Are quizzes and puzzles dying in the era of the internet - where any answer is seemingly a click away - or can they be Google-proofed?

This week, we ask our in-house crossword expert to tell us what types of questions create answers that are hard to search for, how a puzzle question should be structured, and what features make for a good puzzle. We also discuss what makes something un-Googleable and consider the implications for human knowledge sharing. If listening to this podcast doesn’t make want to go to your next local pub quiz night, nothing will.

A few things we mentioned in this podcast:

• Quizmaster devises Google-proof questions to stop pub quiz smartphone cheats
• Google-ability and Google ability
• ‘How Google Works’, according to Google developers (note: this doesn’t tell you how Google works)
• The ‘Tip of my Tongue’ subreddit
• Cognitive Engineering podcast: Why do we like puzzles?

For more information on Aleph Insights visit our website https://alephinsights.com or to get in touch about our podcast email [email protected]

Hello and welcome to the Cognitive Engineering Podcast produced by me, Fraser McGruer for Aleph Insights. In this series of podcasts, we take a look at interesting topics and discuss what we think they tell us about analysis and decision making. I'm here with Chris Lear, Tom Spence and Nick Hare of Aleph Insights. And this week we're discussing Google proof

puzzles. Nick, what are we talking about? What's going on? Quite a long time ago, I used to take part in pub quizzes when I was younger, didn't have kids and that sort of thing. It was a pub quiz up at the Prince of Wales in Highgate, which was famous for being one of the hardest pub quizzes in London. And we'd always come like second or third. And once ever we came first, but that was a really, that was a day of revelry. Who was on your team? It was usually me and my dad and my dad's like journalist friends. Brilliant. Who had normally very good. But the day we won, she had a friend of mine who's a history of art person and they were able to fill some gaps in the knowledge. So you had the breadth of the journalists and the depth in that one thing. Exactly. Yeah. So it was, it was, but, but the, I actually haven't done pub quiz a long time, but a few weeks ago I met up with some friends of mine in Stoke Newington and the pub we were in, they happened to be starting a pub quiz. We didn't hang around for it, but I did manage to get hold of some of the questions because I was interested to see what pub quizzes look like now. And clearly the nature of the pub quiz has changed quite a lot. And of course the major driver is the ready availability of all the world's knowledge in your mobile phone.

But are you talking about just, are we talking about dishonesty here? Is that what you're talking about? Or are you talking about how it's easier?

Well here's the question. Yeah, go on. How do you make a pub quiz immune to being Google-able? But do you mean in the moment to stop people cheating? Is that what you mean? You're saying that using your phone to look things up is cheating. I mean, okay, fair enough. If you call that cheating, which it would be if you said, right, this is a phone, no phone pub quiz, then obviously it would be cheating. But what I'm saying is given how hard it is to enforce that, what would be really good is if we can create quizzes that

are... Yeah, okay. So I would just, no, but seriously, I just want to clarify, because I was wondering if for a moment you were talking about that, how in terms of setting quizzes, there's that, because there's so much, because information is so accessible now, is that it's either too difficult or the kind of information that's accessible is...

Right. So the thing is, you could say, it would be good to be able to say, well, you can use your mobile phone if you like, but it won't help you. Right. So, and the other, and of course, the other reason that Mr. Lear is with us, apart from being generally ace, is that he is one of the team who produced the Magpie crossword puzzle magazine, which is famously extremely hard. And, but at the same time, it certainly cannot, is completely invulnerable really to Googling. At least most of solving a puzzle will not, Google will not help you. So... Chambers might though. Sorry? Chambers might. Yeah, yeah. Access to a dictionary is helpful. But so, so anyway, I thought he might have some views on, you know, whether or not, whether or not Google has, has changed the way that crosswords are set and the kinds of things that you can ask people. But yeah, there we are. But I think it might be interesting to delve into why some things are easily Googled and other things aren't. Chris, your moment.

Okay, well, I'll start with, I'll start with things that are hard to Google. How about that? Yeah. Things that are hard to Google. This would be the quiz rounds that I would do if I was putting on a quiz and I wanted to say, you can use your phone as much as you like, couldn't care less, go for it. I would do a smell round. That would be how I'd start. I'd do a smell round and then I'd do a sounds round. Because although apps are getting good at recognising sounds now, they're still not as good as they are recognising text. And then I might do a pictures round. I give people a bit of paper and they have to recognise the things. I think those would be the rounds to make it hard to Google. And the other things I would do is I would make questions have some sort of intersection. So I'd ask a question like, what links, I came up with one actually, what links, what links the writer of Lolita with the dude from the Big Lebowski? So that's Nabokov and the dude, what links? And if I said, now, I'm just going to give you one bit of information. If I said it's a cocktail, you might get it. Oh, the White Russian. White Russian. Ah. You see, what we're doing there is that, yeah, because Nabokov's White Russian, emigrate, and obviously that cocktail is the dude's cocktail. And so what that demonstrates is that something you kind of, as soon as you're given that one bit of extra information, you can get it. It's a great, great question. If you were to Google it, you would get it with that extra bit of information about the cocktail. And that's how, that's actually how, that's how you can both Google proof a quiz question and make a question sort of, you can make a quiz problem become easier by just adding little bits of extra information, which actually make it more Google-able and make it sort of more answerable. It's like code breaking. The more there is. It's code breaking. You give people information incrementally. And in Magpie Crosswords, which Nick mentioned, he said they are Google proof. They're not, actually. Google is now a search engine. Let's be BBCish about it. It used to be the case that you were expected to be able to solve a crossword using a dictionary and what was known as a standard reference bookshelf. There was this idea of a standard reference bookshelf. The British standard reference bookshelf would include the complete works of Shakespeare. And it might include a lot of classic novels and some Greek myth and things like that. And it would include Brewer's Dictionary of Phrase and Fable and certain sort of big ones. But then everything else you were expected to either know, because it's general knowledge, or you can look it up on one of those. And nowadays, that's out of the window. And the Magpie was one of the first magazines to introduce a thing called the research warning, which seems really quaint. But we'd put a badge on the puzzle saying, this cannot be solved using your bookshelf. The research warning meant basically, you've got to Google this. And it's now, more or less, all the people setting the puzzles are allowed to use any theme they like, on the grounds that people can find out any fact they like, anything. And it is the case. There's nothing which is too obscure. So that's been liberating in a way? Hugely. Yeah, it's amazing. Well, for me, especially because I never had any general knowledge. So I only ever looked everything up. But didn't you accumulate any knowledge from looking stuff up? Or did you just forget it once you've done the quiz, set the quiz? No, if people ask me a question now, and I know the answer, it's because there's been a crossword on it. A lot of people knew the answers before they started crosswords, I'm the other way around. Okay. Keep going. You were still answering a question there, no? That's it, really. I mean, that's kind of what I had to say, that the crosswords are designed, they're designed such that you get given the information you need to research the theme, but at the rate which is appropriate. And occasionally, people can fast track to a solution by using some bit of information. And on the whole, they feel a bit let down if they've got there too fast by knowing something too early. A good puzzle setter gives you the information sequentially in the puzzle, such that as you're solving it, you get a bit more and a bit more and a bit more. A breadcrumb trail. A breadcrumb trail, yeah, to follow. And finally, you've got enough that you can Google effectively and go straight to the point. Bam. Sure. But you also

feel like you've achieved something, getting to the point where you can Google. Yes. You feel

like you've been given the correct search terms, and you get it, and you're happy. And especially you're happy if before that, you've been flailing around searching for something adjacent, but not quite right. But you know you haven't done the work to get the correct information yet. And then by the end, you've got it all. And then you're in. That's a good puzzle ending.

Yeah. Can I talk about one? I mean, this will go out after the solutions have been published. But just to put some flesh on this bone. So one of the puzzles which appeared in the last Magpie time recording gives you standard cryptic clues, which are not easy to solve. But assuming you can do that, it doesn't tell you where to put them in the grid. So they could appear somewhere in a row, for example. But they can appear anywhere in that row. And in fact, they wrap round. So you've got to solve that first. And then there are two entries hidden in there, two clues in there, or two entries, or the two words that appear in that, which tell you something about a building. And it turns out that if you anagramise those words, you end up with a phrase that appears on the building. But that's all you get. So you've got to work, you definitely won't be able to Google it by searching buildings. Did you personally say this one, by the way? No. So once you've done that, then you have to start finding hidden words and colouring in things until the grid has a pattern on it, which is the same as the pattern that appears on the roof of this building. But the point is that, and the whole puzzle is called piece de resistance, which is kind of, I mean, as with all these things, when you know what the answer is, you can see the link. But the point is that you can't backwards engineer from the clues you've been given without working through all of those elements.

I've got, you know, a little side hustle for you. So we'll come on to this in a second, but a little side hustle for you, Chris, okay? You should be doing this already, but if you were, you wouldn't be able to tell me, right? Which is, you should be a recruiter for GCHQ, right? I've already stumbled across something there. And famously, of course, the telegraph and so on were drawn from, wasn't it? Puzzle, crossword puzzles and solving them, so on. So I hope you're getting commission on sort of, that you're a recruiter for GCHQ. I can't talk about that. About that aspect of my career. No, it is true that a lot of crossworders have nostalgia for the good old days when you could solve the telegraph crossword and get recruited by some amazing spy network. But in fact, nowadays, you've got to be a mathematician, unfortunately. Right. Being good at crosswords doesn't cut it anymore. So they haven't been coming to me. I have never been recruited by GCHQ. But it just sounds like you've just called all crossworders bad at maths. That's what it sounds like to me. No, a lot of crossworders are good at maths. And there are quite a lot of GCHQ employees who are quite good at crosswords. But unfortunately, being good at crosswords isn't adequate.

It's perhaps necessary, but not sufficient.

It's no longer going to just land you the job. So which makes me think two things. First of all, when you were talking about those clues, I mean, what kind of person wants to engage with that? I think it's quite an unusual kind of person. Sensible. Yeah, the three of us. All three of you. Yeah. Right. OK. Three quarters of this room. Yeah. But we've got sample bias here.

Chris published that and Nick and I both solved it.

We've got extreme sample bias because the number of people in the world is under 300. One percent of all MacPy subscribers are in this room now. It shows you why I shouldn't be doing it because I said it was 10 percent. But also it makes me think that it's a bit like when people talk about IQ tests and this person got really IQ, high IQ, that one of the things that they say, well, it doesn't mean anything other than they're good at IQ tests. Are you asking whether crosswords is a transferable skill? Yeah. Well, yeah. That maybe just people who do crosswords are only good for doing crosswords. Yeah. No, I think that's more or less true. It doesn't make you better at washing up or growing potatoes.

We have done a podcast about that before. So I wouldn't mind steering us back to the thing of. And I think just I want to talk about how Google works. Yes. The answer is actually no one knows. Google won't tell you. They won't tell you exactly how Google works. However, from what I can tell, it works like this. First of all, Google tries to visit every website in the world that it's allowed to look at. So it does that by following links to websites. That's all it does. It just it doesn't have some big repository. It just finds them because it links to a website, which links to other websites. And gradually it visits every website in the world. And then it the best way. Well, basically, think of it like this. It looks at all the words on every website. And if you imagine a great big load of rows, which are all the websites in the world and a great load of columns, which are all the words. And there's a little X if a given website contains a word. That is not exactly how it works. But if you think of it that way, you'll get a sense of how Google does what it's meant to do. So when you search for a word, Google can just look down the column and find all the websites with that word. When you search for five words, it can just look down the columns for those five words and find the websites that have all of those five words and then does something proprietary and complicated to rank them and show you which one it thinks you're most interested in. And then it shows you the ones that it's paid to show you, etc. But the point is that that's more or less what it does. So if these things that are unGoogleable fall into several categories, first of all, because they're not words, because they're smells or they're pictures that don't appear anywhere. Obviously, Google does have a reverse image search, although in practice, it'd be very difficult to reverse image search a kind of bad photocopy or something. So first of all, because they're not words. But secondly, if there isn't a website which links all of those words together. So this goes back to Chris's example of what links the dude from The Big Lebowski and Vladimir Nabokov. Is that there are websites that link white Russian to both of those. You can say you'll find out that Nabokov was a white Russian. Yeah. Find out that dude likes white Russian. Yeah. But unless you know the white Russian bit. Yeah. That's the only thing that that puts those websites in the same set. So that's what you're trying to find. You can't. There isn't a website which already links those two together. That's so that's why that's Google proof. Those kinds of what's the missing association between these these concepts, those kinds of questions, they can't be Googled because there isn't a website which has all of them at once. I came up with my own. Actually, do you want to have a go?

I'm mad for it.

So what actor what actor links Tom Hanks with Cab Calloway?

I don't know who Cab Calloway is.

He was the guy, the Mini the Moocher guy. So I will give a clue and tell you it is an actor. So we're basically looking for an actor who Tom Hanks has been in the same film as and Cab Calloway is Tom Hanks. Tom Hanks has been in the same film as Cab Calloway has been in the same film.

Well, I'm just going to I'm going to take a stab here and say it's going to be a Blues Brothers thing. And it's going to be either James. So I'm going to say James Blue, she or Dan Aykroyd.

Yeah, but both both good. But it's I don't I don't think Tom Hanks has Tom Hanks been in a film with Dan Aykroyd. I'm not sure. The answer could be it could be Carey. What's her face? It's actually John John Candy, although there may be other answers. And that brings me on to one of the things, one of the problems, which is that it's very hard to demonstrate uniqueness, isn't it? And I don't know how you do it in crosswords. But, for example, why is well, they're both men. Not an acceptable answer to what links Vladimir Nabokov and the dude or any number of things.

So I think this almost puts a second question on top of our first one. First, how do you make a question ungooglable? Then how do you make a question that is both a good question and ungooglable? So, for example, smells has the challenge that something smell very similar. Yeah. Can you definitively say this? This thing that you have smelled is the smell of X unless you're actually smelling a pot of mustard, for example.

Even then it could be could it be wasabi?

Yeah. And I think the same goes for the links. Now, that's not to say this kind of question can never be good. I used to do a pub quiz in Tufnel Park, the Tufnel Park Tavern, where it's quite a tough quiz. That one. Is that a joke? That was a joke. That is really quite a tough quiz because the quite a Tufnel quiz. Yeah, that was another joke. Sorry. Go on. There was a picture round and the ninth picture in the picture round would always be something really ambiguous. I remember one week it was a photo of an egg and the answer was pigeon egg. If you didn't say pigeon, you were wrong. Another week it was a black and white grainy photo of a glass with some liquid in it. The answer was carrot juice. But there's going to be a linkage, isn't there?

Isn't there going to be some way in which if you've got the rest right, you know that there's some kind of.

No. We got the pigeon egg one by pure chance. And it was the first time I had ever been in that pub and seen someone get the ninth picture round. So, you know, as a gimmick, it can work in in moderation. But you couldn't make a whole quiz with that. Yeah, but it makes it Google proof in that instance.

But I think this is one of the things that is sort of machine proof in general, actually, is is this the sort of one way encryption processes which are also related to on Google or actually making things immune from machine learning attacks is if you're able to take things which can only you have to kind of know the input before you can work out the output, for example. So from this picture of the pigeon egg, you it'd be very hard to look at it and go, that's a pigeon egg. But if you if you started with the thought of a pigeon egg, you'd be able to recognize it. I don't think many Toron encodings. Exactly. But I don't think that's. And I feel like what's great about the White Russian example is it feels it feels like that. It feels like if you light upon the White Russian, it feels very clearly like the key which brings together those two concepts in a way where you just know.

You can get the answer and feel I have got this right. Yeah. Whereas you're never going to write down pigeon egg and say, I am confident I'm getting this point.

Yeah, this is. And but but also it doesn't sound very satisfying, the pigeon egg. Oh, it was very satisfying when we got it. How did you get it get the pigeon egg specifically?

Well, we thought it's not going to be an egg. And we thought, let's let's guess an unusual type of egg. Guess a pigeon. So it can't have been that satisfying because you guessed it.

No, but who am I to deny your emotions in that moment?

We've been going to this quiz for months and we'd never managed to guess one of the ninth pictures correct. Fair enough. And no one else had either.

Can I just ask if this includes heat? Of course it does. Yes. Ah, he's smart. Fraser's already made his guess.

He can't change it now. No, that was just asking. No, I did make my guess.

70.8 seconds.

But you didn't give enough information in the question. Well, hold on, you're already complaining about the answer before. He gave enough information for me to get it. You don't know you're right yet.

I'm going to go with 67.4.

Oh, that's not fair. You've just gone just below mine.

But also, by the way, just say even if I'm wrong, because I'm annoyed now, even if I'm wrong. I'll tell you what, Fraser, I'll give you a multiple choice. You see, because the multiple choice questions is another way of making questions which aren't Googleable, but which are satisfying because you kind of feel like, oh, it's probably this. OK, I'll give you a multiple choice. Before you do, let's say even though you may or may not have understood that element of it, right, you could still be wrong. And if you're wrong, you're as wrong as I am, if I'm wrong.

Well, no, it's a numerical question, there's definitely, you know, degrees of wrongness. I could be closer than you. But would you get a mark for this? Depends on the quiz master.

Fraser, what if it turns out Fraser's right? Is he going to apply this same degree of rigour to his own? If Fraser's off by half a second.

I just said to the nearest half second, you see, and I thought it was only fair to give him a multiple choice. Go on then, hit us with the multiple choice. Because the nearest half second is pretty harsh, if you can just...

I'm going to just compel people to go for the multiple choice closest to what they said.

Give us the three options. So the multiple choice is going to be 39.5 seconds, 47.5 seconds, 58.5 seconds or 65.5 seconds.

Well, Tom and I are both going for the last one. Fraser is going for the first one. What's the answer? Yeah, he said 19.

Who made you deputy quiz master? Off you go, Fraser, take your pick. Stop getting up on me, you two. So I forgot what the question was. It doesn't matter, pick one of those numbers. The point is, you're so close. If only you'd gone for 47 and a half seconds. 47 and a half seconds, so two gold medals for 47 and a half seconds work. I think that's pretty, I think it's like it's quite a cool fact. And it's the kind of thing where... I think he works more than 47 and a half seconds. But I can't prove that. Yes, I can. He ran three times in the 100 metres and twice in the relay. See, so that's it. So in total, he did, whatever, 500. Why is that? OK, I get it. Why isn't that Googled? Yeah, yeah, because you have to add up the information, you have to assemble it. Well, it is kind of Googleable. But the reason it isn't Googleable is I had to do some sums and his split time in the heat, I couldn't find at all. So I've just taken his split time in the final for the four by 100 and I've duplicated it and then rounded it. So it's not Googleable because people don't actually care how long he spent in. No one cares about the heat. No one cares about, they just know how much, how long he spent in the final. I like that.

And this is a bit like, so another category is things that are laborious.

But also...

And time constrained. Laborious and pointless. So nobody's worked out how to do it. Yeah, but also... And there isn't, you have to do it yourself.

Because why would you ever want that information unless it's to get an answer in a quiz? Well, that's the whole point of trivia. You know, you don't want the information except to answer the quiz. And then the interesting thing to me is when do you afterwards think, well, I'm glad I know that or I'm glad I found that out? Sometimes you do and sometimes you don't. And that's the tough, tough one. Because if you get something, if you ask someone a question where they, A, can't work it out and B, couldn't care less. So something like, you know, what is the weight of the Empire State Building in pounds or something? Well, no one cares. So they're just going to make something up. And then if you give them multiple choice, they're still not going to care. Because it's just too far removed from anything you can think about. And then they won't remember and won't be bothered. But if you can get it where it's, where afterwards you think, oh, that's kind of interesting.

Like the gold question, which always blows my mind. Go on. Well, if you took all the gold that's ever been mined and you put it into a cube, what would be the length of the side of one of those of the cube? You need to start doing something.

I've heard this before, so I won't say.

It would be about as big as this room would be my guess. So I'm going to say about the cube. All the gold that's ever been mined. Yeah, ever been mined. And what do you reckon, Fraser?

Wait, you thought, and what's the one in a cube? What's one side? Yeah, you only need one side to know what the rest of the cube looks like. Don't be smart. Don't be. No, but I just want to have a guess. 20 metres. Yeah, pretty much that. I mean, I can't remember the exact figure, but it's not. What it isn't is a kilometre.

So I was out. I was out by, yeah, I was out by a factor of four. I mean, you're close.

The point is, we should be thinking in terms of, OK, it's the size of this room, a block of gold, the size of this room, or maybe twice the size. Mental. You were out by a factor of about 100 per cent. Anyway, what is the, how are we doing for time, Fraser? Because I think this has been really interesting. We've covered quite a few ways you can Google proof your quiz, I think. We haven't talked about sort of, well, we have talked a bit about these kind of one way processes. But there was quite an interesting example I saw in this pub quiz, which I looked at. So I looked at the questions and they had this thing where they asked, they fed phrases into an AI image generator. And you had to work out what the phrases were.

I love the thought of feed feeding phrases into AI image generators. Oh, phrases. No, it was phrases, Sam.

So, you know, it'd be like a rolling stone gathers no moss. And it would be a very impressionistic sort of picture of that, where if you kind of got it right, it would kind of make sense. So I thought that was quite clever as well. But again, that's the sort of one way process thing where you can't unscramble it. But it's obvious when you've got the right key to get there. So anyway, I've got one other thing we can mention, but it will be a closing down thing.

Yeah. So and indeed we are coming towards the end. We could end on what I've got here. Does the closing down thing involve all of us? Yes. It's a sort of quiz, actually. So we need to bring this to some, well, not just some sort of conclusion, but a conclusion. Can anyone help me out? I've got nothing.

I'm going to slay you guys. OK, I'm looking forward to it.

So first of all, this one you sort of have to know. But again, you can see how this knowing being able to type this question does not enable you to get the answer in Google. But a literary term used to describe the idea that mentioning French fries in your fiction story implies the existence of France. For example, in Lord of the Rings, when the orcs say looks like meats back on the menu, boys, it implies they know what a menu is. So the kind of implications. Anyone know what the term is for that for the literary term for that? Oh, let's pause for a minute. Let's just take our time. So you can come up with examples, but it's very hard to Google that and find. So there is a term for when you do. Yeah, it's not you wouldn't you wouldn't get it, to be honest, unless you knew it. It's just a good example of that kind of hard to you might know lots about this thing, but you can't.

I think if you had time, you could Google it. But thinking of how to Google it is a challenge in itself.

So the answer is orphaned etymology. Now they get a bit easier now. Orphaned etymology. Orphaned etymology. Yeah. So why is it called orphaned etymology? Here's one question. Because there's no parent etymology for the word to come from. Thank you, Tom. So this one, this is another high, high rated question. Doo doo doo doo doo song. I would associate it with a sort of seafaring style, if that makes any sense. It's also pretty fast paced with the following taking four to five seconds to play through. Doo doo doo doo doo doo doo doo doo doo doo doo doo doo doo doo doo doo doo doo doo doo doo doo doo doo doo doo. What's that song?

But hold on. So do the do's just represent a do, or does the do represent? What does the do represent?

That is as written. I've read it out as written. With the doo doo doo being short ones and doo doo being the long ones. Yeah, I don't know. Is it what should we do with the drunken Satan? No, but astonishingly, this was solved. The answer is the horn pipe. And actually, if you have the home pipe in your head and you go to do to do do. Do you actually work? So a couple more. What is the name of ankles? But above your hands, just at a wrist. Yeah, yeah.

You know, that's actually the second time that that I've heard them describe this hand ankles. Right. Friend of mine did it completely tried to start off and just kept talking as though he hadn't just called wrists, hand ankles.

What's wrong with hankles? As an aside, in Portuguese, or certainly in Brazil, these are your hand fingers, and those are your feet fingers down there.

Awesome. Right? Cool. So the last one, which actually makes quite a good, sort of makes quite a good pub quiz question in itself. This is clearly someone who'd drawn a blank, and it says, tip of my tongue, a profession. What's, I'll read it verbatim, what was that profession called, the one that had something to do with mass arsing, and it sounded like a dinosaur? Something to do with mass arsing, whatever you think that might be, and it sounded like a dinosaur. The answer is really good, you'll know if you get it. Is it, like, being a chiropractor? Yes! The answer's chiropractor. I think it's more that a chiropractor sounds like a dinosaur.

I was trying to go for, you'd make dinosaur-like sounds. Oh yeah, no, so it's mass arge, and it sounds like you're a dinosaur.

So anyway, those, I think that's a place to go if you want to find ungooglable things. Yeah, I like that.

It just makes you think that that's to do with Greek, that's why that one, because the guy was getting mixed up, because I, a lot of, well, I have a lot of dinosaur names in Latin rather than Greek, so.

Sorry, but chiropractor sounds like a dinosaur. If you can have a velociraptor, you can have a chiropractor.

But in crosswords, the reason, a crossword clue, a cryptic crossword clue especially, is ungooglable currently, because it's designed to, it's designed, it's like those things, it gives you a definition and asks you for the word. So you're given a definition, and you need to work backwards to the word. If you can search the dictionary for the definition, which kind of Google gives you that option, then you can get the answer. And you can type into Google, you know, a crossword clue, something or other, and it'll give you a bunch of, it'll give you answers. And so it's ordinary non-cryptic crossword clues, you can just kind of Google them. And it probably won't be that long, although it's hard, for computers to be able to solve cryptic clues.

I agree, there'll be AIs to solve them. But I don't think they'll be Googlable, they'll definitely need to develop an AI to do it. Yeah, they're working on it.

Sorry, Fraser, we need to stop. Okay. Although I think it's interesting, at the moment that I felt was a natural conclusion, that the person who sets puzzles and is quite tenacious with hunting down information, kept going. Chris, so we need to stop, we need to stop. Okay. Do we want to finish off anything? Yes. Go for it. Well, you've got to have the question. You've got to have your question. Yeah, we want the Fraser question. I don't always have a question. I'll tell you what your question is. Tell me what my question is. So that I can answer it. Yeah, brilliant. You're doing a me. Yeah, your question is. I always then change the, yeah, go on. Your question is, what is your favourite quiz question? Yeah. I'm not sure that it is, but anyway, I shall indulge you. Um, before I do actually, I'll try and make this quick. Has anyone, apart from Chris, has anyone set a quiz themselves? Yes. Like a pub quiz, right? I have, I think only once in a pub context. It was a total disaster because I was not in the UK. I'm not gonna say where I was, but I was, a lot of the people in the audience were a lot thicker than I had realised. And so what I felt were quite simple questions were actually way beyond a lot of people in the room. And they really weren't that difficult. And so it almost turned into like a, not a riot, but I mean, it got a bit unpleasant, you know, because some people getting bored and listless. Like our listeners probably. Get to the point. That was the quiz that I set. I pitched it too high. You two?

Are we doing Chris's question? We are, we're coming to that in a second. Your favourite question. I set a couple of pub quizzes in the sort of student bar at university. I've done a couple of the classic lockdown pub quiz on Zoom with friends. I probably pitched them a little too hard, but I try to mix in easy questions as well so that everyone can at least get some answers and feel satisfied.

Yeah, makes sense. Good approach. Nick, anything to say about the quizzes?

No, not really. Although just going to things where you feel like there might be a, you can sort of guess. Um, I did a quiz round about the Holy Roman Empire during Zoom, the Zoom quiz period. One of the questions was, uh, I think now I might get this wrong, but, uh, Frederick I, who was one of the early Holy Roman emperors, who was a wise and kingly man with a long red beard is now known primarily by a different name. What is that name?

I know. Can I say, or are you ready?

Do you want to have a guess? Is it, does it mean red beard?

Yeah.

Then I can guess. Yeah, well, go for it.

It's Barbarossa. Yeah. But I mean, again, that's one of those things where, you know, it's sort of, you could guess it. You could guess it. Yeah, yeah. It's a bit obvious.

No, but it's almost like a, yeah, it's not a total guess though, is it? And that's that nice kind of partial sort of thing. I like that. So, yeah, we'll look. So, Chris, what's the favourite your, no, we'll come to you last actually. Tom, your favourite ever quiz question?

So one of my favourite quiz questions that I've done in a pub quiz was a music round question. I think this fits with the theme of being unGoogleable because often it's just a snippet of the beginning, hard to Google. And it was TV show themes. And I never heard it before, but I thought to myself, oh, this has got kind of, you know, actiony vibes. I could picture this with like a helicopter in the background. So I guessed Airwolf. I've never watched Airwolf, but it was Airwolf. And we got the points and we won the pub quiz that day. So I'm very proud of that one. Nice.

It feels, feels choppery. Probably by Mike Post. He wrote every action theme.

Hopefully that wasn't part of the question. No.

Let's keep going.

Can we use your Barbarossa one? That's your favourite one. Yeah, done that. Yeah, go on. I can't be bothered with mine. Chris. Okay, how about, this is Jamie's one. Which country has the most penguins? Who's Jamie? He's my son. South Africa. Tom's going South Africa. I thought it might be Argentina. Nick's going Argentina. So my initial thought was Argentina, but this is going to be a trick question. It is, you're right. I'm going to say... Trick question and it's fun. Yeah, I'm going to say United States of America. So none of you are right and the answer is Great Britain. Is it the zoos or something? Exactly, because of some islands near... South Sandwich Islands. Well, in particular, the Falklands Islands. The Falklands and some islands near...

I don't think they're technically part of Great Britain though.

Well...

They're an overseas territory.

I'm going to move on anyway. You can argue with me, but I'm not going to back down. I can see that punted look in the quiz master's eyes. That's another way of making a thing ungoogleable, by the way. An answer is an answer and that's the correct answer. And if people are going to argue with you, they're not going to get any points. I think that's a good point to cut this all off. Okay, very good. Fantastic. I really enjoyed that. Went on a while, but I think that was a good one. Let's stop there. Thank you, as always, for listening to the Cognitive Engineering Podcast. I'm Fraser McGruer. We've been here with Chris Lear, Tom Spence, and Nick Hare of Aleph Insights. Until next time, goodbye.