077 = Earth That Quakes and Quick Fire Takes

Hello and welcome to A Problem Squared, the problem-solving podcast, which is a bit like a car, in that it is relatively practical, has a couple of airbags, and only a fool would let a drunk person drive it.

That fool is Matt Parker, a comedian and mathematician, and one of your hosts.

And I'm your other host, Beckhill, a comedian and not mathematician who got home from the agent's party at 3 a.m.

and is very much regretting agreeing to do a 9 a.m.

record today.

The hilarious thing is, I'm in an evening time zone right now.

I should be the one having a cheeky beverage, but no.

I mean, I'm not having one right now.

No.

I just want to make it very clear.

I haven't kept this going through the evening.

I did get up at seven, which means I have slept for four hours.

Yeah, because you're a professional.

I mean, yes,

yeah.

I want, I want to say, I'm not advocating for drinking in any way, form, or fashion.

Kids, it's not cool.

No, no, no, no.

Beck is facing her responsibilities maturely and dealing with all her obligations, despite

at a party yesterday.

I refer to it as borrowing happiness from the future.

It's going to be a fun one.

Oh, gosh.

In this episode,

I've sorted out some earthquake statistics.

I'm going to do some quick fire questions.

And I admit to a problem I've got with farts.

You and me both, buddy.

So, Bec, I know normally you're hosting this episode, so you'd ask me how I'm doing, but I'd like to read you a text message you sent me.

Oh, no.

Which I got when I woke up this morning and I looked at the time.

I got this at 9.49 a.m., which would have been 1.49 a.m.

in London.

Yeah, that's not too bad.

It's not so bad.

You'd actually sent me, there's like a few admin messages and then, and I'm going to read it normally, but it is in all caps.

Guess who is walking home from their agent's party with half a bag of tortilla chips?

They stole from thee, and then there's an attempt to spell the word cupboard.

So

to answer the two questions raised, number one, Beck.

Beck was the person walking home, having three little tear chips.

And to answer the second question, it was

C-U-P-V-O-A-4.

Cupboard.

That's so far away.

That's not even one typo.

I can't even be like, yeah, nah, that was a slip of the thumb.

Starter strong.

B and V are next to each other on the keyboard.

Sure.

And R is where the four is.

I don't know how you got to the number keyboard, but you did.

And then by then, you gave up.

There's no attempt at a D.

Just the word ends.

I think it was because I was typing in caps, but then I accidentally uncapsed it.

And that meant when I tried to fix it, I changed to the number pad.

That makes sense.

Had you had a good time?

Yes, I had a lovely time.

I actually ended up talking about this podcast with a fellow comic and explaining how much fun it is.

So that was nice.

But I had a very long day yesterday.

So I started cold water swimming.

I thought cold water swimming would be my worst enemy because I have terrible circulation.

I feel the cold so bad.

I get seasonal affective disorder.

Like all of the things.

This is your worst nightmare.

And then I'd listen to another podcast.

They talked about cold water swimming and how people had said it had helped them with seasonal affective disorder and things like that.

And so, my friend Nat Lutzima, who is a very funny writer, director, sometimes comedian,

she does cold water swimming and invited me along.

I was like, fine, I'll go.

And so I went to the Lido,

which for people who aren't from Britain is just a pool.

It's an outdoor, they call them an outdoor pool or Lido.

So there's no heating, the water is six degrees.

I'm literally, I'm in normal bathers or a swimsuit for non-Australians.

My friend did say, don't put your head under because I've got very long hair.

The water will just hold onto it and it'll be very hard for like my head and ears and stuff to warm up again.

And so we breaststroke with beanies on,

which is

as it looks as though there was a swimming.

It's my favorite band from the mid-2000s.

Breaststroke with beanies on.

Yeah.

Yeah, yeah.

I love those guys.

But

one thing that I didn't realize is I thought there would be more swimming involved.

Like, I thought you get in, you get used to it, and you do some laps.

It's so cold that you get, you tire the first time I did it, I managed to do one length, and even then, about halfway, I had to stop and walk the rest.

It is so exhausting.

But the thing that makes me keep going back is they have a sauna there.

So we were doing the very scandy thing of doing a length.

Nat usually does more than one.

jumping in the sauna for like 15 minutes, going back in for a length, jumping in the sauna for 15 minutes.

The first time I did it, I hated it, but I did feel amazing afterwards.

I'm not feeling the seasonal affective disorder is bad.

I've only done it twice, so we'll see if that continues.

But that's what I was doing yesterday morning.

Then I went for a very, very long walk and I wrote some script, did some yoga, and it was a very long day.

So I am still recovering.

It's still a rewarding day.

Well,

as we speak, at the time of record,

I handed in my book.

I don't know if you can tell through

the screen.

I, the weight being lifted off my shoulders.

I said to you, you're looking very slick, Matt.

Like, you've got a glow about you.

And I'm not just referring to the actual sunlight that seems to be reflecting off your face.

I've got a window open and the sun is beaming through.

Yeah, and you're beaming back out.

I can see now.

Very happy.

I see that a weight has been lifted.

Yeah, within the last 48 hours, because it was like half past midnight on local Monday.

And my goodness, am I happy to have that book done?

I celebrated by shaving all the hair on my head to be exactly the same length.

I had

same length on your head as in, like, also your beard and your eyebrows.

Yep.

Yep.

Ah, now, spoiler.

Yeah.

So

when I was writing a book, I kind of stopped shaving.

I'll shave if I have to, like, film a video where a beard's not appropriate or do a live thing where I need to look clean shaven.

But for the most part, I cease to care about a lot of things.

Yeah, I stopped showering when I was writing.

I see, exactly.

So to celebrate, I shaved off my writing beard and I did my head as well.

And when I do this traditionally, I need to be careful because I'm very absent-minded.

And if I'm not paying attention, I will occasionally zip into my sideies by accident.

And I'm like, ah, I want to dofus.

Whereas this time I was like, no, I'm going to do everything the same.

It's all short.

I don't care.

It's new, book-free, care-free, matte.

Now, I had not accounted for the third region of hair on my head, which is my eyebrows.

Maybe the eyebrows, I do not want to be at the same level of shortness.

And I caught myself when I took the edge of one eyebrow before I was like, oh yeah, I've got eyebrows.

Well, now I got slightly less eyebrows.

So now I look.

So you're doing that cool thing from like that all the youth were doing like maybe about 15 years ago.

Exactly.

Well, you know what?

I had a moment where I was like, well, my options are I could do all my eyebrows to make them equal or I could do the symmetric notch on the other side just to make it look deliberate.

And I've gone with leave it how it is and just try and style it out.

No, I mean I can't tell.

I want you I want you to lean into the camera.

And the sun.

Nah, nah, you can't tell.

I think it's because I got bushy eyebrows and I've just rendered this bit normal and then the rest is just, you know, bushy.

Yeah.

You've done the thing that a lot of women in the 90s or early noughties did where we over plucked, except the problem is eventually our eyebrows stopped regrowing.

I do what you did with your eyebrow, but with the back of my head, because I do my own undercut.

And so I have to basically, through a series of mirrors and a sense of touch, work out if I've hit the line of the undercut at the back.

And then I keep accidentally cutting into my hairline.

So my undercut keeps getting higher and higher, like every couple of months.

Like eventually, I'm just going to have a fringe.

Just resets around to the front and you start again.

Yeah, yeah, that's right.

It's like painting the fourth bridge.

Yep.

This first problem comes from Bernardo from the problem posing page, which is a problemsquared.com.

And Bernardo says, in Mexico, we're scared of September 19th because there have been quite hard and serious earthquakes in different years.

The first that is always mentioned was in 1985 with a magnitude of 8.1 on the Richter scale.

Since then, every year we honor the ones who died and have a drill.

Nevertheless, in 2017, history repeated, an hour after the national drill, there was an earthquake of 7.1.

What's even worse is that in 2022, another earthquake hit Mexico, this time with an intensity of 7.7.

So that's three earthquakes, all happening on the same date.

Different years, but same date.

I mean,

that's got to be confusing if you've just finished the national drill.

to get ready for an earthquake and then there's an earthquake an hour later.

Yeah, does it make you extra ready for it?

Or does it actually

it's better than it happening before the drill because then you'd be like, oh, well, this is just the drill.

You wouldn't want to be able to do it.

I think if there's an earthquake, you'd notice, surely.

Like, that's true.

You're like, oh no, the ground is shaking.

That's an earthquake.

Well, they also say that there have been two recent earthquakes on September the 7th as well, one in 2017 and one in 2021.

Oh, my goodness.

Yep.

They say they know it's all a coincidence, but they find it odd that it's been happening with such frequency.

They said, what are the odds that this happens?

And how common is it?

And has it happened anywhere else?

Good problem.

I mean, I was drawn to this problem because it falls into the category of things where I'm like, ah, I just have to look up some data.

In this case, I get the data for all the earthquakes ever.

And then I analyze the data to see, you know, how they land on different days, if they happen on the same date in close proximity, and then I'll have my answer.

So I said about that, and I found

that this is going to be an easy thing to solve.

And I'm guessing are you foreshadowing?

Yeah, yeah.

I just thought this would be easy.

Yeah.

And the first step was...

No, well, so far.

So far, past Matt was very happy with how it went because I looked up a database of earthquakes and I found the National Centers for Environmental Information, which are part of the National Oceanic and Atmospheric Administration, which is part of the U.S.

Department of Commerce, which is part of the USA government, which is part of, I don't know, the world government.

So the National Centers for Environmental Information, they have the Global Significant Earthquake Database where they have logged every database that we have records for, which goes back to the year 2150 BCE.

So that's over 4,000 years.

Yeah.

Did they have records that long ago?

Oh, yeah.

Things start like a big earthquake.

People write that down.

yeah but that's a really long it's a very long well 2000 how many years 4 000 years ago we had writing people can write it down and they look up the records everyone's like oh my goodness the earth shook today it was a lot and now within the database not every earthquake has every bit of information some of them do have like a richter scale measurement and some don't because i guess i imagine there's also like proof of it in the rocks and stuff as well, like in the earth.

You know, I did dig through some of the kind of surrounding information about the database, and I couldn't find any documentation on their methods for putting in historical earthquakes.

So, I don't know if they're just working off historical records or if they are taking into account any geological information.

I can tell you that.

Anyone listening works for the National Center of

Environmental Information.

Yeah, all of the, if anyone works for them, let us know your sources.

And they declare a significant earthquake has to meet one one of the following criteria.

It has to have either had moderate damage, which they pinned at approximately 1 million current US dollars worth of damage, or more.

Or it could have caused 10 or more deaths, which is a good point.

I mean, earthquakes cause significant loss of life sometimes.

Or it can have created a tsunami.

So if an earthquake causes a tsunami, that causes other destruction.

That counts as a significant earthquake.

Or it can just be above magnitude 7.5.

Any one of those criteria, in it goes, and I was like, this is it, this is easy.

I've just got to download this database and analyze it.

Now, the next bit of the story is me complaining about how difficult it was to get this data.

Classic, this is going to be easy.

No, it's not.

Because if people recall, we did a problem ages ago about how much does

the

slope, the hills and the troughs of a country increase its surface area, and if that's factored into official surface areas, and I spent a long time complaining about how hard it was to get the kind of geospatial data to work that out.

And this was a very similar thing because I found you could download the map data, but then I had to work out

how to convert that into a form that I could then learn to import into Python.

And I did all that, and I realized they hadn't updated the map data since 2017, and I wanted the data to 2023.

So I was like, oh, so I went back to the website.

And then I realized they've got a search function where you can put in what earthquake parameters you want.

And I realized if I left them all blank and just hit search,

it would just return every single earthquake in the database.

Right.

Yeah, the whole lot.

But then I couldn't get it out of the table on the website because it's a weird JavaScript embedded table.

So all my normal methods of importing a table from a website didn't work.

I couldn't even select the data and paste it.

I was like, you couldn't download or save the website.

I couldn't bring it up in Google Sheets.

Nothing was working.

And then eventually I realized, I was looking at it, I was like, wait a minute, why is there an extra icon hidden next to one of the sort drop-down menus at the very top of the embedded table?

And I moused over

here.

Wait, are you referencing mid-90s Sandra Bullock film The Net?

Yes, yes, I am.

The Net is a classic.

I will fight you on this.

I love it.

The little, what did the little logo look like to you, Matt?

Because the one in the net is a pie symbol.

So I would have thought you would have been excited about that.

A pie symbol would have caught my attention a lot sooner.

Now that I look at it, like it's tiny.

It's very tiny.

Thank you.

I spent a long time trying to find a way to download the data.

I didn't see that.

Spent a long time working out.

Like, I got to the point of, can I write

like a keystroke bit of scraping to take each row at a time?

Because you couldn't select more than a bit at once.

And then I realized that they're not all the same width.

I'm like, how am I going to do that?

And then I found the tiny icon.

So I downloaded the tab, separated values, an hour and a half after I'd started this journey.

And then I spent about another hour getting the data into a nice manageable form where I can start to analyze it.

And there are 6,408 earthquakes.

Not all of them have a date associated with them.

So I took out all the earthquakes that we don't know what day of the year they occurred on.

And that gave me 5,810 earthquakes.

We have a date.

So my first kind of question was, are earthquakes happening randomly on different dates?

Which they should be.

I can't imagine, as a lay person,

I couldn't come up with a plausible mechanism where

like the position in the Earth's orbit,

the tidal forces orbiting the sun could change when earthquakes happen in the the year.

I was like, I don't think it's significant enough to do that.

I feel like we would have heard more about it.

Like, they would say it's earthquake season in the second year.

Oh, it's earthquake season.

It's tycoon season and tornado season and stuff.

So there you go.

Yeah.

Yeah.

Proof by not being in the weather report.

So

I then

said about just counting the number, like for all the 365 days of the year, counting from that database how many there are per day,

which was oh, well, that would have been reasonably straightforward if it wasn't for leap years,

because every now and then there'll be an earthquake on the 29th of February.

And I was like, I don't want to have to deal with that.

So, I went with my classic solution to this sort of problem, indeed, to many sorts of problems, which was to ignore it.

Moreover, to remove it, I just took out all the earthquakes that were on the 29th of January, of which there were three.

So they're gone out of my my data.

Which that's a very like, it'd be interesting to work out whether that would have fallen on the 1st of March and stuff.

But I see what you mean.

Yeah.

It's like, yeah.

I mean, the Earth doesn't care what date it is.

So it still would have been the same as all the others.

We just only declare that date to occur roughly a quarter as often.

We're not quite a quarter anyway.

So, I mean, and I could have written the code to account for that.

I could have scaled it based on how often each date happens.

But if I just got rid of that one, all the other dates are the same.

Oh, and I didn't look into because historically, like, we've not always had all the months.

So I don't know if the very old dates, like back in the negative years, if they used what date it would have been considered at the time.

or if they've converted them all to modern dates, which I assume they have.

But a lot of the older ones didn't have dates anyway.

We just have like years.

And this doofus just got rid of that data.

And of the remaining 5,807, it's pretty even.

The date in history with the fewest earthquakes was the 27th of April.

Only five.

Five earthquakes.

And the date with the most is 18th of April, when there were 31.

Closely followed by the 9th of July has had 30.

And it looks pretty even, just kind of eyeballing it.

But then I was curious to know what it would look like.

Like, what would actual random data look like?

So I made, I generated the same data set randomly.

So basically, I took the same number of earthquakes but gave them random dates and then plotted those and looked at the same distribution.

And so the real data goes from a minimum of five to a maximum of 31.

And my first

time faking the data went from a minimum of five to a maximum of 32.

So,

pretty much the same.

I did it again.

The next time I did it was exactly between five and 31.

So

it's indistinguishable from random data.

Interesting.

So it's a bit like trying to calculate the chances of someone sharing your birthday.

Oh, yeah, 100%.

Yeah, yeah, yeah.

So, well, actually, no, number one, you're completely correct.

And when I was working on this, I was thinking, what's called the birthday paradox?

It's not a paradox.

It's just confusing.

Is it a paradox because it's not a paradox?

It's

paradoxically called a paradox.

Yeah.

Yeah.

But it is exactly the same question.

It's like, if a bunch of people are born, what's the chance that two people have the same birthday?

And you work it out, assuming they're randomly distributed.

Now, human births are not randomly distributed during the year.

There's clumps.

Whereas earthquakes are.

So this is actually more random data than the do two people have the same birthday huh because when two tectonic plates love each other very much that happens

Completely randomly

so yeah, so I had convinced myself the dates are random now.

I had to look at are there places in the world that have multiple earthquakes happening at the same time within within a certain distance.

I mean first of all I ran it and I got like loads of places that had the same date, but like some earthquakes were like 3,000 years ago and then some were now.

And I was like, wow.

I mean, the way, because with these probability questions, a lot of it comes down to where you draw your line around what counts as a match.

And we're only talking about this because it's in the, you know, popular psyche in Mexico.

because it happened three times within living memory, like within one generation, since the 80s.

That's what I was

And I was like,

am I casting the net too wide?

I thought I should try and be realistic and cast it as accurately as possible

to try and match as close as possible.

Because

if there's other ones within exactly the same criteria, then we can say, yes, it is a thing that happens a lot.

And it's just random clumping of data, which you'd expect from random data, random data clumps.

You've got the data of obviously what years they were and stuff.

So you could also.

Could you graph it to to show us whether the dates that they occur on starts to change or do they start to change let's say there were only two on the was it the 9th of july that had the most oh the most overall 18th of april so

it was 31 that have occurred on the 18th of april since yep you know this this data was being collected way back yep it'd be interesting to know if let's say all the way up until 1950, there'd only been two on the 18th of April.

Oh, yeah.

And then like 29 of them happened since 1950.

Yeah, you're right.

I could plot over time, it'd be like one of those animated bar charts where you see different things overtaking and dropping and pulling ahead, like populations changing over time for countries or something like that.

The issue is the data we have changes dramatically dramatically based on how good we are at recording earthquakes.

And it also depends on where we have

earthquake detection and documentation.

And it depends on which past civilizations their records have survived for us to now read them.

Because there are a lot of civilizations

through accidents or malice, we don't have their records anymore.

Yeah,

I would also say it's not again, it might be geological just data as well, but

I was just using that as an example.

Even if you were to do, take the data from the last 50 years and then see if there's any trends or changes in how often they occur on a specific date.

I could say it's indistinguishable from random data.

Ah, all right.

And I did actually plot the number of earthquakes per decade across the entire data set, and you can see a very steady, it's very low for a very long time.

And so I plotted the number of earthquakes per decade, and there's a very low number until about the year 1500, and then they gradually start to pick up.

And then I plotted it from 1700 until now, it's an almost linear gradual increase, just as there are more places that are actively monitoring and logging earthquakes.

Until actually, it takes until quite recent.

I did every year from 1700 until now, and it's not really, and even now, maybe there's still a bunch we're missing.

We don't know if there's like earthquakes happening at sea that we're not logging.

I don't know.

But we are at a point where, as of the

late 20th century, even during the 20th century, it's going up.

It's probably not until

roughly the year 2000, I think we may have comprehensive earthquake detection.

But that's just me me eyeballing a plot I made.

People who know more about this may be yelling at their favorite podcast app.

So what I ended up doing was trying to make it as similar as possible to what happened in Mexico.

The earliest one was in 1985.

So I thought, you know what?

I'll just go from 1980 onwards, which is during my lifetime.

I was born in 1980.

I thought I'll do my lifetime.

Bit of a bias there.

I then calculated the distances between all the earthquakes because I've got the, they're in the database.

I found all the ones from the 19th of September that happened in Mexico.

I then looked at the distances between them and found the most distant two were 483 kilometers apart.

So I thought, you know what?

I'll look for groups of earthquakes since 1980, all on the same date, and all within 500 kilometers of each other.

Okay, yeah.

I crunched the data.

First of all, I can confirm that three

is the maximum.

So Bernardo, the fact that there were three earthquakes all on the same date, all within 500 kilometers in Mexico, there's nowhere else that can claim to have had more earthquakes on the same date than that within the same criteria.

However, it has happened eight times.

As in, like, there are eight different regions of the world that have all had three earthquakes on the same date since 1980.

Wow.

And by regions, I mean 500 kilometers across to match the ones.

So it's happened.

There was a group between Pakistan and China, which 2002, 2006, 2015, all on Christmas

on the 25th of December.

Then you've got, again, I found the three popped up in Mexico.

So it's good to know that I pulled out.

What's nice?

And this is something I always refer to as finding the nugget.

So my maternal grandfather built a machine, which was like a huge automated kind of centrifuge, spinny thing with water, to find literal gold nuggets from a lot of dirt.

So the idea was you you would go out to somewhere with this gold, dig up a bunch of candidate dirt, run it through the machine, it would spit out if there was a nugget.

And to demonstrate this, he had a gold nugget he'd previously found.

And he would put it in the dirt, run it through the machine, and demonstrate that it found it and popped it back out yeah how come i don't know about this this

gold centrifuge man

oh you know he uh he he was a man of many contraptions

would you believe an ancestor of mine yes yeah how about that yeah so when i was at school i did this like extension thing in primary school it's called the primary extension and challenge piak in western australia where you got pulled out of school like once every couple weeks or something or once a term.

And there was a thing where if you made like a bunch of things, you got like some certificate.

And he found out about this and he basically was like, all right, clear your schedule.

And

he was retired at the time.

So we just used all his tools and equipment and retirement time to, we're like, right.

And we had to plan and machined up all these different things, which I think explains something about where I've ended up in life.

But what I actually took away from that is whenever I'm doing this kind of thing, I know I should find the original example in the data.

So Bernardo said we had these three things happening on the same date.

So I know if I separately, like ignoring that completely, if I separately get the database for all the earthquakes and run it in, it should spit out one of the examples should be those three.

And there they are.

They've come out.

They're on my list.

They're position seven.

which is an arbitrary position.

I'm counting up the list to go through them all.

And so I can see there, there's all three from Mexico.

I then got three, which are all in China.

2002, 2009, 2017, all happened on the 8th of the 8th.

So that's the 8th of August.

We've then got three in Japan that happened in 1982, 2005, 2008, on the 23rd of July.

And I won't go through all of these.

There's a few in Iran that all happened on the 22nd of July.

We've got a collection in

Honduras, which all happened on the 10th of April.

The Philippines, three happened in the Philippines, all on the 5th of March.

And finally, on the 3rd of January, we have three across northern India, it appears.

Wow, this is like a really depressing lotto.

Yes, yeah.

While earthquakes are random on date, they're not random when it comes to region because they occur where there's more tectonic activity.

But it is interesting, there are eight different times when we've had three earthquakes all on the same date, all since 1980, all within 500 kilometers.

And in fact, just looking at the 19th of September, yes, there were three earthquakes all on that date, all within 500 kilometers in Mexico.

There were also

two earthquakes on the same date within 500 kilometers in China.

So in China, in 1939 and in 1966, they had significant earthquakes on the 19th of September, and the two of them were within 500k of each other.

Wow.

So I feel like I can say comprehensively, first of all, it is random.

And secondly, it is something that does happen.

I mean, this is a seemingly natural amount of clumping of earthquakes.

But obviously, as humans, it'd be very hard for us to ignore that because we love spotting those sorts of patterns.

And I think what's happened in Mexico is quite clever, turning it into

a recognized day when you prepare for earthquakes.

The fact that humans are going to look for these sorts of patterns and think, oh, is it a conspiracy?

You know, is this where, you know, earthquakes being manufactured artificially?

No, this is just how random data clumps

happens all the time.

That the government would be capable of creating an earthquake and wouldn't think to try it on different dates.

Yeah, someone

when they least expect.

Ah, we reset the machine again.

Yeah, yeah.

So there you are.

Yes, it happens.

And it's happened eight times.

And we should just say, like, we're just talking about the data here.

We're aware that there are massive consequences that happen with these things.

We're not trying to trivialize them.

Oh, my goodness, yes.

Yeah.

Well,

I'm impressed by that.

I think it's up to Bernardo to give us a ding for this one, but I can't see why it couldn't be.

I think that

you've answered all those questions.

So, um, what was my analogy for this episode?

A car?

So, I'm gonna give you a beep beep

up next, and totally unrelated to the fact that Beck is feeling a bit worse for where we're gonna have a quick fire round where we go through a few problems, which were probably not enough for a whole dinglet, but we still feel should be answered, or at least commented upon.

The first one up from someone, well, they put their name in as Cal, K-E-L-L.

They went to the problem posing page at a problemsquared.com and said, What podcast should I listen to while I wait for new Problem Squared?

They said they finally caught up.

I guess they're working through the back catalogue, and now they no longer have a podcast to binge while they're doing things like working or playing video games.

They already have acknowledged we said no such thing as a fish, and they want to know in that case, what episode should they start with?

Well, obviously, one of the episodes that Beck and or I have been a guest on.

Beck, have you got any suggestions?

I mean, you've already answered one of them, but so thank you.

If you enjoy this show, I would thoroughly recommend a show called Science Versus,

which check their sources far more than Matt and I do.

So that's Science vs.

Science VS, that is.

Obviously, recommend a podcast of unnecessary detail.

How many series of that?

That was my recommendation.

That's the only one I had.

Because I don't consume that many podcasts, but I am a fraction of a podcast of unnecessary detail.

It's a lot more sporadic than this one.

This one comes out every two weeks, whereas podcasts of unnecessary detail appears whenever we can be bothered.

So keep an eye out.

Well, I'm also going to give a shout out to We Can Be Weirdos, which is Dan Schreiber's solo podcast.

Dan Schreiber is one of the hosts on No Such Thing as a Fish.

If you're a fan of sort of true crime, but you don't want murder stuff, I would thoroughly recommend a podcast called Swindled, which is mainly looking at corporate crime.

If you're a bit of a crime junkie, you sort of get your crime fixed, but you also feel a little bit better because you're like, oh, I'm being educated about stuff that we should know about.

Right.

If you like deep dives, Dakota Ring is a really, really great podcast.

They've got some really good episodes.

They looked at the history of kissing, of making Out, and it turns out that it's way more recent than we thought.

It's really fascinating.

They had one about cabbage patch dolls, just basically cultural stuff.

They'll do a massive deep dive into it, which is very interesting.

And obviously, I have to recommend the other podcast that will have several episodes out by the time that you're listening to this.

It's called Enemy in Paris.

It's for anyone who either has watched Emily in Paris and wants to hear two people tearing it apart or anyone who doesn't want to watch it but is interested in hearing two people give scene-by-scene accounts and tear it apart.

So yeah, check out the Design Spark podcast with Professor Lucy Rogers and Harriet Brain.

Sort of deep dives into various tech things.

So there are so many great science and math podcasts out there.

I'm going to actually recommend that people post about that in Reddit or talk about it in Discord.

That'd be great because I think we should all share the love.

Our next mini problem.

This one is real mini, but the person seems very keen.

So this has come in from, and their name is, Why Does Beck Hill Know Edgar Wright?

And their problem is, why does Beck Hill know Edgar Wright?

How?

I mean, they've added to it slightly there in the actual problem.

So, Beck, the problem is, how do you know Edgar Wright?

How?

How?

I'd love to say it's because he's a huge fan of mine, like you get to say with some of your fans, Matt.

And look, in a sense, he is sort of basically very early days twitter i i mean very early days when i used to do a stick figure cartoon on the internet i did a bunch of scenes from edgar right films and then i posted it on twitter he um retweeted it and then he followed me we do run in similar circles so i've got friends who know him i'm i'm yet to go further than sliding into the dms occasionally and bonding with him over a hatred of particular books.

That counts.

So that's the answer.

I'm sorry, it's not more.

I'd love to make something up.

Good answer.

I feel like if I make something up on this show,

we solve problems properly.

The final one in our rapid fire is from a person who just put in the little kind of snowman emoji, and they want to know what is the best way to make a snowball.

Wow.

I like to suggest that this snowman emoji means that a snowman sent in this problem, and they are really struggling to create their own genitalia.

But I thought Wow.

I have several answers for this.

One is that you throw it an extravagant party.

Oh, yep, nice.

Hey,

did you get an invite to the snowball?

Great, good work, good work.

Thank you.

My other answer on how to make a snowball is 10 to 15 milliliters of lime juice or lime cordial, 50 milliliters of avocar, and 50 milliliters of sparkling lemonade.

That's a cocktail joke, I believe.

It is.

It is a cocktail called a a snowball.

I like them.

A lot of people hate them.

If we were in the same room, we'd be like, oh, and we've made snowballs.

But that's.

I think today

of all days, I should.

It's a good thing I'm not having a drink.

Hard no.

Yeah, how about that?

And that was Rapid Fire Problem Solving with Beck.

Thanks.

Thanks, everyone.

I'll put more effort in next time, I promise.

And it's any other beep beepness.

Just like a car, this AOB is going to be full of gas.

Good work.

Toot toot.

Speaking of toot-toot, farts.

Yes, farts.

So we heard from Emma, who said, I'm a physics PhD student, unfortunately, writing to deflate one of Matt's solutions to the question of using farts for lifting.

Buoyant force is just two gravitational forces in a trench coat.

I mean,

she's not wrong.

It's a great analogy.

So, decreasing the gravitational force on you will also proportionately decrease the buoyant force from your fart suit and won't help you float.

Gravity shows up once in the gravitational force on you and once in the buoyant force from the farts, so they just cancel.

If you want to gain the buoyant force, you'll have to change the ratio of densities between the atmospheric gas and the farts.

Now, a lot of people wrote in with this, none quite as eloquently as Emma, but plenty of people pointed this out.

For any new listeners, when Emma says fart coat, I was talking about, when Emma says fart suit, I was talking about if you could fart enough to lift yourself off the ground, I talked about if you could trap all your farts and become buoyant.

And then when I worked out how much you would need to become buoyant, I then said, oh, but, you know, if there was less gravity holding you down, blah, blah, blah, blah, blah.

And I totally forgot that, yes, of course, buoyancy is the kind of ratio of densities.

And if you've got less gravity pulling you down, you've got less gravity pulling the air down, the buoyancy won't change.

Yep.

And so, Emma and everyone else who wrote in, you are absolutely correct.

That was a fundamental misunderstanding of the nature of being buoyant because you're wearing an airtight suit that's capturing all your farts.

Thank you.

Thanks, everyone.

It was a lot.

I will also flag up: at least Magnus pointed this out on the problem posing page.

I said Beck weighed 294 newtons, which they're saying is not the correct way to convert 60 kilograms into newtons.

And so I went back and checked my spreadsheet from that episode.

That was a verbal typo.

I have no idea where I got that number from and why I said it.

My spreadsheet was doing it correctly, and it says 588.6 newtons, which is, I can see I'm multiplying gravity by the 60 kilograms I put in.

So I'd like to apologize to everyone for my verbal typo.

My expertise was correct, as opposed to my fundamental misunderstanding of how buoyancy works.

And we also had some any other business from a Patreon supporter.

Yep.

Justin Begley.

So I'm not complaining because I think I'm definitely getting my money's worth from my subscription, but I feel Matt's random picker algorithm may be a touch bias.

What would the chances of having my name read out three times be?

Is there some sort of conspiracy going on?

I download, either directly before or occasionally during the record, I download all the Patreon supporters of a problem squared from patreon.com/slash a problem squared who are active Patreon supporters.

I put them in a spreadsheet.

I assign them all a random number using Excel's built-in rand function.

And then I pick the three Patreon supporters who were assigned the biggest random number.

And we read them out.

So it's as random as random can be using the pseudo-random number generator built into Excel.

Or, or

maybe

people are slipping me a little something on the side to bump up their rankings in the random number generator.

So, you know,

I think that's fair.

Why am I getting slipped anything?

You know, whoever's closest to the random number tap gets the grease.

And that's the phrase, I'm pretty sure.

Yeah, that's what they say.

No, it's 100% legit, everybody.

I don't want to have to go to a system where we have to write people's names on like lottery balls and put them in a mixing Tombola thing.

Because then if someone stops supporting us on Patreon, we'll have to find them and fish them out.

We'll have to find the ball and take it out.

No, no, we're not doing that.

But speaking of which, we have our...

Three assigned random generated number Patreon supporters who we're going to thank on this episode, and we'll do our best to mispronounce every single one of them.

So, we would like to thank

well, okay, they put their name in as Captain Untouchable, but I feel like we have to mispronounce that.

So, I'm going to go with Captain

Unto Chab Lay

Untouched Blay.

That's the one

Bowles Golds tien.

Anton Weissy, art, uh,

and Anto, New York City, Arter.

Oh, yeah, look at that.

And,

and to New York City, art.

Uh.

And if you would like us to completely demolish your names, you can support us on Patreon.

You get more.

You get more than that.

We have a bonus podcast.

A bonus podcast.

Oh, gosh, it's happening now, guys.

It's happening.

This is the quality of the bonus podcast, too, by the way.

I would argue it's better.

Yeah, it comes out once a month.

It's called I'm a Wizard, and that is just for our Patreon supporters as a way of thanking you for making this possible because without it, we can't do the show.

And then everyone else who can't pay or indeed understands that they shouldn't pay for this

can also listen to it.

People who are good with their money.

I also want to thank my co-host, Matt Parker, who has been incredibly patient with me today.

And I would also like to thank our wonderful producer, Lauren Armstrong Carter, who is...

Let's see, if Matt's letting me drive this one, then Lauren is the police car chasing us down the road, trying to get us to pull over and do as little damage as possible.

She's the voice in the sat-nav.

Oh, she's like Kit.

Oh, she's like Kit!

Yeah,

an even more topical reference than the net.

You've gone for Night Rider, yes, because I am nothing if not consistent with my zeitgeist.

Beck, finger on the pulse hill, we call her.

Please, please don't put the word finger in my name.

Too late.

Bec, I'm afraid before you're allowed to go, you have to guess how many dice are in the jar.

To remind you, last time you guessed 420.

Very funny.

And I hilariously said higher.

You did.

We all have a good laugh.

We did.

Yeah.

So I now know it's between 420 and 486.

So I'm going to again try and sort of go down the halfway point.

Oh, efficient.

Yep.

Which I think would make it 453.

Higher.

Ooh.

Okay.

You've just halved your options, though.

Yeah, man.

I'm in this for the win.