Algorithms + Data Structures = Programs - Episode 153: Pascal vs C vs Ada with Jonathan O'Connor

Episode Date: October 27, 2023

In this episode, Conor and Bryce conintue their conversation with Jonathan O’Connor and chat about Pascal, C, Ada and more!Link to Episode 153 on WebsiteDiscuss this episode, leave a comment, or ask a question (on GitHub)TwitterADSP: The PodcastConor HoekstraBryce Adelstein LelbachAbout the Guest:Jonathan O’Connor in 1988 joined Glockenspiel, a small Irish company. C++ had no virtual destructors, but it did have a coroutine library! I spent 2 years teaching C++ and OOP. In 2000, he switched over to Java. But by 2010, he started 7 wonderful years writing in Ruby. In 2016, he returned to a completely different C++, where one never had to see a pointer if you didn’t want to. These days he is helping to make the world a better place writing C++ code for LADE GmbH, a company building electric car charging infrastructure.Show NotesDate Recorded: 2023-10-18Date Released: 2023-10-27Jonathan O’Connor Meeting C++ BioProgtools on TwitterSpicy - aespa エスパ [Music Bank] | KBS WORLD TV 230519Oxide and Friends Episode 93 - Settling BeefAlgorithms + Data Structures = Programs BookStructure and Interpretation of Computer ProgrammingPascal LanguageAda LanguageWhy Did C Succeed Over Pascal?Alan Turing as a RunnerIntro 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

But Alan Turing is one of my heroes. Some might know was an elite runner in his day as well. And there's some crazy stories about him. At the time, he actually qualified for the 1948 Olympics. And I'd like to think that some of his, you know, ruminating about ideas and stuff, you know. I guess today it's because it's one of the few places where you don't have a screen in front of you. Yeah. And so you're not being distracted.

It's true. That's the great thing about playing chess. I've been playing chess for my whole life. It was about as long as I've been programming. And you're not allowed to bring a phone or a watch. And you're there for four hours, no screen. It's just you, your opponent, the board. And yeah, you can just,

it really lets you think and concentrate on a single thing. Welcome to ADSP The Podcast, episode 153, recorded on October 18th, 2023. My name is Connor, and today with my co-host Bryce, we continue our conversation with Jonathan O'Connor. In today's episode, we chat about Pascal versus C versus Ada, as well as some random topics like chess, exercise, running, and more. Listen, folks, we're now at part two of this episode. I don't actually know where I split. I think I probably split it right when we were asking what Jonathan thought about modern C++. Then we took a right-hand turn into Lisp, and now we're in Lisp Quines and Bash Quines and APL Quines. The topic, the reason, Jonathan reached out to us after we, I think it's safe to say that we were drowning at sea while trying to talk about Pascal, Ada, and other system languages. You know, we did our best.

But we got a lot of – I don't think – it might be our most number of, like, comments on Twitter. And I had people not only commenting on the GitHub discussion but opening issues. Please don't open an issue. We have a discussion. If you'd like to post something go to the discussion don't open a whole new issue because what am i supposed to do with that just like you know it's like some sometimes uh on uh like the radio like npr you'll hear them say like no letters please yeah our problem is no issues please yeah we had uh we had someone uh at prog

tools which uh i don't want to dox the, but you can find their name if you go to their website and look at their CV. I mean, doxing them would be saying their name because they don't have their name in their Twitter. I don't want to dox them, but if you want to dox them, you can just go to their website. No, no, no. But so I want to have this individual, and at that point we'll probably say his name, unless if they want to go by ProgTools. But they reached out, I'll read this quickly, and they had, what did they say? C is usually chosen as a FFI target because most OSs happen to be written in C as it overlaps with OS, ABI, etc., etc. Then he goes on to list a bunch of system languages. And so here

we go. Jovial, Espol, Noop, that's N-E-W-P, that's all capital letters for the last three, Meza, Bliss, all capitals, PL1, PL8, PLS, PLM, Modula2, AlgGOL 68, Concurrent Pass CAL, Class CAL, Object Pass CAL, and then finishes with those are some examples. And I think we actually got a comment, which is a, I can't find their name right now. Maybe we'll mention them in the next recording. But they pointed out, how is it that two people with a podcast named ADSP don't know what the birth languages are. And I mean, touche, that is a very good point. And also we haven't read it either. That is a good point. However, I should note that that deficiency

is entirely Connor's fault and Connor's department because knowing programming languages is what Connor is here for. I don't know why I'm here um but hot takes hot takes and this is the perfect i will cut in someone posted a i don't want to it was uh what do they call it k-pop a k-pop k-pop song by some k-pop artists that for a better hot take than that one oh no i'm just gonna insert it all over the place i feel like it's debut it deserves a better i will try to generate

a better hot take would you like well here's your opportunity we i didn't play it folks bryce go hot take i don't have one right now okay i will uh did you know here's here we're gonna get we're gonna get a uh potentially bryce doesn't even know what has happened. Did you know that our rival podcast now, Oxide and Friends, recorded an episode, invited me on, and didn't invite you on? I did not know that. I got to say I'm offended. Connor, this is where you are definitely showing a huge amount of schadenfreude uh i mean it's bryce i don't think that counts i don't think it counts as schadenfreude connor so you're telling me that that i got to trash talk somebody for absolutely no reason

and that you're the one who has to go on to the other podcast and make amends that that actually sounds like a good arrangement for me i mean folks should go listen to the episode it's already out uh but i think i mean did they invite you on the episode yeah and they actually they replayed our replaying of their audio so it's it's getting extremely meta now maybe we should do an episode where we replay them replaying if they're gonna if they're gonna invite just you to their podcast and i'm gonna invite them here we can we can have this out somewhere i don't know i mean uh yes here's our formal invitation we will we will reach out right before this airs uh to both brian and adam if you'd like to come on and and we can unsettle the beef i think bry... No, no, no, but we're only inviting

the one of the two of them that... Anyways, we're way off track here. The point is Jonathan was one of the folks that reached out to talk about Pascal, Ada, Lisp, and we're kind of doing that. We took our right turn into Lisp land. We sort of are taking a tangent into Quine land. Okay, Connor want i want it known that i of all of us here i'm the only person

who's actually read um uh adsp there we go you might be algorithms and plus data our first guest we haven't actually asked folks and i I've read it in hardback. I do think I own a copy in hardcover that I believe Hartmut Kaiser, my... Wait, you own a copy of ADSP? I think. I'm pretty sure. I mean, that's even worse that you haven't read it then. Yeah.

All right. So let's take a little mini book review here. We're here with... We've actually had like 20 or 30 guests, so I wouldn't be surprised if at least one or two of them read it. But this is the first we know of a guest having read Algorithms plus Data Structures equals Programs. Five out of five, four out of five. What would you say? Is it a book that the youth the kids should read at the time i thought it would be it was like a a three to four out of five but i mean it was it was the book that i learned pascal

from um apart from the lectures that we had in the subject in first year um and i that's about it I don't remember anything else about it because it's like it's a long time ago so was it was it a three out of five mostly because it was a only a decent teaching book for Pascal or was it just as a book in general it was good but not not it wasn't it wasn't no it wasn't it wasn't a wow book i mean like um what's that the structure interpretation of computer programs that is that's not even a five out of five that's like a 10 out of five yeah it's it's amazing. And I didn't get that from Virth's book, but no, it was fine. The funny thing is I learned,

that year I learned Pascal and then I had friends in the maths department and there was a guy there. Trinity College always had strange hangers-on around the university. And maybe lots of universities have that, those kind of people. But this guy had been a barrister in London. And his wife, well, she got some serious illness. And they thought she was going to die very quickly.

So they went to Malta for what they thought was going to die very quickly and so they went to malta for what they thought was going to be three three months while she died and instead she took seven years to die so that's after the seven years that's more like seven years to live right there well indeed yeah it was i mean and she had presumably she had a very nice uh nice life uh in those seven years on malta nice weather and food and so on. But he couldn't go back to the law after that. And so he ended up coming to Dublin and he got a master's in computer science. And at some stage, he taught, he would teach a course on C programming at lunchtimes.

And I took that. And I remember it was only after learning C that I could understand how pointers worked in Pascal. And in fact, if you look at the languages, Pascal and C are almost identical in their capabilities. I don't think there's... Maybe C is a bit more flexible than Pascal. What do you mean by that? Almost identical in their capabilities. Okay, you've got... Both languages have structs or records.

Pascal had a variant. C has union. You have functions. You can just write functions. I think Pascal may have added, allowed nested functions, but I don't remember anymore. But you didn't have object-oriented stuff.

Pointers, they had some weird syntax which seemed to me very unintuitive whereas the c syntax seemed very clear um you know star and ampersand um for for dereference and and address of um but in in yeah in pascal it was slightly it's slightly different but basically the same you could almost i'd say you could write fairly easily a translator that translates code from Pascal to C or vice versa. I don't see there being any difficulties there. It's procedural language.

Both are procedural languages. They don't support any other paradigms. So why did C take over the world not pascal oh i would think it was that pdp 11 pd or pdp 8 which were these my uh mini computers in the 70s and like we had we had the first yeah i think we had the maths department had the first um unix machine in ireland um and it was i don't know much cost probably cost quite a lot probably cost like i don't know, $30,000 pounds, euros or whatever. Back in the day, that was a lot. But it meant that, you know, the whole department of mathematics students

and lecturers could all be writing their papers in tech and playing games and, you know know doing silly things um that was that is not what i imagine the primary motivation for getting a computer to be so that we can typeset our papers and play games well that's what it was used for they were they were they used oh yeah trough there was a thing called trough which i i was like a precursor to to tech but um yeah i've heard of it i'm thinking not good things but i've heard of it yeah i mean again it you used what you had and um and you moved on um so yes but that was that was what was available and i think and that's where anyway that's i could learn my c on that and um but yeah c and pascal i considered them like basically equivalent

and in fact i was looking at ada here i've got this this this i fished out the the manual well ada an introduction by henry ledgard um and it was written in 1980 and when we were it was written in 1980. And when we were, it was the second year we had, in 1982, we had a big project. We had to write code, which would never run, to implement some kind of graphic standard. And the name of the graphic standard, I cannot remember anymore. And we had no idea what we were doing.

And our lecturers gave us this as a a class project so there were what 30 of us involved in writing the code for it um i think it was mainly as a as to see like a a class dynamics kind of approach to see how how would work. And we didn't have a compiler. If we wanted to compile our code, we would email it to an address at a New York university. And overnight, the code would be compiled, but not linked. And we would get the answer back the next day as to whether the code was syntactically correct or not.

That's like the ancient version of Godbolt. Rather slower, but probably it works out if you do Moore's Law on that turnaround. It's probably about equivalent. No, I remember coming across a development method that IBM used, this clean house approach. And there, the developers, they would write their code, but they were not allowed to compile it. Or, you know, they just, they wrote it, they pushed it over to people who would build it for them and test stuff and whatever. And apparently this was quite successful, this way of programming, developing.

I would hate that. I don't understand the motivation there. Why? I think it was back in the day, maybe. I don't know. IBM does strange things, or has done. True. Yeah, I can't imagine what benefits you would derive from that.

There must have been some i would definitely want to be in the team writing the code not the team compiling the code it feels like that's the that's the better end to be on true true it's always more fun to write the code but then you know you're gonna make a compile and work yeah i think maybe I think maybe also it's to make sure that you use your brain. Because, again, my first programs that I wrote, we had four terminals between 30 of us. And we had four hours, five days a week access to those terminals. So we had less than two hours a week on average per person to write our code.

So we would go to the library, write it by hand on paper, and then we would execute it in our heads. And then we would debug it that way, and eventually we would get to type it in and try it out. And I think that thought process is useful, certainly when you're learning. And I think people don't do that anymore. It was interesting.

I think you were saying you've been working on this really hard problem, Bryce. And I think you've probably done an awful lot of thinking in your head, but you're probably one of the few people, yourself and Connor are probably one of the few people out there who do that in the programming world. Because for an algorithm like this – and hang on, I got to get a prop. When I'm developing a parallel algorithm like this, I often turn to paper because sometimes I like try to, you know, with a text editor, like write out some, you know, sort of comment style text that'll help me visualize the problem. But for more complex things, I really just think like being able to draw on paper, like it's so much more expensive. Like sometimes I'll go and like make like a PowerPoint slide too that'll help me understand. But oftentimes, before I make a PowerPoint slide to help me understand an algorithm or just before I make a PowerPoint slide in general, when I'm in the

brainstorming phase, there's something really nice about a blank sheet of paper. Because I have a vision in my head about how I'm going to express something Like I don't know if you can see here, but this particular algorithm, it has two scans that go one from the beginning to the end and the other one from the end to the beginning. And I sort of wanted a way to like visualize those scans. And the way that I was imagining it in my head was this sort of pyramid where like the visualization of the scan, like as you go to each subsequent element, you're stacking the pyramid higher and higher and higher as the previous elements are getting summed into this rolling partial sum that you're doing. And it's sort of like these two pyramids that are, you know, one's going from left is low on the left and high on the right.

And one is high on the right and low on the left. And that's just how I visualized it in my head. And I felt like if I can draw this out on paper to be able to see a few different properties of this thread's doing this thing. And then I know that I'll have this memory ordering guarantee from this thread to this other thread. But then the other part of it, some of the more recent breakthroughs that I've had in this, like this morning I was in the gym and I was thinking about this problem, and not with like a keyboard or something. And sometimes when I'm working on developing an algorithm or developing some code like this,

I do my best work in the gym and I will write out on my phone in like notes, like some pseudo code. And then sometimes I'll write the first version of the code on my phone in the gym. And that just seems, I think to some people, like terrible, like who wants to type out code on the phone? But there's something nice about it to me, especially because like, if I'm writing code on my phone, it's really hard to write the code. And so I can only really do it for something that's very simple and very elegant. Like I can't write like, you know, 300 lines of code. I can just write the essence. And that's sort of what I want to do anyways. So it's, you know, even in this day and age, like my, there's something that resonates about what you just described because I was, this morning I went and wrote some code in the gym and I didn't have a way to compile it right away. And then I went home, and then later I copied it over from my phone into the computer, and then I compiled it.

And that's just part of my process. Now, Bryce, if you had been listening to Connor's other podcast, you would understand that I think iPhones now have a BQN IDE available to them. Isn't that right, Connor? And BQN, I'm sure, is much shorter and terser than C++. So you'd be able to write your thoughts and your algorithm down in a very efficient way. I don't want it to be efficient i i want um like i there's a phase before i start actually like interacting with a compiler or actually like writing code where i'm really just in an idea phase and i don't have a desire to um

i do think sometimes it'd be great if i could like just use godbolt on my phone that that's really the thing that that that would be the optimal thing for me because that's what i use on my my computer but sometimes like this particular thing i was just thinking of a synchronization problem and i really just needed to be able to write it as briefly as possible um and also part of that was because i've been texting my old colleague Olivier incessantly about this particular algorithm and problem and asking him various C++ memory model and CUDA memory model questions,

despite the fact that he no longer works at NVIDIA. He's still the best source of truth for me about memory model semantics. But I do think that I didn't want to interrupt and derail the podcast recording earlier, but about 10 to 15 minutes into our start of the recording of the podcast i was staring at my piece of paper and uh i i realized the the final thing that i think is missing that's that's keeping the code from working so after this i'm gonna go try it and i don't know i feel like it's 80 chance it works we'll see famously or not famously one of the i mean i love running i mean most most folks that listen to any one of my podcasts will know that. I can't help but bringing up my running addiction.

But Alan Turing is one of my heroes. Some might know was an elite runner in his day as well. And there's some crazy stories about him. One of them is I think his marathon PB was 246, which is, for those that don't know stuff about running, it's 26.2 miles or 42.2 kilometers. PB is personal best. Yes, and PB is personal best, which at the time, I mean, there's the world record just got broken at Chicago and it's now two hours and 35 seconds, which is absolutely nuts. But at the time he, he actually qualified for the 1948 Olympics. And, uh, the winning time that year was only like 10 minutes faster, um, than what he was capable of running. And like, I don't know the

exact distances, but there were just some crazy stories where he would run like upwards of 50 kilometers a day sometimes. Like he would run like 25 kilometers to whatever college he was studying at and then 25 kilometers back. And I'd like to think that some of his, you know, ruminating about ideas and stuff, you know, Bryce mentioning that when he's at the gym, sometimes that's the, he has whatever, some of his best ideas. Is there something, something to do with like, you know, people talk about being in the shower or whatever, but I think even better is like when you are just, you know, doing exercise, because you get your, your heart pumping and blood flowing. And I guess, I guess today it's, it's because it's one of the few places where you're not,

you don't have a screen in front of you. Yeah. And so you're not being distracted i mean that's a great observation uh it's true that is uh i've never thought actually what is the longest period of time where you're not like glancing down at your phone like it's maybe one thing if you're switching the song or podcast but like where you're not you know know, checking Twitter, or on some social media platform. Like what is what's the longest that you know, either you're in a movie, maybe for a couple hours, or you're doing physical exercise, or maybe if you're at some kind of family event. But even then, you know, people are always glancing down at their their phones below the table, you know, if they if they feel their pocket buzz, you know, it could be could be important. Gotta go check, you know, we could have got sub sub podded there might be another podcast talking

about our podcast um no i think that i think that's that's the great thing about uh playing chess um like i've been playing chess for my whole life it was about as long as I've been programming. And you're not allowed to bring a phone or a watch into the playing area with you because for obvious reasons now, because they're much better than the best players in the world. So you don't. And you're there for four hours, five hours playing.

That'd be a long game now, but you have no screen. It's just you, your opponent, the board. And yeah, you can just, it really lets you think and concentrate on a single thing. When I'm driving or I was a to a store in the city yesterday and um and and just again it's like another another time when i'm largely free from distractions i'm sort of on autopilot and those are some of the times when i do my best work yeah yeah i was i was talking

to to morton kronberg who's the cto of dialogue limited the company that puts out the main apl interpreter when we were driving back from the middlebrook conference that i was at and um or actually it happened during the conference as i was going out for runs during the conference and he was going he actually had a bike uh that he rented and and was going out for runs during the conference and he was going, he actually had a bike that he rented and was going out on bikes. And he made some sort of throwaway comment that was like, honestly, it's better for the company. If, you know, daily I'm able to get out, even if it's just for 30 minutes or 45 minutes to do like some light exercise, I will be more productive. Like even sure I spent some time not actually, you know my desk whatever doing work but uh i'm going to be overall more productive in that day uh versus if i just sit in front of my desk but by because the time you hit you know for the six hour mark or seven hour

mark like the uh you know your productivity level is decreasing just because of you know is lethargy is that the right word i'm not sure if i'm pronouncing that correctly um whereas if you know you shake things up in the middle of the day or even in the morning if it's the first thing you do it like kind of puts you on this like uh whatever the equivalent of like gym high or runner's high or even if you're getting out and going for a walk and like it's sunny out and you know blue skies and birds chirping it just like it does it activates something in your brain or at least for me i used to i would wake up at uh six or even.30 and I'd go do my two hours of cardio and then I'd be done like 9.30, 10. And then I would start my work day and it was just great because one, I would feel like I'd accomplished something.

I'd done my workout for the day even before my day had started. That was great. And I'd be awake. I'd feel energized. That is slightly less compatible with my current life. But yeah. And speaking of distractions, you know, I'm terrible about, you know, I always have my phone with me. I'm always checking my phone. But my girlfriend, you know, she's great about it she um you know she doesn't check her phone you know almost almost ever um almost to the point of being a problem but

she's she's really good about knowing that like she she needs to be free from distractions so bryce have you turned off notifications or audible notifications on your phone? Oh, yeah, yeah, I have. And if I'm trying to work and focus on something, I'll put it in do not disturb mode. And I could never own an Apple Watch or a smartwatch or anything like that. One, because I love mechanical watches, but two, it's just like I wouldn't want to have something on me all the time that's going to give me more of that notification addiction. Yeah, I keep my phone perennially on do not,

or not do not disturb, but like silent, like it never buzzes. It never beeps. But like, I also have a, not like a complicated view, but like, I think we are all cyborgs now. Like the idea of not having a phone, like I need my phone. It's like, we almost exist in the cyber world on our social media platforms or like, think of how many different messaging platforms are like, what's the way that i communicate with my parents and sisters or friends that are in different parts of the world and like my interaction like i would say i interact

socially more with people like on the internet and and online than i do with like people in real life like how i mean bryce how often do we see each other yeah i mean we joke about like me not coming to see you but like we do these recordings I do my other podcast recordings are all completely done like through the power of technology and none of that stuff would be possible I mean they would be possible to the extent that if there's someone else in the same city as you you could get together record stuff but I just like I I have a like conflicted feeling like I don't ever want to be without my phone because I feel like it is an extension of who I am as a person. People back in the day, they used to ask, oh, do you mind if I borrow your phone for

a quick phone call? To me, I feel like you're asking for a part of my soul. No, you could destroy me. Obviously, they have the modes where you, you know, only do it for phone calls, but like, imagine the harm that you could cause to someone with just like an unlocked phone where you have access to every single one of their, you know, social media platforms. And even if it's, even if you're not a public person, but the harm that you could do to someone's relationships with someone, because you, you can essentially become that person just by having access to their phone.

And so, like, I never want to, like, not that I don't trust people, but I don't really trust people, you know? It's like, would you give, you know, the essence of who you are to someone for, like, 20 minutes? And it's like, well, maybe someone you really trust and love, but not to just anyone. Anyways, I can see, Jonathan, you want to say something? Yeah, Connor, because so many films that are out there show somebody getting into huge trouble because they handed their phone to somebody and that person sent a text to everybody in their contacts. Can you actually do that is that actually possible was that a thing many many years ago and i think it's not here phones or something you i mean definitely when

you are uh either if you want to like message blast someone like when you're sending some if you like very easily you can go to your photos app and if you click on a photo and then you go share when you go to like facebook or meta messenger or whatsapp it will give you just a list of people that you can like check mark of like all the people you want to send this to i've never done like a a you know message blast but i wouldn't let's actually just go check i wouldn't be surprised if there is some way to just go like select all. Oh yeah, yeah. I can choose all contracts. Yeah.

Fair enough. My mother seems to solely communicate in group texts with as many people as she can, as possibly could be relevant. So I have this like, I have like 20 group texts of various groupings of people that my mom has texted at the same time and then she'll send like she doesn't know how to send a text to just one person or to send a photo to just one person yeah i mean that's uh i feel like uh you know when i think about my parents uh more specifically my mother i know she doesn't listen to this so she won't she won't mind me saying this.

But, like, I feel like the harm that my mother could do to herself is limited in her to the extent that, you know, she's not going to be figuring out – she's not going to accidentally be doing something. Because, I mean, she famously – I think maybe she upgraded to a smartphone. But for many years, she was going to these places to get flip phones. Oh, my dad was the same way. Although he, and he's, he's tech savvy, but he finally, he finally got an iPhone after many years. Um, it was funny. I had my, my mom was visiting me this weekend. And, um, I mentioned that I've mentioned my mother and my girlfriend in the podcast a few times and that I don't worry because they don't, you know, they don't listen to the podcast and both of them did threaten to listen to i think my mom actually listened to some episodes

she's like oh like where can i find it and i'm like we're available on all platforms she's like but wait you're on apple podcasts i was expecting to sit to have your mom on on the on the the podcast maybe a year ago i think that was a that was a promise yeah never supposed to do that weren't we? What happened? I don't remember what the logistics of that were, but clearly we forgot. I mean, we have forgotten. But at the time, I think we were trying to orchestrate it. And you had said that your mom was going to be in New York for a weekend.

But the problem is that my mom was not going to be. So neither Bryce and I live in the same cities as our mothers. And I think you had made some comment, Bryce, that saying like, it would be better if both of us with our mothers were behind this, like had the same setup. So it's every everyone is virtual, which I think you said, you would prefer not to do or like both of us were with our moms. Because otherwise, if it's like two people are behind one mic, and then two people are behind screens, it's going to be trickier. I mean, I'm not sure if there'll ever be a weekend where all four of us are in the same place.

Be sure to check the show notes either in your podcast app or at ADSP the podcast.com for links to anything we mentioned in today's episode, as well as a link to the GitHub discussion where you can leave comments, thoughts, and questions. Thanks for listening. We hope you enjoyed and have a great day. Low quality, high quantity. That is the tagline of our podcast. That's not the tagline.

Our tagline is chaos with sprinkles of information.