Conor 0:00

And the third and final talk, which was the best talk of the three by far, was Dave Abraham's keynote that just came out a couple days ago from C on C. Welcome to ADSP The Podcast, episode 198, recorded on August 26th and September 5th of 2024. My name is Connor, and today with my co-host Bryce, we chat and don't talk about anything interesting, so I appended technical tidbits at the end, upcoming conferences, and some talks that I have watched recently.

Bryce 0:46

Okay, so I mean, yeah, I I was late, um, but I have very excellent reasons for being late today, which is that my house has just been utterly chaotic the past eight hours. So first of all.

Conor 1:06

When you say house, do you mean apartment, or do you mean an actual house like in a different city?

Bryce 1:10

I I mean I mean um both the place that I reside in, which is an apartment, but also house in the sense of like my household, um, like the people and things that are in that apartment.

Conor 1:25

I see.

Bryce 1:26

Um and that has been utterly chaotic. So, first of all, we woke up this morning and I had to like cancel my first three meetings to go rescue a pigeon.

Conor 1:47

Um, where we go to uh NDC Tech Town and then I might I might have 10x'd all that stuff, but don't worry, I might have 10x all this stuff too.

Bryce 2:00

So if the listeners are not confused, uh so that's why I'm a little bit late, Connor, because uh there's a lot of a lot of chaos going on here.

Conor 2:17

Well, I'm sorry to hear that Ramona's feeling under the weather. Hopefully she will bounce back before your travels.

Bryce 2:23

Yeah, if you if you listen real carefully in the in the audio, there may have been a cough or two there.

Conor 2:28

Alright. The listener will let us know if they heard that cough. Other than the absolute chaos, how have things been?

Bryce 2:36

Things have been good. Things have been good. I've been uh I've been keeping busy. I've been keeping busy. Got a lot of little projects. Some of them are exciting little projects. And uh and yeah, getting all moved in.

Conor 2:50

You've also had a lot of topic suggestions, which we will get into.

Bryce 2:55

I have had a lot of topic suggestions. I think I tried to write them down somewhere.

Conor 2:58

I just don't know how to write them down. But uh before we do that, a correction. Last episode, I mentioned that there were multiple container adapters, and I listed three. One of which this isn't C. One of which turns out not actually to be a container adapter. I listed the three. They were stack, Q, and deck. And it had been a while since I actually looked at the container adapter list. And yeah, deck is not a container adapter. I So I misspoke.

Bryce 3:37

See, I'd do that one. Oh, 2024 has been such a good year for Bryce not looking like a complete idiot during qu quizzes on this podcast. I'm not saying like like come off as sounding smart, just not like a complete idiot.

Conor 3:52

I can't honestly even remember how many quizzes we've had in 2024.

Bryce 3:58

I think that if we scored all the quizzes from like start of this podcast to 2023, my score would be failing. And if you add if you include 2024, I think my score would be a C.

Conor 4:11

We haven't had we haven't had a quiz though. Like if we did any quizzes in 2024, they all must have been in like the first two months of the year because we have we haven't done one recently at all.

Bryce 4:21

I have been doing some research for this podcast. You you you say I'm such a bad podcast host, but I've been doing some research. I've been listening to some podcasts and some of my favorite podcasts.

Conor 4:30

Wait, you've been listening to podcasts now after four years almost, you started listening and now you're telling me.

Bryce 4:40

I always listen to podcasts. It's just I predominantly listen to economics podcasts. I listen to Planet Money and The Indicator, which is Planet Money's daily podcast, and sometimes Freakonomics, but I primarily listen to Planet Money. Sometimes I listen to audiobooks now too.

Conor 4:55

Hello, Ramona. Oh. She's in the background.

Ramona 4:58

Hi.

Conor 4:58

Price is ignoring you.

Bryce 5:00

Hi. Oh, the whole family's here.

Conor 5:03

Um I'm sorry uh to hear that you're fearing feeling uh below the wall.

Ramona 5:11

Oh, thank you.

Conor 5:13

Hopefully you bounce back.

Ramona 5:14

Did you tell him what you made me do this morning?

Bryce 5:17

I did. I she won yes, I I did tell him about the pigeon rescue. But don't you feel good that you that we saved a life? That pigeon would not have made it.

Ramona 5:26

And I also feel great that you put me in a highly pathogenic.

Bryce 5:30

Okay, I I do it. I neglected to mention this part. When we got to the bird rescue place, there was a big sign on the store on the door saying, do not come inside, risk of bird influenza. So there is that. There is that. I did um oh yeah. So, anyways, I've been listening to some of my favorite podcasts like Planet Money and The Indicator. And sometimes on Planet Money, they do a uh listener quiz. And I actually remember that on another podcast I've listened to a lot, which is the British History Podcast, which that that got me through through COVID. It's a it's a uh podcast by an American guy about British history, and there's also an American history podcast by a British guy, but um uh I listened to like 500 episodes just like whenever I would go biking and listen to an episode. And every few episodes they would have a listener quiz. Um I can't say I always did create an listener quizzes because British history, there are a lot of names, but I think we ought to do um some some listener quizzes. And I think we ought to bring uh we should do what Planet Money does, which is we should bring somebody on to do the Planet Quiz live, just like some uh some other great NPR uh radio shows like uh wait, wait, don't tell me. So uh so yeah, I think we should uh I think we should let people enter in and then one lucky listener can come on the podcast, tell us a bit about what they're doing, and uh play to win some sort of prize.

Conor 7:03

Uh yeah, I guess I guess maybe if it's a thing on other podcasts. I mean they're gonna sound shouldn't you know?

Bryce 7:12

Don't you listen to a lot of podcasts?

Conor 7:14

I listen. I'm like probably I listen to more podcasts than anyone I know, but they none of them are like history. In fact, I don't think there's this I've ever heard a quiz.

Bryce 7:25

I think we could do a quiz. I think we could do this, we can make this happen.

Conor 7:30

Yeah, I mean, uh, how about we listen, folks, if you would like to have see the thing is is if we ask the listener, I was just about to say, if we ask the listener to go and leave a comment uh if they would or would not like or maybe I'll actually I'll put a poll out. But the thing is, is it's gonna be heavily skewed because if I was a listener, I would vote yes to that regardless if I wanted a quiz just for an opportunity to go on like my favorite podcast. Like I I would just be like, oh, they're gonna bring on people? That sounds fun. So is it really that people want to hear a quiz, or do people want to have like uh you know their voice highlighted? Like I would love to go on like a bunch of podcasts that I really like.

Bryce 8:12

Well, okay, maybe we just maybe we just pick a random listener to come on to the podcast, but I think the quiz part would make it fun. Come on, you you always you always you always say I don't contribute. Uh here I am, I come up with a whole thing. I'll even make the quiz. Well, okay, maybe that's a step too far. Because then I'd have to actually go and listen in past episodes.

Conor 8:33

Oh, wait, the quiz is based on our podcast?

Bryce 8:38

The quizzes are based upon, like that's what I've typically seen, the format, where you ask some questions upon topics that were in recent episodes.

Conor 8:46

Oh, that sounds way less interesting. My thought would is it was gonna be like either like a programming language quiz podcast or like a C23 quiz podcast.

Bryce 8:57

We can do the quiz in whatever way, shape, or form you'd like. I just want to have a quiz.

Conor 9:02

I mean, we've had quizzes before, and I think the the listener's favorite quiz has got to be quite that I'm not taking a lot of things. No, but that's the best, is because watching and listening to you squirm is everybody's not favorite pastime, but just uh, you know, we uh anyway, so maybe coming to a theater or a podcast player near you, uh user quizzes. Not today though, because we don't have a guest and we don't have a quiz. Uh there were three topics you wanted to talk about. One of them is boring.

Bryce 9:34

I got a fourth one.

Conor 9:35

And the other two were rotate and MCSS, or otherwise known as maximum contiguous subarasum, otherwise known as Cadene's algorithm. What would you like?

Bryce 9:50

I don't remember why I wanted to talk about that one. So let's okay, what was the one that was boring?

Conor 9:55

No, I'm not gonna bring that one up because I don't want to.

Bryce 9:57

Oh, the elevator one?

Conor 10:27

Anyways, what how how long have we been talking for? Uh because we haven't really said anything interesting.

Bryce 10:30

Should we uh you know I know what do you mean we haven't said anything interesting? I think I think I think the listeners will I think that we should have my elevator guy on to explain to us the the algorithms and logics behind elevators.

Conor 10:42

Alright, yeah, if he wants to come on, sure. Uh but no let him know and let him let him know in advance that I'll be hostile and uh I I will.

Bryce 10:52

I I will let him know that. Okay, so let's talk about let's talk about rotate.

Conor 10:57

All right, seeing as that was the most riveting podcast you've listened to on ADSP the podcast. And at the moment, depending on how long this little segment at the end goes, it's probably going to be the shortest episode yet. I checked the shortest episode to date was 16 minutes. The average is 33, so we're way under what it needs to be. But I just couldn't let you suffer through the elevator and pigeon conversations that I was so eager to have with Bryce. We are adding a couple of technical tidbits, a couple conference announcements, and reviews of talks that I have watched recently. So, first things first, I'm gonna be at two conferences in October. The first is Lambda World, which is being held in Cadiz, Spain from October 2nd to 4th. I will be giving a talk entitled The Power of Function Composition. And the workshop is entitled TAC programming in BQN, Wiwa, CAP, and APL. So if you're interested in array languages or functional languages, and you are in the area or not in the area and interested in traveling, be sure to check that conference out. The second conference is a brand new C conference. You might have heard of it already, C Under the Sea, a play on Phil Nash's C on Sea. It's being held in Breda, Netherlands. Hopefully, I'm pronouncing that correctly. It's just outside of Rotterdam. I believe it's on October 11th, is the day of the conference. October 10th is a workshop day. Both Bryce and I will be there. I'll be giving a talk on arrays, fusion, and CPU versus the GPU. And Bryce will be giving, I believe, his one of his Think Parallel talks. I'm not sure if it is the one that he gave at ACCU and CPP North or if it's the second edition. I'm hoping it's the second edition, but maybe we'll talk about that on a future episode. Speaking of the Think Parallel talks, so I'm going to recap very briefly the three talks that I have watched in the first five days of September. And one of those talks was Think Parallel. I can recommend this talk. It's definitely not Bryce's best talk, but it is a good talk. However, it is very dense, and I'm not sure he achieved the goal he was aiming to achieve. I've read a couple of the comments in the YouTube comment section that basically said they enjoyed the talk, but they're going to need to rewatch it a couple times, especially towards the end. It is quite complicated, and there is a ton of code that potentially it could have been split into like a part one and part two. But if you're interested in how scan, inclusive scan, is and can be implemented in parallel, be sure to check that talk out. The other talk, or one of the other talks that I watched was by John Bandela. I believe there was a co-speaker that did not speak. I can't remember their name off the top of my head. And they are two Google engineers. John was the one presenting at C now. And I should add, Bryce's talk was from ACCU 2024. And John's talk was on Repel, which is a sibling library to Absale. And it is an alternative to the Ranges library that is in the C standard library, as well as Flux, which is a competitor to that. And Range V3, which was the initial implementation of C20's Ranges library. The difference is that it is an eager library, not a lazy library, and the API is incredibly different. They don't overload the pipe operator, and therefore you need to pass things in a kind of Lisp style. And in order to compose things, you have to make use of a RPL compose. And in order to apply things, you have to make use of a RPL apply. Honestly, the aesthetics of the library was not a huge fan of. However, um, you know, it could be much more performant because it's obviously a completely different design than the ranges. And it also claims to be a lot safer because it doesn't suffer from a lot of the dangling pointer issues that uh ranges has famously been criticized for. And unfortunately, it's not open source, so I would have definitely taken this library for a spin and benchmarked it against the Flux Range V3 and C standard ranges, as I've done in the past. But it's not open source, so until it is, uh, we won't know which is better. But they said that they were working towards open sourcing it in the future if folks were interested. I'm interested, so if John, you hear this, hopefully Google can work towards that. And the third and final talk, which was the best talk of the three by far, was Dave Abraham's keynote that just came out a couple days ago from C on C. It was about, at a high level, the Hilo language, but actually not so much. I don't think he even showed any Hilo code. He might have shown a tiny bit, but it was more of a uh philosophical talk about MVS, mutable value semantics, and the history of where MVS and Hilo came from in Swift, and he mentions Stepanov and generic programming. Absolutely fantastic. Also, he shouts out Ken Iverson and APL in the first couple minutes of the talk. We absolutely love that, folks. We absolutely love that. And Dave is just a fantastic speaker across the board. Uh, we cannot recommend this talk highly enough, definitely the best out of three. In the pipeline. Oh, also, I should mention RPL is the namespace on the rappel library. Apparently, rappel is a synonym for AbSale, but originally RPL stood for Ranges Pipeline Library. So, fun fact, if you watch the talk, you will also have known that. In the pipeline, two of the talks that I have not started watching but plan to watch in the next week or two Implementing Ranges and Views in C by Roy Barkin from C Now 2024, and Reintroduction to Generic Programming for C Engineers by Nick DeMarco, who is the one of the temp team members on the STL lab at Adobe with Dave Abrahams, David Senkel, and of course Sean Parent. So, links to all those talks will be in the show notes. Stay tuned to next episode where we actually talk about something technical and interesting, stud rotate. Be sure to check these show notes either in your podcast app or at adsphepodcast.com for links to anything we mentioned in today's episode, as well as a link to a GitHub discussion where you can leave thoughts, comments, and questions. Thanks for listening. We hope you enjoyed and have a great day.

Bryce 17:40

Low quality, high quality. That is the tagline of our podcast.

Conor 17:45

That's not the tagline. Our tagline is chaos with sprinkles of information.