Comments by "Vitaly L" (@vitalyl1327) on "ThePrimeTime" channel.

  1. Troll reacting to a troll video - that's meta!
    116
  2. ​ @MrEnsiferum77 developers who did not use version control systems in the 90s were trash developers not worthy of being mentioned. We had CVS ffs, and RCS before tbat.
    110
  3. There's a worryingly massive overlap between people who code in javascript and people who have no idea how binary floating point works. Almost as if web development attracts the most unscrupulous types.
    69
  4.  @redpillsatori3020  logic IS mathematics
    55
  5.  @possumkeys  and these days even assembly can be too high level and too distant from the actual microarchitecture of your CPU. Learning an assembly on an 8bit MCU can be far more beneficial for understanding of how things work than learning an x86_64 or aarch64 assembly.
    54
  6. Saying that one should not learn to code because there are LLMs that code is the same as saying you should not learn mental arithmetic because calculators exist.
    53
  7. I believe the funny myth of 10x engineers appeared simply because of the existence of the 0.1x engineers (thanks to all those bootcamps and alike).
    52
  8. Switch to hardware design, then you'll have over 10 hours builds and multiple days worth of tests running.
    51
  9. In my experience, anyone claiming to be a 10x developer turns out to be an a-hole who make it impossible for anyone to work with him, so he's the only one producing code while the rest of the team recoils in disgust.
    38
  10. Interesting, you really seem to believe that C is a very low level language. It is not. Not any more. Not since PDP11. It is very far from how real CPUs work.
    36
  11. I can sort of see where he's coming from. As a primarily hardware engineer, I do indeed feel this frustration - having all the cool formal verification tools in hardware and seeing software world being so reluctant to adopt even the most primitive forms of it makes one ponder about the future of the software engineering. Software engineering is so resistant to become, well, an actual engineering, where responsibility and safety is a key.
    33
  12. I'm so glad I don't care about cloud computing at all.
    30
  13. ​ @SolidWorksMastery-hr4sg can you imagine what kind of utterly crappy job it is if a bootcamp graduate is fit for it?
    29
  14. And there are things that C does better than C++. For example, restrict keyword - there is no portable C++ equivalent.
    28
  15. enshittification started with Python and JavaScript replacing Perl and Tcl.
    27
  16. Zig does not have procedural macros, Rust does. Any language without procedural macros is infinitely inferior.
    24
  17. I doubt an experience of a java monkey amounts to anything valuable though...
    24
  18. A valid C++ implementation is below 0.2s. Does not look good for JVM, does it?
    23
  19.  @Asto508  Yeah, sure, find me an equivalent of MISRA-C tooling first, then I'll switch.
    20
  20.  @devnexen  I am talking about proper Lisp-style macros, not that preprocessor abomination. Rust procedural macros are much closer to the proper macros.
    16
  21.  @arianghorbani1305  for some people 2+2 is hard. So what? It means they tried to memorise what 2+2 is, instead of understanding Peano axioms. Simple things are objectively simple. If they're hard for someone, they're either dim, or someone deceived them and gave an incorrect definition of that simple thing, making it needlessly complicated.
    16
  22. Inspired? Inspired?!? I say it's just a blatant plagiarism, C just copied javascript and removed all good parts from it.
    16
  23. ​ @teckyify  it can and will happen, but javascript coders are not even aware of memory leaks they are causing, while industrial C programmers have MISRA C. And, anyway, all such things must be coded in PLC languages.
    15
  24.  @baxiry.  nope, it does not. Macros are much more than just some compile time computation. Macros are for implementing embedded DSLs, i.e., macro is a compiler that runs in compile-time, and does all the things a standalone compiler can do, including arbitrary optimisation.
    14
  25. There is a natural difficulty in many real-world problems. When faced with such, both cowboy engineers with a tendency to overcomplicate things and superficially learnt monkeys are equally helpless, and only those with a thorough findamental knowledge thrive.
    14
  26. Sounds like a hideous idea for a sadistic reality show.
    14
  27. exactly, I don't remember any data strucutre questions 10 or 15 or even 20 years ago. That's likely solely for freshly out of university, since what else would you ask them anyway?
    14
  28. It is the LinkedIn equivalent of what spells as "School of Hard Knocks" and "University of Life" on some other social networks.
    13
  29. Ever seen a rotating drum memory?
    12
  30. Is not it a mandatory rite of passage for any rust convert?
    12
  31.  @bitwize  LLVM IR - yes, sure. WASM - absolutely not, it does not even allow irreducible control flow.
    12
  32. I worked with people who earned a million or over a year, who dropped out of this career to burnout (not software development, though, finance). I understand them. "Early retirement" is still years ahead, but youth is going away and never coming back.
    11
  33.  @baxiry.  to clarify - the critical part comptime is missing is functions that take AST as an input, and they don't produce AST as an output. Also, no way to access the current compilation context, making this thing inferior and pretty much useless in comparison to proper Lisp-style macros that Rust is (nearly successfully) trying to emulate.
    11
  34.  @kanji_nakamoto  Huh? It does not allow to do things in a more succint way. Any proper meta-language is infinitely more expressive than a fixed language. A meta-language, language with proper macros, can be easily turned into any language imaginable. It is an infinite power. It is funny when people even compara meta-languages with simple languages. Zig does not allow to construct efficient embedded DSLs, which makes it severely less productive than any language that have proper macros.
    11
  35.  @alexanderlanz9712  The fun part here is that although safety requires an upfront investment, it cost less in a long run. Bad quality code piles up problems and explodes complexity until you cannot maintain it any more. High quality code causes much less problems in long run. Too bad code monkeys very rarely think long term.
    11
  36. even a puny little frontend developer can write a code that hurts people. My eyes and my aesthetic feelings are hurt almost every time I browse web. Not to mention my time wasted on navigating badly designed UI with too mamy steps to do simple things. All developers must be regulated and treated the same way as doctors and civil engineers.
    11
  37. ​ @svuvich everything. Chosing a platform based on familiarity instead of purely techical characteristics is a sign of incompetence.
    10
  38.  @moonasha  machine code is a representation of the same assembly, you're not learning anything new from it. Instructions are still going to be executed in extremely convoluted ways, not in a sequence you wrote them, split into microinstructions you don't even have a chance to know anything about. The better way to learn is to dissect something like nand4tetris, my4th or Ptoject Oberon, where you'll understand exactly how instructions are executed. Or, if assembly is as low as you want to get, do it with some very simple MCU.
    10
  39. What "todays CPUs"? Try cortex-M0, which does not even have an FPU. Also, even the fattest modern CPUs don't have a pipelined FDIV - so in rare cases where you have many square roots in a row, you won't get any reasonable throughput, and latency of FDIV is still dozens of clock cycles.
    9
  40.  @jorper2526  boot loop on a failed kernel is 100% on Microsoft. A proper OS would have reverted to a last known good root fs state (and a last good bootloader state). Not to mention a watchdog that must detect boot loops.
    9
  41. ​ @jordixboy some people have an ability (and a desire) to build systematic knowledge, starting from the first principles and leaving no gaps. The others just memorise (believing they "learned" something) a random assortment of factoids from some easy high level subjects, and then they get confused when presented with a slightly harder problem or just something new. Don't be like them. Learn things properly. Even if it hurts.
    9
  42.  @anon1963  this is a wrong analpgy.
    9
  43. Tomorrow the system need to actually work, today it is enough to dazzle VCs with a shiny MVP that does nothing of a substance. Yes, sure, don"t solve that tomorrows problem, it is scary.
    8
  44. I think people who feel they're competing with self-taughts, bootcamp graduates and stochastic parrots are simply in the wrong industry and should have never been here in the first place.
    8
  45. Going top down is an extremely bad idea and it is going to leave you with an awful lot of gaps that you won't even be able to detect, making you very prone to Dunning-Kruger.
    8
  46.  @possumkeys  if the car breaks down the guy who built it waits for mechanics just like the others, but knowing how did it break and how to fix it if only he had his workshop with tools.
    8
  47. There is a lot of very simple optimisations giving you a good enough baseline. Convert your IR to an SSA, and this alone will give you tons of useful optimisations alnost for free. There are hard things, like loop fusion, polyhedral analysis, etc., but you can get a long way without ever needing them.
    7
  48. SQL lost its mathematical roots. It is not a complete relational algebra. If you want a pure relational algebra, it's Datalog, and it's much better than SQL in all possible ways.
    7
  49.  @momofomomofo  their update process is an utter crap and those who designed.it.are criminally negligent. And Microsoft is equally liable here for not making overlay updates.a default. They don't even have an overlay fs to start with!
    7
  50. Unpopular opinion: the vast majority of people who believe they experience this "impostor syndrome" are impostors indeed, and they're right to feel inadequate. Painting it as an "impostor syndrome" is their coping, nothing else.
    7