Algorithms + Data Structures = Programs - Episode 52: One Year Anniversary!

Episode Date: November 19, 2021

In this episode, Bryce and Conor celebrate their one year anniversary by talking about 0 vs 1 indexing and their favorite episodes so far.Date Recorded: 2021-11-13Date Released: 2021-11-19Conor & Bryce’s Favorite EpisodesADSP Episode 39: How Steve Jobs Saved Sean ParentADSP Episode 21: Galaxy Brain Programming LanguagesADSP Episode 25: The Lost ReductionBryce’s Favorite EpisodesADSP Episode 13: I’m an Excel Wizard!ADSP Episode 17: Special Guest Sean Parent!ADSP Episode 18: Special Guest Sean Parent! (Part 2)ADSP Episode 19: Finding Your Way in TechADSP Episode 37: std::inclusive_scanConor’s Favorite EpisodesADSP Episode 29: From Papa John’s to Google (Part 1)ADSP Episode 47: Combinatory Logic!Other LinksThinking, Fast and Slow by Daniel KahnemanPeak–end ruleIntro Song InfoMiss You by Sarah Jansen https://soundcloud.com/sarahjansenmusicCreative Commons — Attribution 3.0 Unported — CC BY 3.0Free Download / Stream: http://bit.ly/l-miss-youMusic promoted by Audio Library https://youtu.be/iYYxnasvfx8

So, uh, listener, oh boy, I have to say, I'm not, I'm not devastated. Welcome to ADSP, the podcast episode 52 recorded on November 13th, 2021. My name is Connor. And today with my co-host Bryce, we celebrate our one year anniversary of ADSP the podcast by talking about one versus zero indexing and our favorite episodes from the past year. So I think you might have made a tactical mistake. What? How? This episode might turn out to be an episode about counting.

How many episodes do you think we've published? Oh, I know. We're zero index, bro. But wait, wait, wait. Here. Before we do this, I've just realized. I'll be back in a second. Hold that thought. Hold that thought. Connor is departing the podcast recording closet because, yes, that's how Connor records the podcast, in a closet. Was he going to retrieve a drink?

We're back. We're back. Folks, we're back with the black cherry. I don't know why I'm showing Bryce because no one can see this. And here we go because it's our one-year anniversary episode, which Bryce is going to have some thoughts on in a sec. This could end up being a long-winded episode. Here we go.

Black cherry buble coming up. Woo! That's how we get it done. That was amazing. Bryce, share with us what you would like to about counting. Is it our one-year anniversary? Because this is going to be the 53rd episode, not the 52nd. You know what?

This is actually— We're zero indexed. I had like three things sort of planned to talk about in today's episode. Wait, wait, but how is this possible? Because episode zero was published last year on November 14th. And we've published one episode a week. That's not true. It was not published on November 14th. That's what the app says. And that's what the the the app says and that's what the

write-up says it says oh no date oh date recorded yes date recorded november 14th and why folks to take a date date released november 20th correct and what is today bryce today's the 13th we'll release it on six days from now six plus 13 you can do it buddy the 19th which is one day one day on a second this is going to get really philosophical i'm excited to this is so this will be our fourth thing that we're talking about and we'll be doing it first just hang on one moment i'm hanging on there have been 52 weeks since we started this podcast. It's inclusive, bro. But then how will that be?

Because that will be the 53rd episode, episode number 52, because we're zero indexed. How does that math work? Disclaimer. I know that I do have a math degree, but in my defense, I was not a very good student. So this is only 10% tangential. I've really been trying to, I think about a lot of things, but one of the things I think about on a weekly basis is whether zero indexing should be the default in array languages because for those that don't know, in APL, one of the biggest mistakes that's sort of acknowledged

is they had something called quad IO. So quad was like a little rectangular Unicode box, or I guess it wasn't Unicode at all times. And then IO or whatever the one or two letters that followed it would be like a special function. And it stands for index origin, which by default is one, but you can set it to zero if you want. And so it acknowledges like the biggest mistake, not that they chose zero or one. It's choosing one is that they think is the right thing to do.

It's that they let you, it's basically like a global thing that affects the way algorithms works. Like for instance, IOTA, if you have index origin set to one, IOTA five is one, two, three, four, five. But if you have index origin set to zero, it's zero, one, two, three, four. So like the, you know, the state of this quad IO affects the way that certain algorithms work that are dependent on quad IO. But the thing is, is there are, so in J, they got rid of quad IO and they just defaulted sort of index origin implicitly to zero. So if you go IOTA five, you get zero, one, two, three, four in J. And it is my evolving opinion that actually one indexing is the correct default

in array languages. Because so often you're doing things where the exact problem that we're running into here, where like the fact that we have episode zero, one, two, three, and four does not correspond to first, second, third, fourth, and fifth. Like our first episode is episode zero. So we have this mismatch, which is very, it bothers me. I think it's okay in like, you know, and stuff do you have thoughts on i i no i i i i actually think it's probably not okay in c++ i mean i i zero index everything you know if i'm sending an email with a numbered list i zero index it for the simple i i've said this before on the show i'm sure uh you know bryce's law i'd rather be consistently wrong than inconsistent. That's fair.

I live my life by that. That's fair. And so because C++ is zero-indexed, and that's the main thing I do, or at the very least, the majority of languages and programming things that I do are going to end up being zero-indexed. So then, therefore, everything else in my life are going to end up being zero indexed. So then, therefore, everything else in my life is going to be zero indexed. I'm not saying that it's the right thing. I'm just saying I'd like to be consistent.

But I think that zero indexing was probably a mistake because of the hurdle to explain it to the uninitiated. Just like off the bat, like if you are explaining something to somebody who has no programming background whatsoever, if you have to also explain zero indexing on top of that, it just adds additional complexity. Isn't it, I thought, and maybe this is a mistaken impression, that the computer science or software engineering society at large had agreed that for some reason that I don't know the motivation for,

I don't know the reason that zero indexing is the correct thing. Like, do you know where it comes from? No, no, no. There's, there's a reason for that. And I mean, the reason is if you have, I mean, just, just like think about it conceptually in terms of the, the actual memory under the hood and in terms of like pointers, you know, if you have a pointer to some array or if a pointer to some like contiguous sequence of memory that represents an array, what is the address of the first element? Yeah, that makes sense.

Well, the address of the first element is just, you know, that pointer to the array. But I don't think that that physical reality of computers that like, you know, memory addressing works that way. I don't think that that means that that's how we should have

logically represented it in programming languages. And I mean, you know, like if you show, if you try to explain the notion of an array to somebody who knows nothing about programming, just like think about how you'd do it. And think about how much easier it would be if you did not need to explain that there was this notion of a zeroth element. And I think it leads to a lot of common off-by-one errors. I think when you have a zero zero index based world, you tend to do half open sets where the, you know, you're, yeah,

you have a, not sets, ranges. So when you have zero index, when you have a language with zero indexing, I think the tendency is that you'll describe ranges of integers um with the the first element being inclusive and the last being exclusive um you know yeah you know the elements from like zero to n where it's not including the element n right um and and just like little disconnects like the size of an array. Right. If I ask an array, like, what size are you? I cannot use the answer to that question to get the last element. I need to subtract one. Yeah.

And probably thinking about it, one of the big things too, which is why probably a motivating reason why Ken Iverson chose to be one index is it doesn't map to English. When you say, oh, I need to grab the third element from this array, you need to use index two. And that's completely counterintuitive. And this is coming from a guy who named functions verbs and operators conjunctions and adverbs. And like, so he was, he was trying to map computer science and his language to literally English. And I think that gets to, you know, the crux of this decision, whether to zero index or one index. And I mentioned it a little bit a minute ago, which is, it's a question of do you model

the physical reality or the logical concept? You know, when you're building a programming language, do you choose a representation or a design that most accurately reflects how the hardware actually works, how the underlying facilities actually work? Or do you select the design that is most natural to programmers, is most natural to humans? And I think in this case, zero indexing is the option that you pick if you want to model how the computer actually works. Whereas one indexing is what you'd pick if you want to model how we think or model mathematics or, you know, model physics, et cetera.

Yeah. And I think that's one of the reasons why a language like Fortran, which was designed for numerical and symbolic computing, picked one indexing, whereas a language like C that was intended as a systems programming language

that very closely and faithfully maps to the hardware and the operating system picked zero indexing. And I mean, obviously c++ i think could not realistically have have chosen a different option um i mean one couldn't have chosen a different option from c just for like regular raw pointer access because that would have been wildly incompatible um but also i do not think it would have because of that it would would have been wildly incompatible. But also, I do not think it would have, because of that, it would not have been realistic for us to have our containers, our abstractions, not, you know, built in language facilities, but our abstractions be one indexed.

It would be really, again, it's better to be consistently wrong than inconsistent. It would be a very unfortunate world if std vector was one index, but pointers were zero indexed. Like, ugh. Yeah, yeah. And like, then what do you do about iterators? Because, you know, pointers are iterators. So that kind of suggests that they all have to be zero indexed. So it's almost like the high levelness or low levelness of the language

should almost dictate which is sort of you know we just talked about that last episode is that uh apl super high level c++ super low level like the classic example that i run into in j or like an array language that's zero indexed is say you want to like the algorithm, which now exists in C++ 20 ranges, you're not using the zero indexing. Take zero takes nothing, and take one takes the first element. But in array languages, a lot of times you want to find the index at which the first element occurs, and so it's sort of like a take while.

You want to take up until that element you know, element and including that element. And if that index that is returned is zero indexed, you always have to add one to it in order to get take to work sort of inclusively. And yeah, I've always sort of whenever that happens, I'm like, this seems wrong. But anyways, should we get back to episode 53? Are you still struggling with why, I mean, 52 weeks we record or release at the, you know, either beginning or end of the week? And then 53rd is the? Yes, yes.

No, I was mostly being dramatic earlier. But it did, when I woke up this morning and was looking through the list of episodes, I was like, hang on a minute. Did Connor mess up? Oh, no. And did we already have it? And then my brain turned on. And this is the thing. I always state in the introduction when the podcast is recorded because I – this is like a thing that probably applies to nobody in the world except for me but I like I get in the habit of adding a podcast to my list probably the most recent one

that I went back and like listened to every single one I don't know if it was the most recent but I definitely did Cognacast which is a closure podcast so I have a very hard time like jumping into episode 120. I have to go back to episode one or zero and then watch them from the beginning. But because that's 120 episodes and they release once a month, it dates back all the way to 2012. And I like knowing at what point in time, was Rust just coming out or was this pre-Rust or what's the context that we're in here um which is why i always state the date that it's recorded because i think that's actually more important than when it's released

like we're not impacted by when we release we're impacted by uh when we're recorded all right though so uh that was what 20 minutes of uh bryce and i talking about uh counting sounds about right i i i told you this was going to become the episode about counting um but yeah there's two or three other things i figured we could chat about so i'm not sure i gave bryce some homework to um pick what he thought were the top three top five episodes i didn't really do my homework well i mean like i did the homework but like i started doing the homework about three minutes before I jumped on the Google Meet call. I mean, I haven't fully ranked my top three or five either, but I definitely know. I know one of my, definitely one of my favorites is episode 13.

Do you know which one episode 13 is? I mean, I'm staring at the list, so I just skipped to 13. I wanted you to guess. i don't know i wanted you to guess i wanted you to guess what what you thought my favorite would be okay uh i'll stop looking at the list like i mean i know what that one is now because i just looked at it but it's uh the i'm an excel wizard yeah yeah so so i like that episode um because i think one Because I think, one, it tells the listener a lot about Connor. And two, I like it because I think it's very accessible to a wide audience because almost everybody uses Excel, even if you're not a programmer. Yeah, so that's one of my favorites.

Give me the next number. Definitely, I think, my favorite of the ones that's just you and I. Really? I don't even think that would have made, I don't, that wasn't in my tentative top five. Although, I don't even really remember it that well, because I think I was just in a haze because i was too excited um to be talking about excel i'll have to go back and re-listen to it um so episode 17 in 18 the first time we had sean parent on i didn't know those numbers i think that was that was just magical i mean every time we have sean on the podcast it's just magical what of the of i think we've had i think we've we've recorded with him three times and that's been turned into i think eight or nine episodes which of those um because definitely uh you know a couple of the sean parents are in like my top five um but i tried to only choose one and uh of those it's hard i i i really did like the first time we did it. I don't remember which of the like do guests well and like you know sean sort of knew the format then and

it's just like just awesome stories just like sean has just had it's like sure yes sean has had an interesting career but sean has also had like an interesting career that's just been he's just a magnet for these odd circumstances that just lead to great storytelling. Yeah. And so I think you're referring to episode 28, which is the well, actually, we won't ruin it, but it culminates in sort of Steve Jobs, you know, calling the CEO of Adobe with some things to say. So if you haven't listened to that episode, I'd probably say my favorite just because I had not heard. So several of the stories when we've had Sean on, I had heard over a CPP con,

which it's amazing to hear it for the second time because his stories are just so wild. But episode 39 where Steve Jobs, it's called How Steve Jobs Saved Sean Parent. Once again, I won't give away because I had no idea

that what direction that podcast was going. You know what? I didn't know that story at all. That one actually might be my favorite. I forgot that he told that story on that episode. Yeah, that one is.

I won't give anything away, but at one point, I legitimately thought he was about to say that a couple people died. And I like straight up was like, Sean, did everybody live? Because like we can't let you talk about how you potentially killed some people in the past. And he's like, and then he said, no, no, no, everybody survived.

And then my blood pressure sort of dropped a little bit. I was like, okay. But yeah, I think that was just because I had not heard that story and um it's also once again yeah just sort of and that's the thing is that at some point it was before this podcast started that I realized about um Sean is that he's this like peripheral character in sort of all these movies that have been made about you know Steve Jobs and stuff is that like he knows the people that are being portrayed in these movies. Like, he's, like, one person removed, and if he had chosen to stay at Apple a little bit longer,

like, he would have been one of the people in the movie, potentially, which I just think is, I don't know. When I watch, like, the Steve Jobs biopic films, I don't think of it as, like, oh, yeah, like, I know people that are adjacent to that, but that's exactly what, I don't think of it as like, Oh yeah, like I know people that are adjacent to that. Um, but that's exactly what, what I mean. I think it's even, it's not generous enough to say that Sean's adjacent to it. He, um, yeah, he, he's been working at Apple and Adobe for anyways, enough about, uh, enough about Sean. Um, what's your next, uh, or do you have, do you have more episodes?

I liked episode 19 finding your way in tech where we sort of gave our origin stories um again it's a it's an episode i like because it's sort of like accessible to any audience um and i i hope uh i hope was helpful to somebody um and likewise that the episode where we talked about should you drop out of college damn all of your episodes are so wildly different than than the ones that are on my list this is this is interesting well okay so i do have a couple that are that are more uh i think more what you might pick um so episode 25 the lost reduction um all i remember about that episode was being just so so excited and enthusiastic when we were recording it i think that i think that is actually our longest

which says something it's our longest episode oh what no no no it's um yeah it's tied with the ben dean and tony veneered no no it, no, no. It's got it by 19 seconds. Got it by 19 seconds. And I can't say I liked, but I think it was good. I think episode 21, Galaxy being... Oh, yeah, yeah. Where you quizzed me. That's top five for me for sure.

Unprompted was uh was pretty great um and um i liked i liked the inclusive scan um episode the episode 37 um and in general i like all of the times that we've talked about scans or reductions which has been probably at least five or 10 episodes. It's definitely the most retold story. Bryce, we've already told that. It's all right. We're telling it again. And I really liked episode 51,

the efficiency versus speed. Yeah, that's gotten some decent feedback online. We'll have to have a follow-up episode at some point. Someone shots fired. Did you see the one, that's, that was a good one. That's gotten some decent feedback online. We'll have to have a followup episode at some point. Someone shots fired. Did you see the one comment that said, uh, no offense, but do, do you even understand pre-fetching? I don't know. I don't know who that was and I'm not going to look it up, but you gotta be, you gotta

be nicer. You gotta be nicer on Twitter. Um, all right, here's, right. I did not see that comment. I'm going to go out on a limb and say I probably know more about pre-fiction than that person. Watch as that person is like one of the architects of. All right. So here's my. I mean, I don't have these in any order.

I think the one that overlapped was episode 39 although i sort of convinced you that uh that was the most exciting of this the sean parent episodes um super surprised you didn't say episode 29 um i mean i know we love sean but that might even be it it's a close call between chandler's story of going from like living out of his car slash girlfriend's place, you know, and working at Papa John's to his meteoric rise at Google. I had no idea. I mean, I think maybe that was the thing is that you knew Chandler or know Chandler a lot better than I do. And it might have heard that story before. That was all news to me.

Yeah, but like to me, to be fair, I've heard. Well, I've heard most of the Sean stories before. I think I've only heard the episode 39 Sean story once and I'm not even sure about that. Yeah. Yeah. But that's, and also just that whole four part series with Patricia and Chandler, I think is great. Yeah, it was good. So.

We got to have them back sometime. Yeah. Yeah, for sure. And then of the sort of just you and me episodes, the loss reduction I think was up there more because of too like the hilarity of the fact that it was what the third or fourth time that we had tried to record that episode. I completely forgot about that aspect of it. I completely forgot about that aspect of it. I completely forgot about that. That was the one where I texted you after we'd originally recorded and was like, no, no, no, no, no. You got to come back. Yeah, we need to do a 10-minute addendum. This is Bryce.

Now I know that when Bryce says 10 minutes, what he really means is like 30 to 50 minutes. The 10-minute addendum, yes, took about 50 minutes. And then we ended up throwing it away because at the end of it, I think we realized, oh no, we still need to figure this out. And so, yeah, there was- But we like eventually, eventually we got it. Yeah. Eventually we got it. And so yeah, the loss reduction, definitely.

Galaxy brain programming languages, for those that haven't watched or listened to the backlog, that was where I quizzed Bryce on, I think it was eight or nine, uh, quote unquote galaxy brain programming languages that were invented before, I think the 1980s or maybe in 19, I think 1978,

I think was the youngest programming language. And so Bryce had to guess them. And I think you, I think you nailed it. Uh, you were definitely above 500. You batted above 500. Um,

I'm not sure if i would if i would maybe maybe you edited it very creatively to make it sound like i did not no i did not i mean the most disappointing uh part although you wouldn't make this mistake now is that you didn't guess apl until like the seventh language or something yeah that was that was like in the early days of you becoming an aplL fanatic. Yeah, that's true. And that might have even been before it was embarrassing because that may have been prior to me becoming chair of the U.S. Programming Languages Standards Committee. And the reason that would be embarrassing is because APL, of course, is one of the languages for which said committee that I chair is responsible for. I think we actually did something with the APL standard recently. I don't remember what, but I seem to recall something like that crossing my desk.

Yeah, I have no to round out the list, episode 47, where I just got to wax rhapsodic on combinatory logic, my favorite topic of the moment. And then probably my favorite moment just between you and me was on episode 45, when I forgot that you were a big fan of birds. And then in that moment, it was probably the closest I've been that excited since you mentioning that like Sean parent was going to be at a conference I was at or something or something like that. Or maybe that's what it was is you said, Oh, I could introduce him during his keynote at CBP con,

which didn't end up happening. And then anyways, you said oh yeah i love birds and then i was just like oh yeah and then when i was like i had forgotten how excited i got when we recording and when i was editing it and listening to it i was just like killing myself laughing um at how how amped i got um anyway so we'll put all a link to all of those um i mean if you've been listening from the beginning then then you've just heard us talking about stuff you've already listened to. But if you're a new listener, those are the ones that you might want to go back and listen to if you have an empty catalog.

If you're an existing listener, you can go back and re-listen to the episodes. Like, in fact, go back and do it like 10 times. Yeah, in fact. From different devices with unique yeah while you're while you're working from home just uh while you're on vpn hit up our website you know just hit that play button put it on mute and or don't put it on mute you know maybe uh and and make sure to vpn into was it slovenia oh yeah slovakia slovenia one of the two i bet we're doing quite well we were like number 13 at the time we've got to be we've got to be the one like the top technology there's

got to be at least one country there's what 200 is there 206 countries in the world or is that bones in the body i can never remember um i know that there's 54 countries in africa because i read a behavioral economics book where they tell you to guess a 90 percent economics book. It might have been one of the Richard Thaler ones. Yeah, I had to actually read a bunch of those for my actuarial exams. But they have this like my mother before being a lawyer was an economist and so one um when i was like starting like a very young age our dinner table conversation was like brief lectures in uh the fundamentals of economics and two um i have a strange affinity for economics and like a lot of the podcasts I listen to are economics podcasts.

Yeah, I mean, behavioral economics is fascinating stuff. One of my favorites is a bit of a tome, but Thinking Fast and Slow by Daniel Kahneman. They describe probably what's one of my favorite sort of phenomenon, which is called the peak-end rule. And basically it says that your memory of an experience is largely defined by like the end of it and the peak of it so that you know sort of one of the things they say is you should try not to like you know end a great vacation day or something with a fight because you're largely going to remember that day by like the fight or something and the study that they do is they get people to submerge their hand in really ice cold water for 60 seconds and then they do the

same experiment um with the same group of people but they do it for 60 seconds and then uh add 30 seconds afterwards which is also freezing cold but like one temperature warmer and then they say what do you prefer between the two so one of them is a super set of the other one so they should clearly just you know if if they were you know perfectly rational and could experience things you know the way they were they just they prefer the 60 second freezing cold instead of 90 seconds of freezing cold plus freezing cold plus one degree but largely overwhelmingly like i think 90 plus percent of people prefer ending on the slightly warmer, even though it's still uncomfortable, which I just think is fantastic. So peak end rule. I got lots of those.

And yes, 54 countries. I know that because one of the books talks about how people wildly underestimate confidence intervals. So if I asked you to give like a 90% confidence interval of where there's like 5% on either side that you're wrong about some guests, um, most people don't, they, they wildly, they think they know they're way closer than they are. And like the classic example, I mean, I've ruined it because I've already said the answer, but they say, how many countries do you think are in Africa? And on average, people think it's like 20 or 30. And when you ask them to give a, when you ask them to give a 90% confidence interval, they go, Oh, well, 20 or 30. Okay. We'll say 25 plus or minus, like maybe 15 on each side, maybe, maybe 20 on each side,

but even 20 on each side is five to 45. Most people will only do plus or minus 10. So you get a range of 15 to 35 and it turns out to be 54 but the the truth of the matter is is like do you why do you think there's 20 or 30 in africa like the the truth is is that like oh you know there's a few in north america so it's like africa's got a lot more countries right uh so it's got to be like what like you know times 10 or something um but the truth of the matter is is like you really don't know and if you don't know you should just be like zero to a hundred you know that's my that's my 95 percent uh or you know like two to one hundred like i'm almost positive there's no there's not one country in africa and i'm pretty sure half the countries in the world also too we didn't get back to we're totally on a tangent here 206 is it bones in the

body or countries in the world you know i think it's i have no clue i think it's... I have no clue. I think it's countries in the world. Let's look it up. And also, too, how many bones do I think in their body? Probably... Well, that also kind of depends on how you define countries. I mean, Google will tell us a single number, guaranteed. And I'm going to say bones in the body are... 95.

Oh, shit. Oh, sorry. I swore. It's what Google says. Countries in the... Again like i think that there's different counts depending on how you count country there's 195 un recognized countries in the world but i do not i think that the actual number is uh uh slightly more than that i was wrong yeah so it was 206 bones in the body um it's a sad day. It's a sad day.

All right, so we can do... There's two more things. So this is going to be a bit of a longer episode. We'll probably keep it as one, though. So do you want to do a little APL++ play along? Because this is going to wind back to a great point that I want to make.

Why are we doing this on the year anniversary? Because I want to. and it's going to be awesome. Connor is the editor, so I have no realistic say in this matter. He's also got a very evil grin right now. So, listener, oh boy. I have to say,'m not i'm not devastated the last time i was devastated was when i injured my deltoid ligament to my left angle length ankle a month and a half ago

i was devastated then um i'm probably like 20 as devastated as i was back then. But what just happened? So we've been recording for almost an hour and 40 minutes. And at the 44-minute mark, at which point we hadn't even gotten to this huge journey that we went on where I walked Bryce through a BQN solution to a leak code problem. Oh, we missed all that? I'm pretty sure because the part that was where it got lodged was when you were looking up partition in programming pearls. And then I locked,

I think I knocked it out then. And so we hadn't, I think you had probably said the solution in C++, but we hadn't even gotten to the BQN stuff. So I think we lost all that. And it was gold. It was gold. I all that and it was it was gold it was gold i mean and it was it's it's it was done in such a way that uh it's not going to have the same effect the second time around because you're going to under you're going to understand we can do we'll do it but um

uh oh yeah so i gotta say wow it's uh my heart my heart hurts a little bit that's the thing is a week ago or whatever we had to re-record. But we were talking about sort of a... You would think that production quality would go up while you were doing this, but no. My defense is that I was so excited about this that I just was moving all over the place. And oh, man. Oh, man. Yeah.

And that's the thing. When you were about to explain something and then i was like wait a second wait wait a second i i knocked out my mic thing that probably stopped audacity and then i went and looked and then bryce was still like he was like oh i think i know i think this is what's happening and then i was swearing in the background and like i was like bryce no we gotta bryce you gotta stop and he was like oh no, no, we got to stop. And he was like, oh, no, what? Yeah. But it's okay because we're going to have an awesome future episode where we're going to completely redesign Thrust's partition algorithm now that we've discovered that it is subpar.

Thanks for listening. We hope you enjoyed and have a great day.