRANI: A lot of the websites were just not responding.

LATIF: Some websites had just crashed. Others would, like, just keep loading but then never fully load.

SUNIL NAKRANI: At one point they—you know, it was such an overwhelming demand for, you know, news, that they resorted to serving only text.

LATIF: Plain text.

SUNIL NAKRANI: Because there were so many people trying to get to those websites.

LULU: Because there was just so many people trying to get access.

LATIF: So much hunger of people trying to figure out, like, what the hell is going on here?

SUNIL NAKRANI: Correct.

LATIF: And he's like, why is it that at the very moment when I and the world want to access something the most, that's when I can't access it? And this used to happen all the time.

[ARCHIVE CLIP: As millions of people flooded the system ...]

[ARCHIVE CLIP, Ellen Degeneres: Last week, a picture of a dress was posted on Tumblr.]

LATIF: Some website ...

[ARCHIVE CLIP: Healthcare.gov]

LATIF: ... or picture ...

[ARCHIVE CLIP, Ellen Degeneres: The dress that broke the internet]

LATIF: ... or video ...

[ARCHIVE CLIP: Chocolate rain]

LATIF: ... would suddenly become popular.

[ARCHIVE CLIP: Within just 15 minutes of Pokemon Go launching ...]

LATIF: And you get this crash of people.

[ARCHIVE CLIP: ... traffic had already passed initial predictions.]

SUNIL NAKRANI: This flash flood.

LATIF: And it breaks the internet. And this situation ...

SUNIL NAKRANI: A demand that is beyond what they planned for.

LATIF: ... Sunil became kind of obsessed with it.

SUNIL NAKRANI: And so I ended up looking at websites, how they architect some of these infrastructures.

LATIF: And he's like, "There's gotta be a way to fix this."

SUNIL NAKRANI: Yeah. Like, what can I apply to solve that problem?

LATIF: Meanwhile, Sunil's wife is working in Atlanta.

SUNIL NAKRANI: We were doing back and forth in between Oxford and Atlanta.

LATIF: And at a certain point, he's having kind of a hard time with this internet problem.

SUNIL NAKRANI: Because how do you design a system for the future when you don't know what the future is?

LATIF: And one day he just thinks to himself, Georgia Tech is down the street. Maybe someone there could help me.

SUNIL NAKRANI: Point me to some direction that I could take.

LATIF: So he just emails some people in the engineering school.

SUNIL NAKRANI: I said, "Oh, I'm a PhD student looking to, you know, discuss some ideas. Can I come and see you?"

LATIF: Yeah.

SUNIL NAKRANI: And that was basically it. I didn't describe the problem in my email. I didn't really expect anything—anything substantial, but ...

LATIF: Within 30 minutes, he gets a response from a guy named Craig Tovey.

SUNIL NAKRANI: Saying, "Come by my office."

CRAIG TOVEY: And then a very tall guy knocks on my door and says, "Oh, I'm looking for Craig."

LATIF: This, of course, is Craig Tovey.

CRAIG TOVEY: Hi. Good to meet you.

LATIF: Craig's a systems engineer.

CRAIG TOVEY: Operations research.

LATIF: His job is to make huge operations run smoothly—factories, shipping routes, that kind of thing.

SUNIL NAKRANI: So anyway, I walked into his office, we sat down, we started talking, and I started describing the problem.

LATIF: So Sunil is like, "Look, I'm trying to fix the internet. I'm trying to stop it from breaking every time one of these internet flash floods happens."

SUNIL NAKRANI: He didn't say a lot, actually, at the beginning. He just kept listening. And then 25, 30 minutes later, you know, suddenly, Craig Tovey said, "Oh, oh, oh. Wait."

LATIF: Craig stands up ...

SUNIL NAKRANI: And then went back to his desk and pulled out a paper.

LATIF: And he plops it down in front of Sunil. And it's called, "The Pattern and Effectiveness of Forager Allocation Among Flower Patches by Honeybee Colonies."

SUNIL NAKRANI: So at the time, I thought, "Oh, why are we talking about honeybees?"

LULU: Yeah, why are we talking about honeybees?

LATIF: Well, Craig had this hunch that bees had something to teach Sunil—and all of us, really—because it turns out bees are sort of a model of how to thrive in an uncertain world.

LULU: Hmm. Okay!

LATIF: So that's the story I want to talk about today, the story of how a multibillion-dollar tech industry used a trick they learned from millions of years of honeybee evolution to build the internet as we know it.

LULU: Okay. Giddy-up!

LATIF: Okay, so Craig and his bee study, it really began with his collaborator.

THOMAS SEELEY: I'd like you to refer to me as Tom.

LATIF: Dr. Tom Seeley.

THOMAS SEELEY: You could introduce me as "Doctor" or "Professor," but let's quickly switch to Tom.

LATIF: He's a retired professor of neurobiology at Cornell, and one of the world's top experts on honeybees.

THOMAS SEELEY: And I did pretty much pioneer the study of how honeybees live in the wild on their own.

LATIF: And he started doing that work more than 60 years ago, when he was a kid.

THOMAS SEELEY: Yeah, a swarm of bees moved into a large black walnut not far from my parents' house.

LATIF: And Tom says he would watch those bees flying in and out of this knothole in the tree.

THOMAS SEELEY: Like shooting stars.

LATIF: Pew!

THOMAS SEELEY: Zooming off in all directions.

LATIF: And he'd see them and he'd be like, "Where are they—where are they going?" Like, he kind of knew where they were going.

THOMAS SEELEY: There were fields of a dairy farm up on the hillside

LATIF: With lots of flowers full of nectar and pollen for them to eat.

THOMAS SEELEY: And they were probably going up to that.

LATIF: Like, I know where they're going, but how do they know where they're going? Like, what—what is going on here?

LULU: It—are they—they're not—are they—aren't they just, like, flying, and if they see some pollen or nectar they—they bring it home?

LATIF: No. No, it's way, way harder and more complicated than that.

LULU: Why? How so?

LATIF: There's many different reasons. And this is all stuff Tom would eventually learn in college and grad school. Okay, so, like, first, flower patches are not evenly or clearly distributed. They're not just everywhere.

LULU: Okay, fair.

LATIF: They're also not all blooming at the same time.

LULU: Okay.

LATIF: So there might be flowers that are blooming at certain times of the year, or even certain times of the day. Also, you need to find the flowers in bloom that still have nectar.

LULU: Mmm!

LATIF: Because you're—you're in competition with all these other pollinators.

LULU: Right.

LATIF: And you have to do all your food gathering before winter comes, because, you know, then all the flowers go away.

LULU: Huh!

LATIF: And—and you have to get something like—guess how many flowers—if you're a bee, guess how many flowers you have to hit to get, like, a little bottle of honey that you would find in a grocery store?

LULU: Oh. Oh, okay, like one of those bear little squeezy bottles?

LATIF: Yeah, one of those little bear squeezy bottles.

LULU: How many flowers went into that? That's such a nice question! I don't know. Like, ten thousand?

LATIF: Two million.

LULU: Two million?

LATIF: Two million! And the hive needs, like, the equivalent of two hundred squeezy bottles of honey to survive through the winter.

LULU: Wow!

LATIF: All this is an efficiency game. Like, you only have so much time when the sun is shining, when these flowers are blooming that you can—you can hit them.

LULU: Okay.

LATIF: Okay, so they have to do all of this. It's really hard, and they have, like, nobody—they have no boss. Nobody is in charge.

LULU: Queen bee isn't in charge?

LATIF: So that's the thing. So—so back in Aristotle's time, that was the idea.

THOMAS SEELEY: For many, you know, centuries, people thought the queen was a ruler. But no, that's not correct.

LATIF: The queen's whole job is to make babies.

LULU: Yeah.

THOMAS SEELEY: She's not telling anybody else what to do.

LATIF: Nobody is!

CRAIG TOVEY: A colony is intelligent in some way that the individual bees are not.

LATIF: And when Craig found out about this ...

CRAIG TOVEY: It started with NPR.

LATIF: ... from a radio story ...

CRAIG TOVEY: Tom Seeley talking about honeybees.

LATIF: ... he was just like, "What?"

CRAIG TOVEY: I was just in awe.

LATIF: Because at the time, Craig was working on robots. He was trying to get a group of them to build a car.

CRAIG TOVEY: It was an unsolved problem.

LATIF: And he was kind of stumped.

CRAIG TOVEY: I had no idea how to get these robots to work together to be as a group, like, more intelligent than the individual stupid robots. But here were these honeybees.

LATIF: Doing exactly that.

CRAIG TOVEY: And I'm thinking wow! Let's understand how the bees are doing that, and then we can copy that and apply it to robots.

LATIF: At first, Craig thought this would be pretty straightforward.

CRAIG TOVEY: I thought that the biologists had figured out everything.

LATIF: But turns out they hadn't.

CRAIG TOVEY: That's right.

THOMAS SEELEY: If somebody asked me how much we know about how a honeybee colony works, I'd say, 50 percent.

LATIF: And, you know, 50 percent's not nothing. They knew, for example, about the waggle dance, which you might have heard of.

LULU: Yes, I've heard that bees will waggle. And—and what is the waggle, though?

LATIF: I mean, it's quite sophisticated. It's a dance that bees do to sort of show other bees where they just came from.

LULU: Oh, so it's like a—it's a choreographed map. It's like a map dance.

LATIF: It's like a map dance. And this is like a piece of the puzzle.

CRAIG TOVEY: Oh yeah. Well, it first won the Nobel Prize, but what the biologists had not figured out ...

LATIF: Is the bigger picture. What Tom would call ...

THOMAS SEELEY: The wisdom of the hive.

LATIF: And that is what Tom was working on when Craig called him.

CRAIG TOVEY: And Tom said, "Well, you know, come help me run these experiments."

LATIF: And the two of them would end up doing an experiment together that would give us a little peek into that wisdom of the hive, and would eventually become the paper that many years later Craig would slap down in front of Sunil in his office.

CRAIG TOVEY: It's still vivid in my memory. In fact, this is one of the best weeks of my life.

LATIF: It's July, 1991.

CRAIG TOVEY: Okay, now this is in the Adirondacks, upstate New York.

LATIF: Like, way upstate.

CRAIG TOVEY: Close to the Canadian border.

LATIF: Craig had to drive on these dark back roads.

CRAIG TOVEY: For an hour and a half, two hours.

LATIF: Looking for a little sign for this research station.

CRAIG TOVEY: Cranberry Biological Station.

LATIF: Cranberry Biological Station.

CRAIG TOVEY: Yeah. Cranberry Lake. I'm sorry, Cranberry Lake.

LATIF: Cranberry Lake.

CRAIG TOVEY: Yeah.

THOMAS SEELEY: I'm always amazed that he pulled it off, that he'd found this biological station up in the middle of the woods, deep at night.

LATIF: And the reason Tom studies bees at this place way out in the middle of nowhere is because ...

CRAIG TOVEY: There are no naturally occurring honeybee colonies.

LULU: What? Wait, why would you study bees in a place where there are no bees?

LATIF: Right, because the absence of bees lets you set up kind of controlled experiments without any other bees messing it up.

THOMAS SEELEY: That's right.

CRAIG TOVEY: It's a BYOB—bring your own bees ...

LATIF: [laughs]

CRAIG TOVEY: ... set up. So Tom brought an experi—you know, a little colony ...

LATIF: [laughs] BYOB.

CRAIG TOVEY: ... of about 4,000 bees.

LATIF: Okay.

CRAIG TOVEY: And, you know, after breakfast, we go out to where the hive is that Tom has set up.

LATIF: A wooden box.

CRAIG TOVEY: Two feet high.

LATIF: Kind of like a beekeeper would have. But with one big difference.

CRAIG TOVEY: The front is all glass.

LATIF: Basically a transparent hive.

CRAIG TOVEY: Yeah.

LATIF: So you can watch the bees inside.

CRAIG TOVEY: And each bee had to be individually recognizable.

LATIF: How do you do that?

CRAIG TOVEY: So you put a little two-digit number on their thorax, and a little daub of paint on their abdomen.

THOMAS SEELEY: Until they're painted they're anonymous members of a colony. But once they're painted, you get to know them. Some are nervous Nellies, others come ...

LATIF: Really?

THOMAS SEELEY: Yeah.

LATIF: They have little personalities?

THOMAS SEELEY: Oh, definitely. Definitely. Some get up in the morning and get things going, others hang out 'til eleven o'clock. [laughs]

LATIF: [laughs]

LATIF: Anyway, when the sun comes out on that first morning, a big wave of painted bees whooshes out.

LULU: Except for those slackers who are sleeping in.

LATIF: [laughs] Fine, except for the lazy ones who are sleeping in. But pretty quickly, the ones who have gone to work, they start flying out to these feeders, these fake flower patches that Tom has put out for them.

CRAIG TOVEY: Imagine a glass Petri dish where we have high sugar content water.

LATIF: It's a little buffet for them.

CRAIG TOVEY: Yeah, that's right.

LATIF: And so the way the experiment works is that there are two fake flower patches.

LULU: Mm-hmm?

LATIF: But they're not equal. One is better than the other. And what they want to see is like, how do the bees suss that out, and how do they make it kind of a collective decision to send more bees to one rather than the other?

LULU: Whoa! Okay, so this is—this is it. This is where you see maybe it's not just random. Like, they might get to glimpse hives' intelligence.

LATIF: Yeah, because remember, this is the whole thing, right? Like, winter's coming, they're on a clock, and let's see how they prioritize how they make that decision.

LULU: Huh! Wait, and then how—how do they make one fake flower patch better than the other?

LATIF: Well, there's lots of ways.

CRAIG TOVEY: You know, we can put 1.5 bowl of sugar.

LATIF: One feeder might have sweeter sugar water than the other.

CRAIG TOVEY: We can change it from one and a half to two and see how the behavior of the bees changes.

LATIF: Or one might be bigger and one smaller.

CRAIG TOVEY: If you had a lot of bees coming to the same feeder, some of them would have to wait.

LATIF: But Craig says, for the purposes of explaining this, just imagine this situation.

CRAIG TOVEY: One feeder which is close.

LATIF: About five minutes from the hive. And another one that's ...

CRAIG TOVEY: Ten minutes to fly out to the flower patch, fill up her stomach with nectar and fly back.

LATIF: ... a little further away.

LULU: Okay.

LATIF: So ...

CRAIG TOVEY: Tom is at the hive recording each bee as it leaves the hive and each bee as it comes back.

LATIF: Also, are you a—have you ever—I mean, probably not but, like, have you ever been stung by a bee? Are you afraid of bees at all?

CRAIG TOVEY: I am somewhat allergic to bees.

LATIF: Oh, I did not know that!

CRAIG TOVEY: But—so I am scared of them, but ordinarily, honeybees are not all that aggressive.

LATIF: Anyway, right away, the guys notice ...

CRAIG TOVEY: There are a lot of bees going to this five-minute feeder.

LATIF: The feeder close to the hive is blowing up. And when they follow those bees back to the hive ...

CRAIG TOVEY: To my eyes it just looks like chaos, but if you're Tom Seeley, you'll spot ...

THOMAS SEELEY: The waggle dance.

CRAIG TOVEY: ... a waggle dance.

LATIF: The bee effectively saying ...

[VOICEOVER: Hey, go that way.]

LATIF: And pretty soon they see another bee from that close-by feeder come in and dance.

[VOICEOVER: Hey, go that way.]

LATIF: And another.

[VOICEOVER: Hey, go that way.]

LATIF: And another.

[VOICEOVER: Hey, go that way.]

LATIF: And each time those bees dance, they're bringing more bees back to the five-minute patch.

[VOICEOVER: Hey, go that way.]

LATIF: And as all that's happening, every now and then ...

[VOICEOVER: Hey, go this way.]

LATIF: ... a bee comes in and dances for the ten-minute patch.

[VOICEOVER: Hey, go this way.]

CRAIG TOVEY: But ...

[VOICEOVER: Go that way.]

[VOICEOVER: Go that way.]

[VOICEOVER: Hey, go that way.]

CRAIG TOVEY: ... there'll be more bees going to that five-minute patch, and there'll be fewer bees going to the ten-minute patch, just because the bee's coming back twice as often.

LULU: Oh! Okay, so the closer flower patch gets more bees saying, "Go that way. That way. That way." And so more bees go that way instead of this way.

LATIF: Right.

CRAIG TOVEY: So now here comes one of the beautiful parts of it.

LATIF: And this is where it gets really interesting.

CRAIG TOVEY: If there are a lot of bees going to this five-minute patch ...

LATIF: Eventually, Craig says ...

CRAIG TOVEY: ... there will be more and more depleted flowers.

LATIF: This patch, it starts to run out of nectar.

CRAIG TOVEY: That means that a honeybee, she's gonna take longer to fill her stomach.

LATIF: Right.

CRAIG TOVEY: So a five-minute patch, if it's crowded, is no longer a five-minute patch.

LATIF: Right.

CRAIG TOVEY: It might become a seven-minute patch.

LATIF: And as the patch gets more and more picked over ...

CRAIG TOVEY: Now it's an eight-minute patch.

LATIF: And then ...

CRAIG TOVEY: A ten-minute patch.

LATIF: And at that point, it's taking the bees the same amount of time to go to the close-by patch as it is to go to the one that's further away. And because of the dance thing, the hive is sort of evening out the number of bees it's sending to each patch.

[VOICEOVER: Go that way.]

[VOICEOVER: Go this way.]

[VOICEOVER: Go that way.]

[VOICEOVER: Go this way.]

CRAIG TOVEY: Then you're in equilibrium. Even though the bees don't have stopwatches, they equalize the round trip time.

LATIF: The hive has taken into account distance and crowdedness, and figured out the way to get the most nectar in the least amount of time. And in the real world ...

CRAIG TOVEY: The bees may very well be going to five different patches.

LATIF: Or a dozen.

CRAIG TOVEY: Not just two.

LATIF: And dealing with actual nature, which means taking into account different types of flowers, and weather and predators. But no matter how many other variables the bees have to deal with ...

CRAIG TOVEY: The allocation ...

LATIF: Of bees.

CRAIG TOVEY: ... amongst the flower patches ...

LATIF: ... is astonishingly efficient.

LATIF: It looks like there's an air traffic controller, or like the hive is thinking.

THOMAS SEELEY: Yeah, that's right. I think, though, when we use the words like "thinking" we're thinking in human terms. But you say "thinking," it's just a matter of taking in information and processing it to make decisions, then I think that definition of thinking applies.

LATIF: And the way that they were processing this incredibly complex set of information was by following one simple rule.

CRAIG TOVEY: If one flower patch has a smaller round-trip time than the others ...

LATIF: Send more bees there. Like, whichever patch bees are coming back most quickly from ...

[VOICEOVER: Go that way.]

[VOICEOVER: Go that way.]

[VOICEOVER: Go that way.]

LATIF: ... that's where you send more bees. And that's it.

CRAIG TOVEY: It's gorgeous, isn't it? A friend of mine once said it's very zen.

LATIF: But it's also weirdly—it's very bottom line. Like, it's like, I don't care where you're going, I don't care what you're—like, it's like what really matters is when you show up with the goods.

CRAIG TOVEY: Mm-hmm.

LATIF: Like, show up with the goods, and then we'll talk.

CRAIG TOVEY: [laughs] And then we'll negotiate.

LATIF: So this system Craig and Tom observed at Cranberry Lake, Craig wrote it up as math. He called it ...

CRAIG TOVEY: The honeybee algorithm.

LATIF: What does it look like on paper?

CRAIG TOVEY: Equations I wouldn't want to show a fourth grader? F sub n of x of n, F sub n of x sub n divided by x of n is equal to F sub m ...

LATIF: And when Craig tried to use this algorithm on his car-building robot problem ...

CRAIG TOVEY: It didn't apply.

LATIF: It was completely unhelpful.

CRAIG TOVEY: It was such a different problem.

LATIF: Which is why it was so exciting to him when Sunil walked into his office more than a decade later with this other problem, the internet problem which, to Craig anyway, looked basically the same as the one the bees were facing.

CRAIG TOVEY: And so after 15 minutes I said to Sunil, "Let's imitate the bees."

LATIF: How could imitating the bees help Sunil stop the internet from breaking?

LULU: I am desperate to finally understand this, but first we need to take a quick break.

LATIF: All right. We'll be right back.

LULU: Doot doot doot doot doot doo, ha! Lulu.

LATIF: Latif.

LULU: Radiolab.

LATIF: Back with ...

LULU: Bees.

LATIF: Bzzz. Yeah.

LULU: And you were going to tell us how they're, like, in our internet, or something?

LATIF: Yeah, basically. Pretty much. But to do that I needed an example, so I found something in my own life which I apologize in advance for how annoying this is gonna be.

LULU: Okay.

LATIF: Do you remember the hamster dance?

LULU: [laughs] Uh, no, I don't think I do. Is this like a website?

LATIF: Yeah. Yeah, yeah, yeah, yeah. Okay, I'm kind of surprised you don't remember it. It was a very early web kind of thing. It was—but all it was, it was just a song with a webpage, and the webpage was just a bunch of gifs of cartoon hamsters dancing.

LULU: [laughs]

LATIF: It's the most annoying song that you will ever ...

LULU: Can you still sing it?

LATIF: I can. Would you like to hear it?

LULU: Yes, please.

LATIF: It goes, dee da dee da dee da doo. Dee da dee dee dee. Deedly deedly deedly—anyways, something like that.

[Hamsterdance music]

LATIF: It's very, very annoying.

LULU: Okay.

LATIF: And I remember when I was, like, in high school, I thought it was so funny. And one day I tried to show it to my cousin, because I was like, "Oh, this is so funny! You have to see this!" And I remember it, like, taking a really long time to load. And I was like, "Oh, because there's so many people who are ..."

LULU: [laughs] Who need to get their eyes on this.

LATIF: Right. And so what was happening behind the scenes was—well, okay, basic internet 101.

LULU: Yeah?

LATIF: Every website on the internet, like, for example, what was at the time Hamsterdance.com exists geographically somewhere.

LULU: Like, in the cloud, or ...?

LATIF: No. No, like a single, earthbound computer. The owner of the website paid for it to live there, and that computer, which is called a server, is not in a house or an office. It is in a special building dedicated to servers, a server farm, right? Or a data center.

LULU: Okay, so it's at a data center in, like, Ohio.

LATIF: Yeah, wherever it is.

LULU: Okay.

LATIF: These server farms are all over the world, and connected together they more or less make up the internet.

LULU: Oh!

LATIF: And so ...

SUNIL NAKRANI: When the request comes in ...

LATIF: ... when, you know, young me types in "Hamsterdance.com" and presses enter, that server ...

SUNIL NAKRANI: The actual infrastructure that's holding it ...

LATIF: ... who was just sitting around in Utah or wherever on the server farm gets a little notification.

[VOICEOVER: Oh, someone wants to see that Hamsterdance.com website. All right, I'm not too busy with anything else.]

LATIF: So it does a little computation.

[VOICEOVER: Go ahead and get that for you.]

SUNIL NAKRANI: Some execution of some java code or something.

[VOICEOVER: Let me see here. I think it's around here somewhere.]

SUNIL NAKRANI: And then push that content back to you via internet.

[VOICEOVER: Here you go!]

LATIF: Where it pops up on my family's desktop computer back in 1999.

[Hamsterdance music]

LATIF: And then I can enjoy it. Then the server just goes back to hanging out on the server farm. But then ...

[VOICEOVER: Oh, another person seems to want to see that website. All right, let me just get that for you. There you go.]

[Hamsterdance music]

[VOICEOVER: All right, where was I? Man, I just love this country—oh. Oh! Oh no!]

LATIF: And as it starts to go viral ...

SUNIL NAKRANI: The number of people are all bombarding with requests.

LATIF: ... 'til that one server is like ...

[VOICEOVER: Oh my God!]

LATIF: ... aah!

[VOICEOVER: Won't somebody help?]

LATIF: Can no longer serve up this website in a timely manner.

SUNIL NAKRANI: And so now people have to wait in a queue, right?

LATIF: And this would happen all the time, because back then, the way the servers were allocated was, like, you'd get how many servers you paid for.

SUNIL NAKRANI: Right.

LATIF: Like the owner of the Hamsterdance website probably would have been like, "Hey, I'll just pay for the one server."

SUNIL NAKRANI: Because I don't expect my website to experience a lot of demand.

LATIF: And how wrong they were.

SUNIL NAKRANI: Right.

LATIF: Do you remember what that was like, though?

LULU: Yes, I do!

LATIF: Like, how slow it was?

LULU: The, like, slow loading and the, like, ugh!

SUNIL NAKRANI: If something took longer than five seconds, generally the human psychology was that people give up on the website.

LATIF: So as Sunil saw it, the problem was relatively simple. There are a bunch of servers ...

SUNIL NAKRANI: And sometimes they can be not very busy.

LATIF: Sometimes they're just sitting around doing nothing. But other times ...

SUNIL NAKRANI: They could be overly busy.

LATIF: And so how, in a system that is changing so quickly, do you get those servers who are doing nothing to help those servers who are doing too much?

SUNIL NAKRANI: Start moving these servers around where they're needed.

LATIF: And this is the problem that Sunil brought to Craig.

SUNIL NAKRANI: Yeah, hi. I'm Sunil Nakrani.

CRAIG TOVEY: And I mean, heck, within 20 seconds I saw that the problem was similar to the honeybee problem.

LULU: Like immediately? Like that fast?

LATIF: Yeah, because to Craig, it was like he'd been holding on to this rusty old key that he was holding for more than a decade and Sunil showed up with something that looked like it might be the exactly matching lock.

CRAIG TOVEY: Yeah. Yeah.

SUNIL NAKRANI: Like, evolution has solved this problem in some way, right? And now you're saying, okay, can it do it in the artificial domain?

CRAIG TOVEY: And so ...

LATIF: They got to work.

CRAIG TOVEY: Right. F sub n of X sub N.

LATIF: Dusted off that old math equation.

SUNIL NAKRANI: Mapping the two structures of what the bees are doing.

LATIF: And sitting there, they started to connect all of these dots. Like ...

SUNIL NAKRANI: Collection of bees make up the beehive. Collection of servers make up the internet server farm.

CRAIG TOVEY: Yes. Exactly.

LATIF: And wait, so that would mean—oh my God, the stuff on the internet ...

CRAIG TOVEY: Are flower patches.

LATIF: Oh, bingo!

SUNIL NAKRANI: At that point, it got really exciting.

LATIF: Wow!

LULU: Wait, I'm sorry. Just please stop that.

LATIF: Sorry.

LULU: How is what the bees are doing anything like what the server's doing? Like, lay out the parallel.

LATIF: Okay, why don't we—okay, let's—why don't we Cranberry Lake this thing, okay?

LULU: Okay.

LATIF: So instead of a meadow filled with wildflowers.

LULU: Yeah?

LATIF: Okay? Picture the entire internet.

LULU: Boop-boop-boop-boop. Got it.

LATIF: It's the beginning of the day. You're just signing on to your computer, and the first thing you decide to do on there ...

[ARCHIVE CLIP: [laughs] Charlie! Charlie bit me.]

LATIF: ... is watch a video of a baby biting his older brother's finger. Naturally, right?

LULU: Charlie bit me. Yes, okay. How I love to start my day.

LATIF: By procrastinating, of course. Okay, so you watching this video is like a flower opening in the meadow, right? The meadow of the internet.

LULU: Why is me watching analogous to a flower opening?

LATIF: You watching is like the flower filling up with nectar. It's your desire—which they can capitalize on, basically.

LULU: Oh, me watching is money for—it's—oh, oh!

LATIF: It's money. It's money for them.

LULU: My eyes on the video ...

LATIF: Your eyeballs and your attention, that's like the server company's nectar.

LULU: Oh! Okay, I get it. I get it. We are the flowers.

LATIF: Yeah, and as more and more people watch and then share this video with their friends who watch it too, and share it with their friends who watch it too, that's like more and more flowers opening in the flower patch. And the server hosting that video is like a single bee being like, "Holy motherlode!"

[ARCHIVE CLIP: Charlie bit me!]

LATIF: Bonanza of nectar over here. I'm gonna need some backup.

LULU: Oh, so now it has to recruit more bees, aka more servers!

LATIF: Yeah. So it does like a computer version of a waggle dance, this server-to-server digital nudge. Okay, let's call it a ping. That means, "Hey, I need some help. Come get some of this stuff."

LULU: Okay.

LATIF: And so servers buzz over to the "Charlie bit me" video and start servicing all these flowers, right? Helping all these people to watch this video. And then it kind of keeps going, right? The video popularity grows, flowers keep opening. More bees needed, more servers needed.

[ARCHIVE CLIP: Charlie bit me!]

LATIF: And very quickly, all these servers are just servicing the "Charlie bit me" video.

LULU: Okay.

LATIF: But then, wait a second. Now there's a ping, but it's not for the "Charlie bit me" video.

[ARCHIVE CLIP: A giant container ship wedged from bank to bank ...]

LATIF: Now everybody wants to see pictures of this huge boat on the BBC website ...

[ARCHIVE CLIP: ... blocking one of the world's most important shipping lanes.]

LATIF: ... that just got stuck in the Suez Canal.

LULU: I remember that! I was a flower!

LATIF: Right. Right. So like you, all these flowers opening up all over the world All of a sudden, the servers that are serving up these boat pictures, now they're the ones who are saying, "Hey, we need backup." And again, there is this server-to-server, peer-to-peer, bee-to-bee waggle dance recruitment ping calling out to the idle servers to come help. And those servers come, and they ping out. And those servers who are no longer needed with the Charlie bit me thing are now like, "Oh, man, I gotta go help with this Suez Canal thing!"

LULU: Okay, got it. But then in the actual internet, this is happening times, like, a gazillion, right? All the time.

LATIF: Right.

LULU: Literally billions of people, billions of flowers. So could this algorithm, this little fix, like, actually work?

LATIF: Well, that—they had to figure that out.

SUNIL NAKRANI: You know, initially, excitement kind of sets in, but then you go to prove what you thought would be a good solution, right?

LATIF: To test this idea, Sunil and Craig decided to compare it to another algorithm they invented.

SUNIL NAKRANI: That algorithm, we named it Omniscient, where if it could see the future, what would it do?

LATIF: Like, if you knew ahead of time everything that everybody would want to see on the internet, how would you organize your servers to meet that demand? It's—it's basically God mode.

SUNIL NAKRANI: This is the best you can do. You can't do better than that.

LATIF: So they compared this perfect model to the way that humans had been allocating servers already.

LULU: Like, the guessing?

LATIF: Right. And they compared it as well to the bees' algorithm version. And what they found was that the bees ...

SUNIL NAKRANI: Even without knowing the future, they were coming within, like, 20, 15 percent of the optimal behavior.

LATIF: Wow!

LATIF: In a bunch of their tests, the human algorithms didn't even come close.

LULU: Whoa!

CRAIG TOVEY: Yeah.

SUNIL NAKRANI: It turns out that ...

LATIF: At least theoretically ...

SUNIL NAKRANI: ... it worked really well.

LATIF: So based off these results, they publish a little paper. And then Sunil goes back to Oxford to defend his thesis.

SUNIL NAKRANI: I presented the results to the committee.

LATIF: He tells them about the problem he's seen with the internet, and how the bees could help solve it. Talking for, like ...

SUNIL NAKRANI: Forty-five minutes, an hour.

LATIF: And when he's finally done, the first question that he gets is ...

SUNIL NAKRANI: Have you patented this idea or not?

LULU: Oh, wow! That's a good question to get!

LATIF: But he was like, "Well, I just published it."

SUNIL NAKRANI: You can't patent your own thing if you published.

LULU: Oh, no! [laughs]

LATIF: And so he did not patent it.

LULU: Oh, he gave it away?

LATIF: He gave it away for free to everybody. And over the following months and years, server farms all over the world worked this bee algorithm into the internet.

LULU: Wow!

CRAIG TOVEY: The honeybee algorithm made it 10, 20 percent more efficient.

LATIF: Which means we have the bees to thank for the internet being this place where ...

[ARCHIVE CLIP: US climber Alex Honnold stunned the crowds in Taipei after ...]

LATIF: ... you can get whatever you want ...

[ARCHIVE CLIP: Video from outside the Capitol shows the beginning of the storm ...]

LATIF: ... right now.

[ARCHIVE CLIP: The hurricane pushed Lake Pontchartrain deep.]

[ARCHIVE CLIP: It's peanut butter jelly time!]

LULU: I mean, that is just a moment for nature, that that could outperform so many thousands of human brains working on solving all kinds of problems, and just, like, evolution had figured this out through its own longer timeline of trial and error.

LATIF: Yeah. And not only that, this honeybee algorithm—or variations of it—have been lifted into so many other industries. I have found people researching how to use it to forecast exchange rates, design electric cars, detect defects in wood before using that wood for construction, even for sharpening MRI images to better detect breast cancer tumors.

LULU: Bless that, little bumblebee! I mean, I guess what's so—I think about humans, one of our gifts and our curses is that we can jump to the future.

LATIF: Mm-hmm?

LULU: You know? Or the past. We can jump out of the present. But we can worry. And that's like entire industries: prediction, forecasting. Whether it's weather or money or, you know, whatever. Like, but we spend time worrying about the future, but it feels like, what this—the elegance, it's like, not only to me is it beautiful that it came from bees—because I'm a nature lover and I love it when nature outsmarts us or has more elegance ...

LATIF: Yep.

LULU: ... in its design, but it's also, like, one of the insights, if you wipe away all the technicalities, is like, it's—it's throwing away the future. It's not basing its decision of where to move based on a guess about the future. It is only responding to the present.

LATIF: Hmm. Right. Right. It's like it's kind of at this very subtle—like, it's like, right at the line between responding to the clues of the moment and predicting the future.

LULU: [laughs] It's on the edge of the present?

LATIF: It's at the edge of the present. And it's like, all we gotta do is, like, pay attention to what's happening now and then—and, like, build it into these feedback loops so that we can—yeah, so that we can address just the next moment. Like, just the next moment and just the next moment after that, and just the next moment after that, and just the next moment after that.

LULU: I mean, that's—if anything, that might be a thing I, like, take around with me.

LATIF: This episode was reported by me, Latif Nasser, with reporting help from Maria Paz Gutiérrez, production by Maria Paz Gutiérrez, Annie McEwen and Pat Walters. Edited by Pat Walters and facts checked by Diane Kelly.

LULU: We also got a lot of help for this episode from Radiolab and Terrestrials' resident bug correspondent Sammy Ramsey, and we couldn't have done this one without his help. Big thank you to you, Sammy. And if you want to hear more about bees, and hear Sammy talking about them, we have a Terrestrials episode called "The Crystal Ball: Honeybees Who Predict the Future." You can go listen to him over there.

LATIF: Other major thank yous to John Bartholdi, John Vande Vate and James Marshall. As well, we want to thank the folks at AAAS who administer the one and only Golden Goose Award. If you remember, the award goes to government-funded science that sounds kind of silly or bizarre but then goes on to change the world. This research won a Golden Goose Award back in 2016, which is how I first heard about it. So thank you to all our friends there: Erin Heath, Gwendolyn Bogard, Valeria Sabate, Joanne Padrón Carney and Meredith Asbury.

LULU: And bees. We barely scratched the surface of how amazing they are. If you want to learn more, read any of Tom Seeley's books. His most recent one is the incredibly titled memoir, Piping Hot Bees and Boisterous Buzz-Runners.

LATIF: I think that should be it, but if you are still here, let me leave you with a bit of tape that has been haunting me.

LULU: [laughs]

LATIF: Sunil! Like, dude, over the last 20 years, I've used the internet a lot! If the internet today took the same amount of time as the internet in the '90s, like, I can easily imagine not just minutes, like, I mean, hours, full days of my life. Like—like, you know, a second at a time that this could have saved.

SUNIL NAKRANI: [laughs] You're right, but also what most internet service company wants is stickability, for you to keep using their service, right?

LATIF: Okay, so you're saying the opposite. You have made the internet so enjoyable that you have cost me days and hours, potentially even years of my life. So I should be—I should be mad at you.

SUNIL NAKRANI: Yeah.

[Hamsterdance music]

LATIF: And that's it.

LULU: See you next week.

[LISTENER: Hi, I'm Gabby. I'm from San Francisco. And here are the staff credits. Radiolab is hosted by Lulu Miller and Latif Nasser. Soren Wheeler is our executive editor. Sarah Sandbach is our executive director. Our managing editor is Pat Walters. Dylan Keefe is our director of sound design. Our staff includes: Jeremy Bloom, W. Harry Fortuna, David Gebel, Maria Paz Gutiérrez, Sindhu Gnanasambandan, Matt Kielty, Mona Madgavkar, Annie McEwen, Alex Neason, Sarah Qari, Rebecca Rand, Anisa Vietze, Arianne Wack, Molly Webster and Jessica Yung, with help from Gabby Santas. Our fact-checkers are Diane Kelly, Emily Krieger, Natalie Middleton, Angely Mercado and Sophie Sanahee.]

[LISTENER: Leadership support for Radiolab's science programming is provided by the Simons Foundation and the John Templeton Foundation. Foundational support for Radiolab was provided by the Alfred P. Sloan Foundation.]

-30-

 

Copyright © 2026 New York Public Radio. All rights reserved. Visit our website terms of use at www.wnyc.org for further information.

 

 New York Public Radio transcripts are created on a rush deadline, often by contractors. This text may not be in its final form and may be updated or revised in the future. Accuracy and availability may vary. The authoritative record of programming is the audio record.