General statistics
List of Youtube channels
Youtube commenter search
Distinguished comments
About

p11
ThePrimeTime
comments

Comments by "p11" (@porky1118) on "ThePrimeTime" channel.

4:20 Open Source is NOT Democracy. Democracy is when the people vote about what's done, and then some solution is chosen which doesn't satisfy anybody. Open Source is multiple dictatorships. But if you don't like it, you can fork it. And if the dictator does something wrong, people might switch to a fork, a different dictator.
122
13:50 You don't have to adapt your speech when talking to children. Children arent as stupid as most adults think. You don't have to change your voice, you don't have to talk slower. As long as it's not highly technical, you just talk to kids as you would talk to any other person. And if something is too complicated, you explain the details. Just like you would when talking to adults.
33
13:30 Usually I only work as much as I have to, don't make over hours, sometimes make longer lunch breaks, and might even work a little less than I should. But if there's some timeline we need to hit, I sometimes work until late in the evening, or on weekends. Sometimes I even work on evenings or weekends when I have some exciting idea myself which I want to add to our project. This way it's rather relaxing most of the time while my boss still values my commitment. I think that's a good middle ground.
29
21:16 Primagen is happy that naming doesn't have meaning like in go (upper case is public) A few seconds later it's mentioned how naming has meaning (functions containing test are tests)
25
But nobody actually calls it 0. It's usually E.
20
8:00 There are things you could do to avoid this. 1. Define trains for all of these lines, which are automatically implemented on everything 2. Make a trait ("Traits") to include all these traits (trait Trais { type A: AbusePolicy<Self::D::Tx> + Clone + ...; type C: Clock + Clone ...; ... }) 3. type SimpleDriver<T: Traits> = Driver<Traits::A, Traits::C, ...>;
16
I did the same test with the Gab AI. I got this answer: To remove the female silhouette and recreate the background, I recommend using an image editing software like Photoshop or GIMP. These tools allow you to select and remove specific elements from an image and then fill in the background using various techniques like content-aware fill or manual cloning. As for your second question, there are two "R"s in the word "silhouette".
15
@isodoubIet Every language has some kind of parser. In case of Lisp it's just a simple parser, which is not really related to the semantics. It only parses lists, and the semantics are defined in terms of these lists. I also don't think, Lisp is necessarily about the syntax. You could have Lisps using totally different Syntax, as long as they are internally parsed as lists (preferably stored as arrays, not linked lists).
12
52:15 In Rust, I recently wrote a extension method "select" for bools. So the "ternary" expression "if c { a } else { b }" can also be written as "c.select(a, b)", so when I just choose one of two values based on some condition, the formatter will never turn this one line into 5 lines anymore.
9
And the intuitivity issue can be fixed by fixing natural language and teaching children to count from zero from birth :)
8
16:40 Rust is also pretty authoritarian. You /could/ use a fork of Serde, but then YOUR serde types are not compatible with the serde types of the other library. You really have to decide. Do you want your own system, or do you want to use the system everyone else uses. You could also fork every lib you use and make them use your serde fork. It's mostly because of the orphan rules. I love them, but sometimes I hate them. I guess, that's the main reason why people coordinate to use the same crates. In most languages you would just implement serde for all the types you import, if they don't already implement them. In Rust, it's pretty annoying, so you'd rather demand everyone to use the serialization library you use for your crate. That forces everyone to use the same library. If you don't use it, you are excluded by parts of the ecosystem. And I consider that some kind of authoritarianism. It might be effective to coordinate on some specific systems, but it takes away some of your freedom. You either have to obey, or you have to do your own thing almost completely. You can't just implement your serde fork for each type form other libraries. You /could/ do it if you derive every library you use inside your serde fork itself, but that's far from scalable. You're never done.
8
I could answer the question: The simplest way to implement negative indexing would be this: value = array[mod(index, array.length)] If you implement it like this, the first element wold still be 0, and the last element (using negative indices) would be -1, just like it usually is now. So that's why negative indexing starting with -1 is rather an argument for positive indexing starting with 0.
7
8:20 "Yank" basically doesn't do anything. Cargo just doesn't use yanked versions automatically. If you put it in the cargo lock file manually, it can still be used.
6
7:15 Yeah, I created a Petri Net Editor. It's probably useless to many people, but it's useful to me.
6
I'm pretty sure when I clicked it, it said "Asian" instead of "Asiah". So I guess I'm too late.
5
@spacelem Yeah, indexing would be the proper term. But since the maximum index represents the count, it's just the same. But still, you'd still count the same way since you usually don't care about the indices. But calling myself the "zeroth" (I don't like the term) would make sense. If the rank is just the same as the number of people who are better than me, that's just more intuitive.
5
I hope it will be possible to create multiple scenes containing the same character. I kind of doubt it. Or make small modifications to a scene. Like saying "I want her to have red hair instead" without getting a red shirt as well.
5
2:10 I think, if it's only internal, it's a mess. If I have a personal project, and then decide to publish some component, I clean it up and put it into a new repo as a library, which is then very clean.
5
You could also put a linked list into an array, where you store the next and the previous index in each element. And the removed elements, which are still inside the array also form a linked list in the same array. No custom allocation, pretty cache friendly, and almost all operations are O(1) until you resize. And sorting doesn't need to copy the elements itself, but only the indices.
5
56:00 The way I deal with this is that I just ignore async, and Rust stays fun.
4
5:30 Why does an open source program even have to comply to national law? Can't they just change country?
4
3:20 Writing code with AI is probably around as good as letting an intern program, who just copy pastes stackoverflow without knowing what that code does. Oh, I hate when people just copy paste code. I like to copy paste code, but at least I do some basic things like doing an auto formatting, rename the variable names to make more sense, change everything to fit my coding style, maybe introduce intermediate variables to make it more understandable if too much is done in one calculation (or remove intermediate variables if every small step has a variable), and remove all comments.
4
15:00 That's why I hate pattrens. They give things I do everyday, which aren't special to me, names, and call them patterns. And I'm supposed to know these patterns now. I had no idea what the strategy pattern is. I just pass functions to make my code more generic and that's it.
4
11:00 I encountered something similar in N64 games. The infinite stairs of Mario 64. You go up for a few minutes and then you turn around and you're still at the start. Or in Quest 64 in the last area, you sometimes enter a door which leads to a room with not much in it. But when you leave the room again using the same door, you suddenly ended up in a different place than you came from after a short load time. As a maybe 6yo child, I was mindblown by this.
4
24:50 I still don't get why any framework is even necessary. Couldn't you just write plain HTML?
3
53:10 But in Rust, there aren''t real ternaries. There's just if-else, and in this case, something like this is acceptalbe. But I still prefer early returns (breaks/continues) nowadays, where it's possible instead of using nested ifs. Sometimes I create new functions just to be able to return early.
3
5:00 I got my job after a two week internship, where I used Unity for the first time and programmed their AI using A* which they tried to solve using some deep learning shit. They knew someone working on a game using Unity looking for a programmer, so I applied to him with 2 weeks of Unity/C# experience, did two test work days and that's basically how I got my job.
3
0:07 „What should I do?“ Why are people like this? I started programming because I wanted to create specific things. I always knew, what to program, when I tried a new language. I never did tutorials or test programs. I usually tried to create stuff I failed at in other programming languages. Basically Petri Net simulators, World of Goo remakes and Geometric Algebra libraries. I don't program because I want to program, but I want to create specific projects. If you don't know what you should program, just finish the projects I have been working on.
3
@spacelem Yeah, intuition is the wrong term in this case, since intuition might be subjective. My point is that it reveals more information. The number has an actual meaning (how many people are in front of you), where calling the best rank 1 doesn't provide any useful information.
3
@drdca8263 I also had the thought of prononcing the 0-element "first", since it linguistically doesn't really look related to one. But I would then continued "oneth" or something like that, else the prononciation will get weird later (26th, prononced "twenty seventh")
3
8:00 That let assignment is nothing new. It also works like this in rust.
3
9:38 Clojure should use SLN, the Scopes List notation. It is an indentation based syntax, where indentation implicitly creates inner list. And the outer list doesn't require any parentheses. This code would look like this in SLN: defn square [x] * x x println take 25 map square (range)
3
@nonamehere9658 It does support insertions between elements in O(1). It only has empty elements where elements have been removed. And I only can do one insertion per index, or multiple insertions before and after indices. Indices are not related to the position in the list, though.
3
@isodoubIet What doesn't matter? I said, it's not about the syntax. You could have a direct mapping between C and S-Expressions. I also disagree about writing AST by hand. You write (+ 1 2 3) in Lisp, and the AST after macro expansion would be something like FunctionCall("+", PrimitiveValue::Number(1), FunctionCall("+", PrimitiveValue::Number(2), PrimitiveValue::Number(3))) Besides that, it's a good thing if the syntax the user sees is just a simple representation for the AST, which is the internal representation of the logic. This makes writing AST based macros much easier.
3
@milanstevic8424 I think it comes down to the distinction of ordinal and cardinal numbers. Ordinal numbers (first, second, basically indices) should start with zero. The offset from the start. Cardinal numbers (1, 2, 3, basically counts/quantities) should also start with zero. But zero in this case means no elements. An array of length 0 doesn't have an element 0. And ranges should be inclusive at the start and exclusive at the end. And these should probably be different types, too.
3
40:20 I guess I'm stressed after all. I really don't notice. I have sleeping issues and get headaches once in a while. There have been times where I got headaches almost daily. (but I also realized that I get headaches after coffeine, which slowly started a few years ago and got worse and worse, so after not consuming any cola at all, it's much better)
3
3:00 I think, the best argument for command line is , that you can just copy-paste commands instead of clicking through GUis. Especially if you don't know, what you are doing, you can just google the command, or a coworker can send you the command, and then you paste it and it probably does, what you want. Maybe you have to insert the correct filenames or something. Or especially when using AI, you can just ask the AI "how to resize all your images in one folder and subfolders to a maximum of 1920×1080 without changing the aspect ratio in the fish shell", and it probably gives you a command, which just works.
3
@haemolacriaa rustfmt::skip is even more verbose. And the main reason is, that functional style seems more elegant for such simple cases.
3
34:30 Lifetimes work well in Rust. The trick is just not to use them at all if possible. Or only in structs. And when you use a lifetime in a struct, there has to be a good reason for it. The best reason I came across is when you want to create some temporary type T, which has mutable access to some other type O, forbidding the user to access the value of type O as long as a value of type T exists. In my case it was some edit struct T, which enables editing O safely. Because after editing, the states have to be updated. So the safe way to edit is using a value of type T. And the drop method of T updates the states of O. And only after T has been dropped, the user can access O directly again, which will be in a safe state again.
3
 @cybernetic-ransomware1485 I was thinking about elevators. It's often E and U. But I guess 0 and -1 exists, too.
3
8:20 Whenever my program gets a little more complicated, I use a lot of goto anyway. In rust I can have named blocks now and then either break from them (goto after the block), or continue them (if they are declared as loop, so a goto to the beginning of the block). And loops shouldn't loop by default. Most of my loops have a break at the end anyway. They usually look somewhat like this: for (;;) { if (...) continue; ... if (...) continue; ... break; } This wolud be written easer with goto: loop: if (...) goto loop; ... if (...) goto loop; ... And the break at the end isn't even neccessary.
2
8:15 Yup, that's how things should be done. Always be more restrictive by default, and in order to allow more, you have to add more code. That has been my philosophy since I'm using Rust. Or when I think about it, even before I started using Rust. I defned some language for stories, where you can define something like this: a -> b | c -> d Which means "Scene a first, then either scene b or c, then scene d" And if you defined "a -> b & c -> d", it means the same, but "both, scene b AND c" But when you define both of these at once, always the less restrictive variant is used, which means, after a comes b AND c, but d only needs b OR c to be finished. Not exactly the same, but probably similar.
2
Reminds me of APL/J
2
16:30 I recently had a similar problem. My solution was buying bigger screws.
2
20:36 I totally agree on that. There's no point in having multiple screens when you have workspaces. You will only watch at once screen at a time anyway. And having to switch where you watch is much more complicated than pressing a few buttons. Especially when you have different screen sizes and rotations and placement heights and resolutions and whatnot. And often I just use my one big TV screen (43 inch, 4K) for work. Bigger screens are not better than using multiple screens. Having more than two windows open at once still doesn't feel as good, even if the screen is bigger. Especially because the more windows there are, the more difficult it is to place them properly, especially in tiling WMs. But normally there's no good reason to have more than two open windows per workspace at once anyway. Either you work in a single window, or you compare contents of two different windows, or you copy/move something from one window into another, or one is the text editor and the other one the terminal.
2
2:22 That's just not true. I almost never have borrow checker issues. Because most of the time, when I'm doing lifetime stuff, the design of my library/program is probably "wrong". So I just design my programs to never use explicit lifetimes most of the time.
2
1:20 You started webdev with PHP and SQL? I started learning to make a server a few years ago by learning html and some http lib in some language I already knew. Or maybe I already parsed the headers myself.
2
2:30 I didn't realize this isn't a reaction until now.
2
10:30 I just go with Rust when I want to get something done quickly nowadays. Or once I went with C++, which I'm not that familiar with, but with a bunch of native C stuff, because I was at a vacation and got an idea, but there was no internet, and Rust probably wasn't installed on this cheap notebook yet. I still got it done very fast, in a few hours. I even had to open manpages :)
2
@fredoverflow caar :)
2
10:05 I like fd a lot, but most of the time I run "fd" first, but then I often run a "find" command instead.
2
8:10 Not by retunging stuff from expressions. Just try it in rust. let x; // not mutable if true { x = 1; } else { x = 2; } println!("{x}"); x = 3; // error: x needs to be mutable
2
13:05 I never do PRs on my own projects. But before I commit I usually do "git add -p" and am likely to look at every line of code. Sometimes I also look at the diff in Gitk. Often I find some lines which I don't want to be committed like debug logs, unimportant changes like variable renames, uncomment of parts of the code, or I find obvious mistakes like just having copied some function (like insert_before to insert_after), but didn't replace all the internal calls (like push_before -> push_after or get(i - 1) -> get(i + 1))
2
4:27 The saved time could be used to find and fix the 1000% more bugs you accidentally produced.
2
2:22 Oh, I didn't know that's a thing in America. I was wondering why deepl put the dots inside the quotes and put them outside maually. I think it's different is someone says something. In these cases I'd most likely put the symbol for ending the sentence in the end of the quote, but still inside, especially if it's a questionmark or exclamation mark.
2
21:30 If an intern is able to do something, I haven't been able to, I'm just happy about it. It's something I (hopefully) don't need to do anymore. I as the only (almost) full time programmer at my company, so maybe I should feel bad when an intern accomplishes something, I haven't been able to, maybe for weeks.
2
I also never really cared about these many parentheses. But I always wondered why I can't just write "print x" instead of "(print x)" in Lisp. That's why I prefer SLN (the Scopes List Notation) over pure S-Expressions nowadays. And in Scopes, you can do both, infix ("x + y + w + z") and prefix ("+ x y w z") notation for most operators.
2
@yjlom I know why it's not possible in most Lisps. But it would be an useful feature. Especially the Repl could implicilty just add parentheses before and after the expression in case there are multiple arguments in one line. Besides that "print x" in a repl is not equivalent to (progn (print) (x)) but to (progn print x)
2
12:20 It could derive something different. For example it could send all your program data to some server, whenever it's serialized.
2
@milanstevic8424 I guess I just don't understand. It just doesn't make sense to define centuries to include the last year (1900, 2000) while the decades of the century don't.
2
15:20 I also always use rust, even for very simple tasks.
2
The never nester video made me rework most of my Rust programs. I now use early returns or early continues everywhere. My loops usually look like this: for ... { if (...) continue; ... break; } I started to question why loops loop by default.
2
18:40 Nobody NEEDS syntax highlighting.
2
@josephs1732 You might have fallen into a cult. Maybe you're one of these "ruined" developers now ;)
2
4:15 Whenever I learn a new language or tool inside a language (like ECS or a game engine), I usually try to replicate something I've done many times before, like reprogramming ball physics and connections like in World of Goo, or trying to implement compile time dimension generic geometric algebra.
2
9:00 It's not a mantra. It's what happens when you use Rust for long enough. Doesn't mean it always works perfecrly once it compiles. But probably 90% of the bugs, which might happen in other languages, are just forbidden by the compiler when you don't do anything fancy. And it's pretty nice if you don't have to worry about most stuff. And you still have opt-in unsafety in case the safe way is annyoing. (there seems to be a trend of "you should never use unsafe"; but I think, sometimes it's okay to do unsafe)
2
0:20 When I got into my first job, I took over some codebase. Most of the things my predecessor did seemed stupid, and I still think, they were stupid, but some really weren't that stupid.
2
@fearedjames Yeah, kids usually quickly like me. Just showing serious interest in them is often enough. My mom is even worse. Sometimes children from other visit us, like the childern of a friend of my mom, and usually they just don't want to leave anymore. My mom always wonders what she's doing, because when her children (like me) were somewhere else, we were usually happy to see our mom come back. But maybe it's also just because we don't put so much restrictions on them, like they would be allowed to play video games for hours. (I'm actually happy if children are able to focus on something for more for an hour. The children I mentioned earlier ususally can't focus for that long on a single game)
2
7:30 Maybe just hire people just for these things without giving them a management or system design role.
2
10:38 I would have written this code around the time I started programming. Probably 10 years ago. One thing I would have done different: I would have used one instead of two spaces for indentation.
2
I also prefer the rust way. But I'd like if rust allowed one liners, at least for early returns. @Khwerz
2
16:50 I wouldn't call myself a sophisticated c++ programmer. In one job I had I used C++. But I never did anything sophisticated. And once I didn't have internet connection when at vacation, and for some reason I couldn't use rust (either it wasn't installed or I needed a specific crate which wasn't installed, or I just couldn't read the documentation), so instead I used C++ reading the documentation using manpages and mostly using C except for generic types.
2
11:00 My own website has 4 threads and no databases, and I wrote most of it myself, including the http handling. And if all threads are used, my threadpool just closes all threads and restarts. I thought, only 4 threads might not be enough. When navigating my site, you might click a link every 5 seconds.
2
I didn't know the Avatar had a circumcision.
2
6:26 "ASAP as possible" ;)
2
28:58 People don't know how to grep? I usually do "grep -l" and then pass the resulting file list to something else. Maybe "sed". So "sed" doesn't have to change files which don't even have that word inside. Probably overkill?
1
2:45 Lisp was my first real programming language. I learned most of programming using Game Maker with its own programming language GML. And also RenPy. Around the same time, I was learning Java at school, but it was very basic. I also tried to program games using Java, but didn't get very far. After school I tried C++ with old OpenGL, but I didn't even know of the stl or libs in general yet. I was looking into a bunch of languges, and then got stuck with go for a little. I liked it very much at first, but the lack of generics quickly made me lose interest. And then I stumbled upon Common Lisp. It seemed so minimalistic, and I liked that. (I liked to think about minimalistic langugaes anyway, both programming and spoken languages) At first I didn't think, it was possible to prograam using this. I couldn't see if something was a function or a definition or whatever. So I bought a book, the Land of Lisp. It took me a while to get into it, but soon I thought, I'd never want to use a different language. I was using it almost exclusively for at least three years, except in university. Whenever I saw shortcomings, I thought they could be fixed using macros or something. I played around with a few other languages, often lisp inspired. What annoyed me most was that Lisp was so high level, and I couldn't trust the compiler to optimize stuff for me. I just want my stuff efficient by default. Nowadays I'm mostly using Rust, but there is a Language, which has the real Lisp experience how I would have liked it before. Scopes. It's basically Lisp, but with more powerful macros to make it more like C++ in some ways (for generic types). It has a Rust inspired borrow checker. The low level stuff is lower level than C (there's a distinction between local variables and values). Syntax is indentation based like python. And the general feel is a little like Lua (multiple value semantics).
1
22:55 Oh, yeah, some input forms are just bad. Especially reddit. Sometimes the input completely broke. Copy-Paste and Ctrl+Z is really dangerous. But also twitter has an annoying issue, where the text cursor always moves to the beginning of the line when you switch line.
1
47:40 I don't agree at all. But if he reads this, I guess it's worth a try. I also know that Rust is not prefect. It's very limiting in some ways, and the features that get add aren't important. But it has many good things, so I just stick with it.
1
10:00 You COULD write your code in some way that it's easy to change, which would be hard otherwise. For example if you wrote a game, where some subsystem are 3D (Rendering) and some are 2D (physics, UI), and then you decide to change the dimension of one subsystem (physics), it might be very easy or very difficult, depending on how you wrote it. In the simplest case you would have to change a 2 to a 3 at one place. But in the worst case, you have to change ONLY the types, which are related to physics into their 3D equivalents. In 2D you just used a float to represent rotation, while in 3D it would have to be a quaternion. You have to rewrite most of the algorithms, since you didn't use algorithms based on dimension generic operations (like the dot product), but instead you used dimension specific operations (like accessing x and y of the vector). Or if you convert from 3D to 2D, you would probably have used Vectors for rotation speed, which would be scalars in 2D. So you don't immediately know which vectors have to be changed into floats and which would be 2D vectors insdead.
1
1:30 Natural language is wrong there. That's why nobody counts like this :) 2:50 Counting is different. The number of elements starts with 1. The lenght of a list with no elements is 0. The length of arrays with one element is 1. Ordinal numbers should be treated different than cardinal numbers. Programming languages should have ordinal number types for indexing and cardinal number types for array length. The first index of a range should be inclusive, the last index exclusive, so that "last-index - first-index = count". 7:20 This subscript logic also might apply to real life. If you take part in a race, and you are at rank 3, how many people do you have to overtake until you reach the best place? In one based, it's 2. In zero based it's 3. Which is more intuitive?
1
Oh, that's i3, mine looks totally different
1
19:30 In Rust, I import simple crates like derive for Add/Sub/etc. and replace-with, but I wrote my own http server. I even put a lot of simple stuff into separate crates, since I don't want to rewrite/copy these simple utilities all the time (like my own color type).
1
31:45 How I handle most of my errors in Rust: let _ = writeln!(stream, ...);
1
17:00 Ternary operators just shouldn't exist. Just have if return values like Lisp and Rust do.
1
7:55 My approach: I just don't license my stuff. That means I still own it. And if anybody else uses it, it's technically illegal.
1
4:55 I have to make a fork!!!
1
@demoncorejunior I'm glad I didn't. Later in the video I saw it's already allowed by now.
1
4:40 Is it really so special to build an editor? I could write one in a few weeks and it would be better than nano, right? Maybe I would just fork an existing minimalistic Editor in Rust and add new features into it, when I need some.
1
10:30 I'% pretty sure the Bitcoin source code is written in C.
1
10:05 A typical conversation at my job: I: Should we really change this thing to work totally different? It will be difficult to restore it. I'll remove all the stuff I don't need anymore. > Sure, just change it. We decided for the perfect design, it won't change anymore. A few months later: > I got an idea, for a new design, we should do it like this. I just think, that's almost exactly how we already had it.
1
18:50 I'd probably say something along the lines "Of course there needs to be some degree of teamwork, but normally I prefer to work by myself and don't interact with others more than necessary"
1
15:40 Maybe I'm programming for 5 hours and don't get anything done. And then I get an idea and am 10 hours productive and don't even want to eat. And then I want to sleep, but I just want to finish it, I'm almost done anyway. If I tried to sleep now, I couldn't anyway and would just watch stupid stuff, and still be tired the next day. And the next day, all my motivation would be gone. Now I have the whole day to finish it, and it's almost finished anyway. I'll first have a long breakfast, then I watch a few episodes of my current anime, then I slowly look into my project and realize I have no idea where to start, so I eat lunch and then go outside. And at 9 in the evening I think, I should just finish it, but since it's not finished yet at midnight and I have work tomorrow, I go to bed and haven't finished it yet. If I just finished it yesterday, it would have taken another 5 hours, but at least it would have been finished. Something like this isn't considered in this algorithm, right?
1
16:50 It shouldn't be higher. Maybe the beginning should even be lower. I almost have no ideas about Latex. I just google how to do something and then do copy paste. I often use templates, too. When I want a specific feature in Word, which isn't as obvious as making bold or colored text, I have to google it anyway. But I can't just copy paste, but I have to click through menues. But these menues are often aren't in the same places as in the tutorial on my system, especially when the tutorial screenshots are a few years old. Or I have a non-english system, so everything is called different. It's difficult to use templates. I might use a whole template file, but just importing a specific element to use at multiple places? I think you could import a whole file and copy paste?
1
Two YouTubers I liked to watch very much at once :)
1
55:55 I only use tests for small libraries. For example I wrote tests for some collision detection library to have at least one edge case. It was a capsule collission detection library. So I checkked collissions between every possible tuple of these components: - point (capsule of radius 0, where both ends are at the same point) - sphere (capsule of radius greater 0, where both ends are at the same point) - cylindrical part (a real capsule will be touched at its cylindrical part) - spherical part (a real capsule will be touched at its spherical part) And I had collissions for the most common dimensions (2D and 3D). Technically works in other dimensions, too.
1
37:45 I was visiting some crypto related event at some university in austria, and afterwards there was a contest about use cases for smart contracts on a blockchain for games or something. And everybody was very stubborn on their ideas. Everyone of us had their own views and we were arguing so hard. After a while everybody was a little annoyed that we don't make any progress. But in the end we came up with a solution everybody was happy with. It was some card game, it really could have worked in a decentralized way using only smart contracts, and it required the blockchain to work. Our idea was the last rank of three. The ideas of the other groups were pretty bad, but still ranked better because the judges were stupid. I only remember one idea. It was just a simple game, and the blockchain is only used to track the scores. They had no way to prevent anyone from cheating, which is at least difficult to do in a decentralized way, and probably impossible using smart contracts. However, that's how I learned how valuable hard arguments can be if they get resolved, if not everybody just gives up and we either do something inbetween or just do what one guy says.
1
21:35 Stupid typescript for making implicit copies. In rust you could easily do something like this without any copy. And splitting an array and then copying is actually pretty unintuitive.
1
9:20 I still prefer to build the thing. But it depends. Especially for the simple, boring things I use libraries. For example if I want to derive some traits in Rust, I use the derive more crate. If I want to implement custom error types, I use the thiserror crate. But for vector math, I implement my own crates. And fo rserialization, too.
1
55:30 If the fundamental data type might change, make it a struct maybe? Or make functions generic? for example in Rust, instead of passing an "usize" around all the time, I create a "struct Index(usize)", so when my index changes to "Box<str>" or something, I just have to change the struct and the constructor call(s) ideally.
1
25:00 I don't know if people expect me to respond. Usually I just don't respond. I think once a few months ago I woke up early in the morning on my day off, I planned to sleep again, but then I saw a mail that some external company needed something, which I didn't know about, so I didn't fix something in time, so I just fixed it this in the morning.
1
8:00 No, if used like this, singeltons are pretty stupid. I just use them when I can't have global state, or I don't have control over the main function or something similar. (only in Unity) And I never define a static method when using them. I just use a public static parameter, or different static methods to access it.
1
8:40 „A, Double-U, S“ I think English speakers should start calling Double-U just „We“ (pronounced like „meh“), just like Germans and probabld some other European countries.
1
31:10 I sometimes use LLMs to write the first prototype of some program. Usually it's something like "Parse this file line based and do something with these lines based on some condidions". And most of the time, that's basically it. Maybe I fix some bugs or refactor it, and then I have some command line tool for some specific task I need once in a while. Sometimes I even provide what I already know that I want. For example the data structures. I'm pretty happy with them. But I think, they are just better in Rust since there's less bad Rust code out there. At least compared to Unity/Csharp, which mostly absolute beginners use.
1
49:45 Oh, the video kind of agrees to my last sentence
1
40:52 When I write a Rust library, I sometimes write test for the API, but only when I realize it doesn't work properly, or when I write documentation, so I decide to write documentation tests.
1
2:00 Skill issue? I use Rust for everything. Especially for short scripts it's very nice and useful, and I never had a problem. And I also wrote multiple GUI applications. The GUI is a little cluncy, since I don't use actual GUI frameworks, but it's still better than most of Unity Engine, especially when it comes to usability and shortcuts. My Rust games aren't that great, but at least I feel comfortable to continue working on them years later.
1
5:30 I never had to do coding challenges. But I never explicitly asked about the recruitment process and coding challenges. I just showed interest in how they work and what I would do, and talked about some of my personal projects and open source development maybe. For example when I caused WGPU to continue development after a few months long break, because the internal API changed a lot (I think that's the only contribution I did for that project. I had no experience and just added some additonal parameters which seemed plausible, and the change was accepted).
1
53:45 Looks very annoying to me. There's a return. Why do languages still have return? It's not an early return, so it's kind of useless.
1
45:45 In Rust I sometimes want getters, which return modifiable state. It almost can't happen accidentally. And you would usuall call it `get_mut`. This immutable by default is such a nice feature. I'm so glad, I'm using Rust.
1
39:20 Does it already count as a pre meeting if I talk to my boss before we have a meeting with somebody external?
1
24:00 A square, which inherits from rectangle is such a stupid idea. If I want a square I don't want to store width AND height. If I want my square to be a rectangle, I don't use different types at all, but instead use a rectangle directly. If I have a square and I do setHeight(1) and then setWidth(2), which is totally fine for a rectangle I get results I don't want, and probably want to get an error. If I want square and rectangle to be different types to be sure that squares will always be squares, I'd use an interface with getters for height and width, but still different internal representations.
1
6:25 Being able to work when it counts also has some value. Even if you basically don't do anything, if you CAN do something when there is no time and nobody to do it, that might be worth it.
1
I prefer licenses, which allow me to get money in case somebody makes money from it. I'd like even more if everybody just gives a fuck about licenses, since if it's there, you can technically use it, and as long as there is no legal system to enforce this shit. But as long as there is such a system, people should just abuse it to their likings. I would like to make it difficult for huge companies to use my stuff, while making it easy for people I like to use it.
1
I like how programmers have a completely different view on AI than artists. Artists usually hate AI and sometimes are scared of it. Programmers (who didn't only copy-paste to begin with) are not really scared of AI, often they use it themselves for specific tasks.
1
@username7763 Yeah, that's a good take on it. At my job we often have interns who are either artists or programmers. And it's easy to see if somebody is a good artist in advance. But it's difficult to see if somebody is a good progrrammer. And I'm a programmer. And even if programmers seem to make a good job, it's not always the case. The code is sometimes very bad, and it's difficult to work with it. They do some weird things (like "if (condition) { ... } else if (!condition) { ... } else { /* error, invalid state */ }"), name things badly, rely on some specific configuration (which isn't explicitly named) like creating an array of various things, where each element has a specific meaning (so specific parameters would make more sense). But the boss only sees if it works well.
1
26:35 I sometimes would like to be that guy. But when somebody does in a way which I don't like, I usually tell them that I think it makes sense to make it like this, and explaind why I think like this, and ask if they agree. Usually they don't have an opinion about it, though. And probably even more often I just don't say anything, and just change it the way I want in case I have to work with their stuff. And what I also do a lot is that I just talk about it with some family member and laugh about it.
1
8:50 I think, YOU have this view on the world, since you are pretty deep into web.
1
pacman -R linux
1
3:30
1
44:08 When did he say, reservations are fair? I don't think he cares about that. It's just the pragmatic solution to a problem.
1
No, the offset IS the index. But I'd appreciate a language where you have some specific index type, so there might also be a different index type.
1
@dc443 I just came up with another idea. If you put a dot after a number in some language, it will be the index/ordinal type. Like 1. (first), 2. (second), 3. (third)... But 1. will internally just represented as 0, and if you do math with ordinal numbers, it just uses the internal representation. So most of the math you need for fancy indexing operations, which is easier for 0 based, still works.
1
@rikschaaf Yeah, in the simplest way, you don't even need a different type.
1
11:00 Rust: let mut sum = 0; let Ok(file) = File::open(filename) else { return; } for line in BufReader::new(file).lines() { let Ok(line) = line else { return; } let num: i32 = line.parse().unwrap(); sum += num; } Something like that...
1
RIIR (GitHub)
1
10:15 Isn't normal distribution just as simple in every language? I just looked it up for Rust and already found a library.
1
...and C#? I won't trust this guy. I'm using C# at work (Unity) and else Rust whenever possible (mostly for private projects), and Rust is definitely the better experience.
1
9:10 At my job, we have a skype meeting running all day, which combines most of the communication benefits of in-office while also having most of the home-office benefits.
1
I work at a small company, around 5 paid workers and around 5 interns most of the time. Since 2020 we work remote most of the time. And since then, half of the interns we get don't even live close. We have a group calls running all the time. Even if we go to office once in a while (around once per week currently, only the boss, me and one other paid worker), we can't turn off the call, since not everybody can just come to office. And coming to office makes communication with people even more difficult, since only one person per room can join the call, else we get echo effects. And with the continuously running group call it's almost as if we were in office. If you need something, just ask somebody directly. No need to wait for meetings or write a mail or something. If I want to tallk to a single person, I can just enter another call with just this person. This also makes work with interns pretty comfortable. They can just ask if they have a problem, and they can share screen.
1
10:00 Sounds like this also applies to using engines instead of using a language.
1
11:35 What? The guy who types the dick operator is called porky10g, and I'm porky11.
1
28:24 I just disabled this lint altogether. I guess it isn't even enabled by default when using clippy.
1
22:55 I once worked in my bed a few months ago :) Also I work in the same place as I relax, but even if I planned to work a little more, I often just won't do it. First I used different users for my work and free time. But I stopped this because then I would just not work in my free time at all. And I still don't do it most of the time. That's not my issue.
1
29:45 Orios already exist for that long? They only came to Germany around 10 years ago I think.
1
11:40 Something like this happened before I was using Rust. In the recent years I look at my old code and aways wonder how smart I was back then.
1
2:00 I use Rust for almost everything. And sometimes fish for bash scripts. And C# for work because I have to. And maybe I will come back to Scopes one day, which is the best and most powerful programming language ever.
1
7:00 Why does he talk about such basic stuff anyway? He's a post doc (which means he finished the doctor degree?) And now he's just explaining how Java is stupid, and Rust is better.
1
21:45 I'm not sure about that. Maybe it's just because I don't let other people how to do things, since I already know better myself. But to me it feels like there's usually a best way to do it. And if you do it in a different way, it gets annoying or at least inefficient. You could of course use Arc<RefCell> everywhere. But in most cases it's just a workaround. I kind of gave up when I tried to understand it, and just designed my programs differently, and I'm happy with this approach.
1
3:45 I like polymorphism. But only at compile time. Rust impl trait! And if you need compile time polymorphism? Rust Enums!
1
41:30 I kind of get this problem. If I show parts of a program online, people sometimes just tell me to do something different, even if I know that I have to do it in a specific way, and even explained the reasoning.
1
17:50 Third comment about Scopes on this video: Scopes also being a Lisp like Closure is statically typed, but you almost never specify any type. All functions where you don't specify types are generic functions. Here some comparison of the same code in Rust and Scopes: Scopes: fn mul (x y) * x y Rust: fn mul<X, Y>(x: X, y: Y) -> <X as Mul<Y>>::Output where X: Mul<Y> { x * y }
1
33:00 I think, event handling is a workaround for not having function as parameters in OO.
1
4:40 I think, OO is, when you (can) define specializations of non-abstract types.
1
11:35 "Global Marx in Vim" :)
1
12:50 I got into Linux because once I copied a directory into a subdirectory of itself on Windows, and then the file paths were too long, and so this directory couldn't be touched anymore, and defragmentation also stopped working when it reached that directory. And Linux could just delete this directory without a problem.
1
2:10 I am the only programmer in my company, so for programming related tasks they always think of me first. I prefer if I'm not the person of choice for something. 2:30 I sometimes want to be nitpicky on PRs. I mean we don't do PRs. As long as they work on separate scripts, interns just work on master. And when I look at their code and notice something weird, I tell them when they did something I don't like, but usually I just fix things myself. I just think as long as it works, it should be fine how they did it.
1
14:40 When we had to plan tasks at some job in advance, I always underestimated the effort and only completed around half of the tasks most of the time, and felt bad for it. I tried to multiply the estimated length, but it didn't seem to change anything.
1
Thanks, good to know. If something takes longer than expected, I'll just point to this :) @peterkerj7357
1
27:30 When I wrote some C library some years ago, I planned to intigrate it with Rust from the beginning. But I didn't try to make my program compatible with Rust. That's an issue that I handled in my Rust wrapper. And I think, I even made my C library less Rust-friendly over time. For example I removed "unsafe" markers in method names from my C code. I think, I also allowed more by default in the C library and delegated all the safety features to the Rust library only. And recently I also removed most of the convenience features from the Rust library, so it's just a direct wrapper, which marks some exported methods as unsafe. These can be implemented in separate libraries.
1
@spacelem Not sure what you want to say with your test example. The score is counting, not ordering, I agree.
1
@funkstication That's a bad idea. The len is much more important than the max element. For example if you have one index, but multiple arrays, you could do something like this: while (index > array[arrayindex].length) { arrayindex += 1 index -= array[arrayindex].length } return array[arrayindex][index] Something like this is easiest with zero based indexing and a direct accessor to the length.
1
@voskresenie- I don't remember most of this conversation. When you include zero in your natural numebers, which often makes sense, 0 is the first number. Starting to count from 0 doesn't really make sense. If you count, it's like jumping over the elements. You start at zero, but you don't say it. I would only make ordinal numbers start from zero, not cardinal, just like it already is in most programming languages. So I start at 0, this is my "first" (0st) element. Then I move to place one, there still is an element, so it's element 1. Then I go to place two, and if there still is something, it's 2. And I repeat this until there is no element anymore. The resulting number is the count. The first index with doesn't have an element. That's one way to view it.
1
17:25 Or even better: First all types, then all names: fn add (int, int) (a, b) { a + b }
1
I also didn't find a proper use case for using LLMs for programming. The best use case would be getting started with a new tool by letting it generate a base program. But even then it usually uses an old version of my library or includes some minor mistakes. Or it even tells me about features which don't exist at all. So the only programming related task, where LLMs are actually useful is as an alternative to googling/ducking (which basically means reading documentation or stackoverflow). If I want to edit some documentation file I rather ask "How do I edit my documentation file to do this?", and often it just works. Or for simple commmand line programs. Something like "How do I resize all images in a directory with fish shell?"
1
5:10 Is having multiple monitors really that helpful if you already have multiple workspaces? I only use one screen and just switch around between workspaces with no delay/animation on my i3 with one button combination. Switching workspace takes me less than half a second. And if I have to cycle through multiple ones, I can probably cycle through 5 per second. So I think having multiple screens would just make things more complicated when you already use workspaces. I would have to know which workspace is on which screen. It just adds another layer of complexity.
1
@sophiophile I never used xargs except when copying scripts from somewhere. I think, xargs allows interpreting the input pipe as arguments or something. But it's always possible to do this in a nested way. Example: gedit (grep -l "struct Point" (find src/ -name "*.rs")) (that's fish shell; I think in bash you would get the same behavior by adding a dollar sign before the opening brackets)
1
20:50 That's why I hate to work with Unity inbuilt stuff. Their inbuilt Camera for example uses multiple levels of inheritance, each of them just overwriting a single method. That's why inheritance is stupid. Or their shaders. They use so many macros, so if you want to rewrite a shader and you have to change one little thing, you have to expand the macros manually. And in the expansion of the expansion of the expansion you finally find the single line of code you wanted to change...
1
20:50 I guess hundreds of these already did explode, but they probably were just annoyed for a moment and then just got it back up or maybe just didn't need it anymore. I mean, something like this happened to me recently. I realized that my domain name expired, so I bought a new one after a few months. I just didn't keep track of it.
1
7:00 If the unwraps are the problem, then I see a better solution: There should be no way to quit a program. Remove unwrap and panic from the language. Or at least make it require some explicit marker similar to "unsafe".
1
1:00 Oh, I had a similar feeling when seeing Lisp for the first time. I liked the idea, but I didn't understand how one could ever use this for real programming. Then I used it for real programming and realized I never really knew a programming language before. Technically Lisp, Rust and Scopes are the only languages I really know. At least when it comes to the standard library and using external libraries. I never used a package manager before. And also didn't use most inbuilt packages. Maybe my C knowledge comes close, but that's different. There isn't much in the stdlib, so the stdlib and the package mannager is the OS. 2:20 Still sounds like my experience with Lisp.
1
5:35 Is this really the case? I don't believe JS is better than Scopes or common Lisps.
1
One other case where AI helped me was some function. I told it some results: f(0, x) = 0 f(x, 0) = 0 f(0.5, 0.5) = 0.5 f(1, x) = 1 f(x, 1) = 1 And the AI gave me exactly what I wanted: f(a,b) = ab / (ab + (1 - a)(1 - b)) I didn't expect that coming :)
1
3:50 Yeah, if you have to use unwrap, you probably did something wrong. In most cases, you can get the values using if-let. If you expect something to fail, avoid using unwrap, but either forward the error to the next function or generate some error message for the user, if possible one which doesn't kill the program. If you know, it won't fail, then unwrap is acceptable, but I'd still use except, just to be sure, or use unwrap_unsafe, if you're really sure.
1
7:20 I'm triggered by C# more. I know C++ can do things, Rust can't.
1
9:10 Yeah, it should be Go or OCaml if you just want to get something out.
1
25:00 I never used copilot. I only copy-paste my code into a chatbot. And in some cases it's very helpful, in others it's a little helpful, and in some it's not helpful at all.
1
25:20 I agree to some degree. Rust traits or similar are fine. These traits are abstract and you can even have muliple inheritance. But when it comes to specific implementations, you should not be able to inherit from the objects. Or if it was possible, it should not be allowed to override existing methods, except it's just meant as a default implementation.
1
1:30 It's obvious that using JSON to communicate between machines is a stupid idea. Just send a binary representation of it at least. One byte (or int) for the element count, and then the elements. And if you want to be able to parse while the data arrives, first send the metadata of the parent struct before senditng the actual data. You don't have to be a genious to get to this point. Converting things to some basic binary representation should be basic knowledge.
1
7:30 I think, Rust does it the right way. By default overly protective, but unsafe possible.
1
7:50 What I hate most about inheritance is being able to override things multiple times. Something should either be abstract or maybe virtual with a default implementation, and you only override it once. Else something like this could happen. There's a class A. Class B inherits class A, since it's basically the same, but almost everything is slightly different. Class C inherits from class B, since it's basically the same, but almost everything is slightly different. Why not just let class C already inherit from class A and duplicate the one method from class B, which is still the same? Or even better, turn this method into a function. Maybe just create an abstract super class for A, B and C, all three of them, and only make the parts, which are not exactly the same abstract methods?
1
14:40 I should probably use emacs again.
1
14:12 I first heard him saying the N-Keyword :)
1
13:25 And these functions don't even return a value, they modify global state, and maybe you even have to call them in order.
1
19:00 I expected facade being something like "struct NodeId(usize);" Internally it's just an integer, but you can't use it like an integer.
1
15:20 This whole assumption that functional programming is better for parallelization is complete bullshit. Even map is not parallel. There might be parallel alternatives to map in many functional languages, so a refactor may be less difficult. But I think, that's all. There might be languages which also have a parallel loop construct. I guess the main reason, they don't exist, is a lack of macros in most languages.
1
I had a similar experience for some time. When I got into Lisp, I didn't want to use another language for a few years. I ocassionally tried a few other Lisps or Lisp inspired languages, but it took me a few years to use other languages again. Now my main language for private projects is Rust, but Scopes being a Lisp is still the most powerful language I know. @josephs1732
1
@freesoftwareextremist8119 Reader macros don't seem that useful, though. Is writing "#(1 2 3)" really more difficult than writing "(# 1 2 3)" (where "#" would be a macro which generates a vector at compile time). Or writing "`(list 1 ,b 3)" wouldn't be more complicated than writing "(qq list 1 (uq b) 3)", right? (I think Dale and Scopes do it like this)
1
@freesoftwareextremist8119 I think Scopes also allowed some custom syntax at symbol level. For example "." isn't a special character in Scopes. It's just a valid character for symbols. So when I define a point p and a value p.x, p.x is the value, but p.y is the y component of the point. It does some post processing of non existant symbols.
1
@gagagero Macro generating macros aren't bad. That's the point of macros. You define a while loop in terms of goto, you define a for loop using a while loop, and you might define an iterator based for loop using a for loop. Not sure why you bring in reader macros now. I just mentioned I'm not the biggest fan of reader macros.
1
If somebody sent me such a message for real, I'd think they just want my money.
1
4:44 Serde is shitty if you only want json. Also you shouldn't want json. You should want pure binary. And for binary, just use the binary-serialization libary I wrote (data-stream). Doesn't have proc macros yet.
1
@Encolas What do you think my political beliefs are 😉
1
@hungrymusicwolf But in a democracy everybody has to choose what the majority thinks is good. In open source, you don't have to choose what the majority wants. What the majority wants doesn't matter.
1
@hungrymusicwolf I think A and B are closer to capitalism. If you create something that's popular, people will use it and you get rich. It's a little difficult to view capitalism and democracy separately in western countries, I guess. Both are huge parts of our societies. Democracy to me is more like some person who doesn't know about something wants to do something. People who also don't know about that vote for them. Smart people have the same amount of votes as idiots. But it's easier to convince people of things, which sound good, even if they aren't realistic.
1
If you want a powerful language with low level features (like manual memory management) and compile time stuff, I highly recommend Scopes.
1
@milanstevic8424 What's tricky about that? It obviously belongs to the 21st century (1-based). If there is a definition which doesn't put the year 2000 into the 21th century, it's just not such a good definition.
1
 @milanstevic8424 I don't get your point. The whole premise is weird. The year 2000 just doesn't belong to the 20th century. That would solve the problem, right?
1
@drdca8263 To most western people this probably seems weird. And maybe in the future people will think "Traditionally they called the year before they got one year old year one, not year zero"
1
@Norsilca Yup, he mentioned it in another comment. But that does't make sence to me. The 1th year has to be the oneth year if 0st is first.
1
23:10 Theoretically it's pretty easy for someone to understand how complex programming is. Just write down everything that has to happen. But you really have to think of everything. It's only a few buttons an an image. But the image has to be read from memory in a specific custom format, and you can switch it using buttons. The images need to be cached for efficiency, since there might be many different images. And the texts of the buttons need to be translated to each supported language. And most important: Sometimes bugs appear. Often they can be fixed in a few minutes, but you first have to find the problem. And finding the problem sometimes only takes a few minutes and sometimes a whole day or even longer. And even the same bug could take a factor of hundred longer depending on if your first guess is right. And that's almost random. But more often than not, it's not the first thing you think of. Else you probably would have checked it while programming already.
1
I had to click this because the shape had 3 edges, while the duration of this video was 33:33
1
6:25 Sounds like Zig has Lisp macros.
1
20:00 I don't want JSON. So I'm happy Rust doesn't have this built in. I don't want to be encouraged to use protocols or formats. People should always question which protocols and formats they use. Only for something like TCP/UDP, it's probably good that it's built into Rust. There aren't really alternatives to that.
1
6:45 There's one thing I understand about licenses: It's a weird concept that it's not allowed to use your stuff by default until you allow (license) somebody to use it. This is only possible because of governments or other institutions to enforce this.
1
@MNbenMN The point is that you're legally not allowed to use my lunch if I don't explicitly allow you. But I think this shouldn't be the case if you just make a "copy" of my lunch as long as it's technically possible for you without damaging it, stealing it or physically hurting me, that should be totally fine. You probably can't make a full copy, but you might be able to make a partial copy by making a photo, which copies the look only. And even if you have an automatized lunch copying machine, which almost doesn't cost anything and the result is the exact same lunch, I don't think it makes sense to call it stealing. The only case where I consider copying to be a bad thing would be if you open my lunch box by force. But even in that case, the problem isn't the copy itself, but that you opened it by force in order to copy it.
1
@MNbenMN I tried to be clean with the sandwich example. You don't need my permission to copy my lunch. As soon as I open the lunch box myself, you can at least make a photo. And similar, as soon as I upload something to public, you can copy it. As soon as you view something online, it's already a copy. It doesn't affect the original data. So it doesn't harm anyone nor damage the original data.
1
16:10 A while ago I used the "=>" feature in C# (which allows defining functions without the curly braces). So instead of "bool something(...) { return ...; }" I'd just write "bool something(...) => ...;" Then I had something like this: "=> (a && (b || c)", where a, b and c are somewhat complicated expressions. But then I refactored it with a lot of early returns. { if (!a) return false; if (b) return true; if (c) return true; return false; } Much nicer, since that's what I was thinking anyway when interpreting the &| expression. I look at the original code and think "So if a is false, this returns false. Otherwise it goes in the or-expression, so if one of them is true, it returns true" or something like that.
1
30:05 I like immutable by default. But I'm not a fan of "Everything is immutable" If I have a list, and I add an element to it, I just want to mutate my current list instead of passing my current list to a function and then get a new list back. Maybe I don't care in this example, but when also having move semantics... Maybe I should just pass everything by value in rust, and see what happens? That would basically mean immutable data structures.
1
34:40 Maybe a text editor should have a keybinding tutorial. A fixed number of key bindings will always be shown in the lower right bar. And once you used that command maybe 10 times, it will disappear from the info, since then you probably know it. Or it's more like having vocabulary cards, and it comes back once in a while, but it always takes a bit longer until it comes back each time.
1
9:57 Not sure about that. You could have type specific instances of modules. Instead of having generic vector types (for various floats, ints, and fixed points numbers), you could have generic vector modules with non-generic types.
1
AI never feels real as long as it always replies immediately and exactly once. It should rather reply, and if I don't write back, reply once more. Or sometimes just not reply. Or just write me after I didn't reply for a few months.
1
48:45 I don't agree to this either. I don't even think, you get done faster in an untyped language, where most errors occur only at runtime. In Rust I usually don't have bugs, and when I have they are at places where I handled the error myself (like using unwrap or unsafe code).
1
30:30 If I made a game, I would like to have some shitty reviews as well. As long as there are some hardcore fans, they would comment bomb the bad reviews, and the controversy makes it more popular.
1
19:00 > The only thing that matters is the end product People (especially artists) who criticize AI often don't understand this.
1
This is the answer: find /path/to/folder -type f \( -iname "*.jpg" -o -iname "*.jpeg" -o -iname "*.png" \) -exec convert {} -resize '1920x1080>' {} \; But I didn't try.
1
 @Karurosagu Sure. Being able to copy-paste scripts is an argument, why command line might be even more beginner friendly, especially with the raise of AI.
1
@yjlom And yeah, of course you should check your program. If something looks weird, don't just run it. But still not a bad argument against command line for beginners.
1
1:55 I sometimes use Emojis, but only normal ones and the clown. No mmonkeys or weird nuclear bomb heads.
1
I also never "work" at work. I program, which is basically my hobby, and I tell interns what to do. You can't call that work, can you? 🤡
1
31:20 I use getters and setters in C# a lot. I often implement set like this: ```C# set { if (_foo == value) return; _foo = value; RefreshSomething(value); } ``` I get, why a method might be preferable. But there's some issue, becasue I use Unity: I use Events from the GUI, so if I turn it into a method, all these events just won't work anymore. I can fix it, but they don't cause compile time errors.
1
But I agree that you should not have getters, which don't just return the actual value. These already caused me some problems. The benefit of using computed getters is, that I only recompute the value, when I call the getter. But the issue is, that I always recompute the value, when I call the getter.
1
I'm the only real programmer at some small company, and we never do sprints or agile.
1
4:30 Maybe you just shouldn't commit fraud. It should just be a local change on the system, which does the build, so it will be easy to revert it.
1
I did some similar naming for a rust library. It's called rslnp. rs is Rust of course SLN is the Scopes list notation p is parser
1
Wouldn't it be a good idea to have ONLY packages with one type or function in it? Sounds like an interesting approach to me. You only download the functions you really need.
1
3:30 I'm very confident that I produce bugfree software in Rust. When I model my program to use only the types to represent valid state, I can remove many potential issues at compile time. And since it's easy to put things into libraries, I tend to do this. So when I do this, I want to make it a proper library which I want to reuse with proper error handling using enum types, which only represent the errors that actually occur, and I write a bunch of tests, so I can be sure it never fails.
1
AI even helped me fix a bug. I configured the network interface enp2s0, but it didn't work. I showed the AI the list of network interfaces, and it was just what I expected. So I told the AI, the network interface is enp0s2 just as expected. Then the AI told me that enp2s0 is not enp0s2. First I thought, I just made a typo, but the data in the config the actual device name actually were different.
1
7:00 There is code, which doesn't need to change. Maybe some very specific task, which you put into a library. If you want to change the library, and it's not an easy change, it might be because you do something totally different, and so you probably want a different library.
1
2:55 I also do a lot of grep, even for Unity.
1
@Tstorm731 I think my younger cousin didn't really view me as an adult. I think he was 8 when I was 20. I didn't talk to him using baby langugae. I just talked normal to him. I think the main reason is that I usually talked to him as if we're on the same level. Probably similar to me and my uncle when I was young.
1
6:30 I just discovered some actually useful use cases for using AI in coding. 1. Simple tasks like making every struct field public 2. Writing documentation 3. Adding simple features, which I would have to look up anyway (for example adding a new parameter to a Rust macro, which I don't write often)
1
6:50 I thought, it's called "Democracy"
1
4:15 If I want a language with more freedom than Rust, I'll use Scopes. It's the only language besides C++, which allowes me to implement whatever I want. And in many ways I consider it better. I've never been able to implement a compile time dimension generic geometric algebra library before. And the only language I know of, where this exists, is C++. But my version is better since it tracks if a multivector is a versor or a blade at compiletime, which allows some optimizations. And some functions are only allowed for blades.
1
2:55 Or I fork the project. I might maintain a fork of your project having my own changes while still regularly synchronizing.
1
7:40 I use a lot of Rust, but I'd consider myself libertarian/anarchist. So I wouldn't agree. I want my Rust programs to be as efficient as possible. I want control. So if I release a program, I can disable safety checks. I can use unsafe rust or disable some safety checks (like asserts) globally.
1
2:40 They really do? I think, Fortran, C, Rust and Odin should have the same speed. So probably some bad programming or weird optimization involved.
1
33:00 I like to use unsafe in Rust. Whenever I'm confident that something has a value, I do some unchecked operations.
1
11:11 I think, assembly langugae would be very simple if you used a simpler or more familiar syntax and more useful names like "move_register" instead of "mov" or "register_a" instead of "eax" (not sure if that's even the meaning, I'm not that deep into assembly).
1
@thewhitefalcon8539 Because I want to format consistently. And using the default rules is just the easiest and it's pretty good.
1
@thewhitefalcon8539 I was talking about consistent formatting.
1
1:35 Yay, Lisp! I think he has a point there. Lisp was my first programming language, I really got into. It seemed so weird at first. Everything just looked the same, no matter if it was a function call, a macro call, control flow, type definitions or function definitions. But the different syntax is just an abstraction for humans. It's not necessary. Before I only used Java in school. And I did a bunch of programming in GML (Game Maker) to create many unfinished games. Then I started to get into C++ for Graphics Programming (with old OpenGL), and also tried a few others, one of them being Go. But I never really used a standard library. But when I got into Lisp, I really learned all of the language features, including the more obscure non-standardized ones like the CLOS MOP, and I also used a package manager for the first time (quicklisp). I used Common Lisp for around 3 years almost exclusively. Whenever I didn't like something about the language I thought, I could simply fix it by using macros, preventing me from really getting into other languages, especially if they weren't S-Expression based. Aferwards I had another phase, where I tried a bunch of PLs, most of them being inspired by Lisp, like Stanza, Dale, Scopes, and also started to get into Rust. I still think Scopes has potential to be the best programming language. It's basically Lisp, but using an indentattion based syntax, being more low level than C, having a borrow checker, havnig three kinds of macros, useful multiple value return semantics like Lua. Basically Common Lisp, but everything is better. I think, the best programming language is basically Rust, but with S-Expression based intermediate format. I think, the programming language should not be monolithic, but be devided into two layers: The semantic layer and the syntax layer. The semantic layer defines the core features, and it's defined in S-Expressions. It only is about thinks like if it's statically or dynamically typed (by default), if it's garbage collected, does it have classes, interfaces, etc, a borrow checker, etc. The syntax layer is just a syntax, which could work for different languages. It is just a different format for S-Expressions. So a C-like syntax, a python-like syntax, a node based visual language, a voice based programming language, etc. all map some format to the same S-Expression format. I hope, this will happen to Rust one day, too.
1
24:15 I totally agree to that. Scopes might be the more powerful langugae. And Rust is also very good. But Rust definitely isn't as powerful as C++. And while Scopes is way more powerful than C++, it doesn't really have good libraries.
1
15:35 Democracy...
1
4:50 If you don't dislike C-like syntax in general, syntax is a valid reason to dislike go. Maybe you only want indentation based languages. Or you want some everything in Sexpressions.
1
7:50 You just shouldn't get too attached to your code to begin with. What's wrong with him? If you use version control, your code is never truly gone. It will always be part of history, at least until you rewrite it.
1
7:25 I recently created a C program, which has collections of methods that share a state. The struct is called `PnsNet` and the "methods are called `pnsNet_<method name>`, and the a pointer or const pointer to `PnsNet` as first argument.
1
I hate SQL. It's a bad language to represent a beautiful logic. SQL is basically just set operations. So it would be useful to write it like set operations in a functional style. Just write something like "join(set1, set2, condition)" instead of "JOIN set1, set2 ON condition", or "select(key, where(cross(a, b), condition))" instead of "SELECT key FROM a, b WHERE condition"
1
11:00 You misunderstood that sentence. I didn't immediately get it as well. But he's talking about people who din NOT start in the last five years, people who program for AT LEAST five years and do vim shortcuts.
1
22:15 Nooo, my distro is missing...
1
@JrMonkey Void
1
Is using vim motions similar to using the neo keyboard layout?
1
10:55 I don't put sqirly (?) braces behind my code, but usually if it's a short statement, and I put it in the same line. while (true) { if (condition) continue; if (otherCondition) ++x; if (whatever) return f(x * x); ... }
1
@Khwerz I was pretty sure you were talking about Rust here, since it doesn't require () but {}. if condition { return; } And I'd prefer to be able to write it like this: if condition: return;
1
I don't think, it's best to have as many keywords as possible. I'd rather prefer less keywords. I really like Penne in that regard. It only has infinite loops (you can use break or continue) and goto (only forward). And everything else should be implemented as Macros (like Lisp macros). The macros could also provide some optimizaiton details to the compiler. @Khwerz
1
Part of why I don't like vim is becasue it uses specific keys to represent specific meta keys, which aren't even ordered in a useful way. Besides that, I already have most of my meta keys on a different layer of my standard keyboard position, like arrow keys, numpad, enter, backspace, etc. (I use neo layout). And when using a non querty or similar keyboard layout, keybindings don't make sense anymore anyway. Programs should be keyboard layout independant, if the positioning of the keys is more important than the symbol they represent in the current layout. And it's not that difficult. Just use the scancode, which your windowing library hopefully provide (winit does).
1
4:44 I think, open source is highly libertarian. Not sure if it's capitalistic, since it's mostly about power and opinions, not necessarily about money.
1
26:05 I never do tests of random functions inside a program. I only split off useful parts of a program or write libraries, which might be useful for different programms, and only then I do unit tests of the public interfaces.
1
1:00 I talked to somebody at an indie gamedev meeting, and he told me I should use established programming languages and SOLID ;)
1
2:07 Oh, are you also responsible for all the other bugs at Netflix? Like after watching advertisement, audio and video aren't in sync anymore. Or when watching on web, I first have to start playback before the "Back to main menu" arrow appears. And I don't know how to get the season and episode list, but sometimes it just appears when reopening a tab where I was watching a show.
1
I still like the idea of natural language inspired programming language. I especially like the implicit nesting aspect of natural languages (subordinate clauses), and the implicit temporary variables (pronouns, articles). So a function could be defined like this: fn inerse_of_product_of_sum_and_difference (x: int, y: int) let sum = x + y; let diff = x - y; 1 / (sum * diff) When using pronouns and articles, you could do it like this: fn inverse_of_product_of_sum_and_difference: sum the first int and the second int diff the first int and the second int product this sum and this diff div 1 it "the" would access function parameters. "this" would access the result of the most recent call of this function. "it" would access the most recent result of any function call (the previous line/sentence). Also "the" and "this" would implicitly expect one single objects after them. Also "first" and "second". So the implicit nesting of the sum could be "sum(the(first(int)) and the(second(int)))"
1
23:45 So how would a test work for a game, when the game only should be fun and look good and control properly? How would a test do this? It can't. I can write a test that the character really reaches max speed after half a second on normal ground and after four seconds on ice, but if I change these parameters, I also need to change the tests.
1
25:48 So I'm not allowed to add a fetish story as an example story for my visual novel engine?
1
31:30 My laptop always just shuts off when I try to join a zoom call while running a Unity build.
1
3:25 I can't think of any case, where OO is better than anything else. But maybe I just don't know, what OO really means?
1
18:00 Yeah, I hate comments. When interns add comments to their code, I just remove them when I review/refactor their code. I only add comments in very special scenarios like "It seems like you could write list.Clear instead of list = new List();, but that list is actually being used somewhere else, so don't change this line".
1
When I was a bit over 20 I had used Java, GML, Python, C++, Go, Common Lisp, Scala, C, Julia, Dale, Scopes
1