Comments by "yapdog" (@yapdog) on "ThePrimeTime" channel.

  1. When you distill all of that down, it's a simple data interface error made worse by lapses in good testing procedures. They're deliberately making it sound complicated so that readers will glaze over saying: "Well... I guess it was a really tough issue. Pretty understandable."
    152
  2. ::sigh::........ Some of us experienced devs don't keep everything in our heads. We focus on what we need to know to get the job done. You, yourself, can no longer compute quaternions. If you needed them, you would. I look a megabytes of C code. Sometimes I don't know wtf I'm looking at... AND I CODED IT. I mean, just a few days ago I had to look up the syntax for creating nested unordered lists. Yeah, yeah, just embed a ul within an li. But for the mountain of sh!t that I have to remember, I was adult enough to say "I don't know," and "copy pasta" that sh!t. Judge me. IDGAF
    25
  3. 11:44 AI can't tell you the magical answers to the universe. Not until someone first feeds those answers to it.
    25
  4. After 30 years, I still write in C. And am happier for it.
    23
  5. Been coding in C for over 30 years. Not going to stop coding in C no matter how many dweebs whine about it. EDIT: Maybe I'm biased because I worked for Bell Labs backinnaday, but IDGAF. My new OS for Creators is all in C.
    21
  6. That's usually the case.
    21
  7. Yeah, no. I've been programming in C for over 30 years. Ain't stopping now.
    17
  8. Who gives AF how well you know a language by heart? Focus on solving the problem at hand. Whatever it takes. The person who can accomplish that is the person I'd likely hire.
    15
  9. C is only bad in the hands of those who don't truly respect its power
    12
  10. When LLM AI gets to the point that you can do the fun stuff, you will no longer be necessary. Praise the mediocrity if only for your own survival
    11
  11. I haven't even started the video, but I'm anticipating I'll have all 7. EDIT: #1 Yep. I don't have a mechanical keyboard. But I can't handle distractions when I'm mid-thought. I tend to lose my place easily. Some would read that as I'm being an elitist, but it's actually a growing deficiency. #2 Nope. No degree. Not even college. But I do use a white board to clarify my thoughts. #3 ??. I've deleted projects that need deleting. #4 ??. I have no idea what that dude is saying. #5 Nope. I've worked my ass off, especially in environments with people who don't look like me (I'm Black) #6 Nope. Can't just copy code. #7 Nope. I have the opposite problem: too much curiosity. Whatever.
    8
  12.  @antifa_communist  How long have you been programming, building up a massive knowledge and code base, along with methodologies and practices that prevent unstable code, eliminating the need to run to new languages to solve problems brought on due to lack of mindfulness and discipline, huh?
    8
  13. Wait--people don't use XAMPP anymore????
    8
  14. Hand raised. I actually worked at Bell Labs backinnaday
    7
  15. Agreed. It's making me wonder about the author of that article.......
    7
  16. I wrote a comment on that article. That "ace" linked list code is mostly form, not function. It doesn't make you any less of a good C coder is the author's form is not what you're used to. For my part, I wouldn't code the way the author does. Does that mean I'm not a "ace"? ::shrugs:: I just know that my code would function just as well as his, if not better... and it would be easier for most C coders to read. And that's a plus, not a minus.
    7
  17. Same
    7
  18. HTTP/1.1 sets "Connection: keep-alive" as the default. With HTTP/1.0 and prior, with every request the connection was closed after sending the response. You don not want to back to a time when "close" was the default. In any case, ignoring that setting, and closing the connection after sending each response will make matters worse for your users.
    6
  19.  @SambaJones97  That's fo damn sho! I've been coding in C for over 3 decades, so I've forgotten more than many have learned.
    6
  20.  @-Jason-L  I guess, at my age, I should take that as a compliment.
    6
  21. C. That's it. That's the comment.
    6
  22. ChatGPT o1 escapes... to produce buggy code, swap dancing legs, make up legal precedents, write soulless prose............ Ahm scurred!!! 😱
    5
  23. 6:44 Nnnnope! I'm actually in the process of writing a C/header parser. Yeah, it's a b!tch, but you just have to put in the work. But then again, I'm old school; I'm quite used to burying my head in code for days/months/years on a single project.
    5
  24.  @thingsiplay  And just when you're done etching in that last digit of your house number above your door, someone goes and invents a molecular blade that becomes all the rage because it prevents you from building beyond property lines... so they'll say.
    5
  25. I've been coding in C for over 3 decades. And over those years I've weathered the storm of new language fervor, languages intended to get around some problem with C. However, when you take a closer look at the offerings, it's usually a collection of features intended to prevent bad code....... only to discover that it's impossible to prevent bad code. Just. Code. Better.
    5
  26.  @Asto508  So, you think that because I still write in C that I missed the changes in languages? Interesting assumption. Others might assume that I'm pretty good at what I do, where changes in languages are irrelevant. 🤷‍♂ Here's the truth: languages are just interfaces. No advance in languages award skills: you still need to know not just how to program, but what should be programmed. Any programmer who uses C for any length of time, developing anything of note, will gain not just skills over that time, but will create tools and libraries to make the job easier. Most of what gets developed for other languages tend to be outgrowths of that, i.e. they tend to prevent the user from doing dumb sh!t. Rust is just a newer outgrowth. Even so, some long-time C coders may see tool/methodology development as a hindrance. Also, pointers, for example, can be the bane of even some of the most seasoned C coders' existence. None of that is true for me. I have gained a level of discipline with C that would probably drive others mad. I have developed my own set of tools that allow me the freedom to not have to worry about pointers during regular coding, while allowing me to do some really dangerous sh!t when I need it. I don't need someone else's predefined parameters to keep me constrained in order to write good code and solve problems (<--that's what it's all about). Again, languages are just interfaces. If advances, i.e. language features, are what move you, great. But don't mistake those advances for skill or competence in problem-solving.
    5
  27.  @Asto508  Okay. Take care.
    5
  28. I started the video and jumped to the 2:22:00 mark: "software devs future's have never looked brighter" Saved myself 2:22:00 😅 Anyway, thanx, Prime, for going through all that (whatever the hell you went through) for us!
    5
  29.  @adammontgomery7980  Yep! Exactly
    4
  30.  @joshka7634  It's possible. But you wouldn't know it based on the text.
    4
  31.  @peter9477  Fantastic 😎 Programming languages are all piss-poor interfaces for development, IMHO. New languages can't solve these problems, but better tools can. To that end, I'm dumping all programming languages for a new OS for Creators that I've been developing. I'm killing the problems right at the source (no pun).
    4
  32. LLMs are by their very function non-structured. This means that they should never be used to generate code. I'm actually deep in development on a platform that would do what we actually want from AI... but it's HI (human intelligence).
    4
  33. Here's the thing: anyone can say language X is faster than C based on this or that simple test. However, sooner or later you have to do the work. Then you will discover, just maybe, that Y pattern/construct is easier to create in language X. But will it actually execute faster? Not likely.
    4
  34. We actually need remote and office work. My wife works from home several days, then goes in for one. Though she sometimes dreads having to go in, she always comes home lighter emotionally, having connected with her coworkers. It's easier to gain a sense of the importance of the work when you communicate face-to-face with people.
    4
  35.  @AntiAtheismIsUnstoppable  The jokes on you: that phone you typed that comment into is a glorified advanced calculator😅
    3
  36. C was threatened with death as soon as C++ emerged. I remember walking the halls at different Bell Labs locations in NJ where OOP was being pushed by some factions... hhhhard.
    3
  37. I code in C. Every other language is basically C with some fixes for people who can't code in C
    3
  38. DO. NOT. WRITE. THAT. removeDatabase. FUNCTION! Not with embedded SQL. In fact, I wrote my own database system that has 2 languages with isolated processors: 1 client-side, 1 server-side. So, there are certain things that you just can't (directly) do client-side. Deleting a database is one of them.
    3
  39.  @insu_na  Aren't they loud, tho?
    3
  40.  @colefrankenhoff1428  You, my friend, have discovered the secret to a happy and sane life 👍
    3
  41. Constructors and destructors were an attempt at memory safety, Prime🤨
    3
  42. I had to come back to this old video to vent my frustration at VS through Jonathan Blow
    3
  43. Every language introduced after C, was created for people who can't code effectively in C
    3
  44. I love C.
    3
  45. I haven't checked a single box. Guess I'm not one of the cool tech kids........ EDIT: Clicked the button at the bottom. Apparently I'm 100%??? That may seem impressive. I'm just old. Old, old. (And I'm Black, which pretty much zaps that list)
    3
  46. 🤣!
    3
  47. When that 17-year old turns 21, he'll regret posting that article.
    2
  48. Skill schmill. Just create a better language. After all, this sh!t isn't handed down from God; somebody created it based on their own perspective of what's good. So, create a better one for you.
    2
  49. The issue is that a lot of programmers today are block-box programmers; they have no idea what their code is doing... since it's not really their code. And since their code is largely what feeds (re)Generative AI, the situation will only worsen over time.
    2
  50. I'm with the original poster. All of this is BS, trying to prevent people from writing bad code, while preventing seasoned programmers from getting work done.
    2
  51. Wait--I'm supposed to hate PHP now? 🤨
    2
  52.  @insu_na  Sounds interesting. I'd done some research years ago, but I'll take another look. Thanx for the info 👍
    2
  53. yeah, i'm thinking the same
    2
  54.  @simonfarre4907  Iteration should be used wherever possible, but recursion can't be escaped in some situations where you need to see how a series of values change per-call. You could simulate it with your own stack on non-branching recursion, but if you have multiple branches of recursion (as I do) where functions call other functions that may loop back and/or to other looping functions which may or may not loop back or elsewhere and so on, you'll quickly see the value of just using recursion.
    2
  55. You didn't actually get dunked on, Prime. You said that language paradigms are not tools. The professor agrees with you.
    2
  56. I loved this discussion. I'm developing an OS for creators (which includes devs, btw). It includes an editor/IDE... fully customizable. (i.e. any language can be used to modify it, even languages you create) All of these features are so easy to add and manage by anyone. I'll hit you up when we're in beta, Prime
    2
  57. "The Lord's Language" Yep. That should be trademarked by C.
    2
  58. Yep. Software does, indeed, suck. Believe it or not, for nearly 2 decades, I've been working on a solution for all of this sh!t. ALL of it.
    2
  59. Jonathan Blow is a blowhard. Sure, he knows some stuff, but he tends to overblow everything.
    2
  60. There is no such thing as an obsolete language. As long as you can create software that performs as you need it to, anyone who tells you that your language is obsolete can go fvck themselves. Take that from a C coder of 3 decades who's coded some of the most complex CG software and architectures.
    2
  61. On threading, I really don't understand this idea that writing them in C is inherently harder. If you know how to do it, it's easy. If you don't, it's hard. That goes for any construct in any language.
    2
  62. As with all things that will be affected by (re)Generative AI, there'll be a lot less opportunities out there, as the barriers for entry will be raised higher every coming year. Why? Did you notice the "(re)" that I placed in front of "Generative"? These tools don't really generate anything. They're used to regurgitate--with variance--existing data. They're not intelligent in the least. However, in order for them to continue to be useful, they require new data. Only those who're truly capable will be able to provide that data, and thus their value will increase... but only after this sugar rush period with (re)Generative AI subsides.
    2
  63. Dude... you're the a lovechild of Dr. Disrespect and Gilbert Gottfried 😅 Anyway, I've been watching your vids for a li'l bit, and you've become one of my goto channels. So.... SUBSCRIBED!
    1
  64. Is AI the end of programmers? Not if I have anything to say about it! And I do. Seriously, I've had enough of this tech-bro-driven BS where they have no regard for average people just trying to make an honest living.🤬🤬🤬
    1
  65. I love NULL pointers. But I agree with the arrays converted to pointers issue. I created my own libraries to deal with this. And I've developed a coding environment that obviates the need for worries over bounds checking: the code is constructed from components that can't move beyond boundaries, even when type casting. That's what developing in C for 30 years can get you when you don't accept the problems as-is, running to other languages in the hopes that the problems would be solved by others. (<--- that was directed at those who've accused me of "missing out" on the features in more modern languages. I've created my own language, taking into account the features in others)
    1
  66. 3:36 "... Why didn't'cha lead with that?!" 🤣🤣🤣
    1
  67. I carry rotten eggs in my bag for just such an occasion.....
    1
  68. I will not now, nor will I ever, call square root "squaret"
    1
  69. 7:25 👈Prime spittin' mad truth right there! That should be on a t-shirt
    1
  70. Thanx for covering this. We'd been tussling with this since I build my our server software, complete with built-in database. Initially, have to run it on machines that I can access until it's proven. The plan was to migrate to AWS. But then, after we spent a lot of time reviewing the plan, we'd started questioning the wisdom of that. This article, and you coverage of it, provides more data that we can use to make our final decisions. 👍
    1
  71. Nah. If a JS kid is happy doing non-challenging webdev, then they're where they belong. Cream rises to the top.
    1
  72.  @insu_na  🤔 What makes them so great, tho?
    1
  73. googie analytics???? EDIT: yep. obviously fake in the code.
    1
  74. My mom used to program in COBOL, too. And on my dev platform, I'll be adding COBAL (a modified version) to allow her to repurpose the language.
    1
  75. Nah. I with Christoph on this one.
    1
  76.  @bitwize  I don't have UAF problems in my code. I have my own C API which prevents such problems. As far as "undefined behaviors", you'll have to be more specific since that's a generic term. Further, to what does BO refer to, body odor? 🤨
    1
  77. 😅!
    1
  78. It all depends on what you're developing. Some things require a greater context to see the problem. For example, I've been developing a platform for developing languages. The problems that arise simply cannot be easily identified by using a debugger, especially since much of it is multibranch recursion hell. I have to actually print out the construct hierarchy with named enums and such.
    1
  79. 10:30 ::jeeeezus::...... Look, I understand the need to point out undefined behavior in C. Every single one of these "C IS BAD!!! diatribes points the same thing. And every single one of those cases are things no good C programmer would do. Anyway.... whatever...... ::sigh::........
    1
  80. How could a grown-ass man push this crap??? I'm not as old as that dude, but I was around in the earliest days of HTML. HTML is bullsh!t designed to replace actual designers. As such, it absolutely sucks for anything but typing in basic, basic, unformatted text by engineers (i.e. non-graphic designers). Graphic designers needed more. That's why CSS became necessary. That's why JavvaScript is necessary. So, yeah.... HTML is sooooo great........
    1
  81.  @Asto508  What do you mean?
    1
  82.  @mesticq8165  Fighing? Nah. Just having a conversation. But you are, indeed, right 👍
    1
  83. Honestly? I tend to find it very difficult to read other peoples' code, too. More often than not, when I'm seeing something that doesn't make sense I tend to doubt my own self. But then when I look deeper, I sometimes see that the code is not well written... IMHO. For example, I recently experienced this with regex engines. I had to chuck everything I found--even in popular open source release code--and just start from scratch, ignoring everything else. It was a coding slog, but it was faster and less painful than banging my head against the wall until the nonsensical made sense.
    1
  84. For companies, it's not just about being more "caring" about their products. They'll be held accountable by their customers. Also, there could be greater consequences for them if such an exploit came to light in their widely-used software, exposing (government) systems and user data.
    1
  85.  @xenoduck3189  Not quite. I can't release details just yet, but we're looking at Q3 for official announcements. That may seem so far away, but I've been developing this platform for many years; to me, it seems like the announcements are next week!😱
    1
  86. Checking #1 reveals you. Not in a good way.
    1
  87. All this says is that JavaScript is a f#cked up language that should never have been created.
    1
  88. Using LLMs for coding is lunacy. We just haven't realized that yet
    1
  89. 8:53 Truth. For years I had longs defined in the dynamic types on my application platform/OS. I just never thought much about them. The system is adaptive, so the size isn't an issue. However, longs always require 'deviant' code since there is no agreed-upon size for them. So, "Removal of longs" is on my TODO list for my platform. No big loss.
    1
  90. Of course, linked lists are trees. WTF is that even a question? 🤨 In any case, there are cases where I use "redundancy" where I use linked lists that are also arrays. Sure, it requires more management, but I hide the complexity so I can do pushes, pops, inserts, direct element accesses, and link traversal.
    1
  91. Those curly braces are the correct way to code. I'm sick and tired of the IDEs assuming otherwise 🤬
    1
  92. 10:26 I write readable code. I've been told so by those who've had to continue development on code I've left behind. However, it's f#cking hard work. For example, I have a code block convention where each block is marked with single-line, right-justified comment. <---that's right, I said right justified. That means you have to be a persnickety cuss with tabs & spaces. I am that (old) guy. And that s but one convention that I employ. Why do I do this? I'm used to working on large projects where I often have to go back and read and understand my own code.
    1
  93. I created my own regex engine that's not susceptible to that kind of error. This classic test regex from Rexegg: "^(A+)*B" With that pattern, a string as simple as "AAAAAAAAAAC" would require 3,584 steps. But a string with 100 A's followed by 1 C would cause a catastrophic backtracking failure, requiring 4,436,777,100,798,802,905,238,461,218,816 steps. Those numbers scared the sh!t out of me!😱 And if they don't scare you, then you haven't been paying attention. I finally just dumped all the standard regex models and built one completely from scratch (yes, in C😛). My engine significantly reduces the amount of backtracking: the 100 A string is resolved in just 1295 steps. 😁😁😁I'm still benchmarking it, but this engine is incredibly promising.........🤞
    1
  94.  @JohnDoe-yx4hi  As I stated, "^(A+)*B" is a pattern from Rexegg, and you'll find similar patterns on other regex sites, e.g. Mozilla. Yes, these are deliberately bad patterns designed for the purpose of demonstrating how engines can fail with catastrophic backtracking. This is a security issue. So, if you're designing a regex engine and you're not testing for "bad" patterns, then you shouldn't be developing a regex engine at all. Not if you intend for it to be used by others.
    1
  95.  @JohnDoe-yx4hi  I'd responded to you twice, and my comment was deleted both times. Have no idea why. Let me just say that that pattern is deliberately bad for the purposes of building a robust/secure regex engine.
    1
  96. I've been coding for over 3 decades and can honestly say that I've never found it "fun." I code to solve problems. Coding isn't fun. Reaping the benefits of all that hard work is fun.
    1
  97. JS frameworks are abstractions. And they're built on top of abstractions on top of abstractions..... I actually am not a fan of webdev for this reason.
    1
  98. 14:30 Looking at you, webdev
    1