• abcd@feddit.de
    link
    fedilink
    English
    arrow-up
    5
    ·
    1 year ago

    This is so true!

    Last year I had a project to upgrade the PLC of a machine to the newest generation. As usual the customer was not able to tell me the requirements they had. They told me to look in the old software…

    It turns out it was 30+ years old software where you had to program in a cmd line (Siemens S5 if you know PLCs). I had to migrate everything to the next generation (S7) to be at least a little bit productive. Then I thought come on lets try to migrate to the current generation (TIA) to be even more productive.

    At the end everything was nearly ready to be compiled and uploaded to the PLC. So I fixed some minor compile time issues, deleted around 50-75% of the old program (old stuff which went obsolete), changed some variable names, refactored some stuff and here we are: The same 30+ year old software is running strong. 24/7 since 6 months without issues 😁

    • birdjazz@lemmy.world
      link
      fedilink
      English
      arrow-up
      3
      ·
      1 year ago

      Most projects written as a Demo, are still in production 3 years later. Still waiting for the “real” solution.

      If you spend a lot of time thinking of future options, nobody will ever need them.

      If you spend more time on the API with a partner, the partner will be replaced, probably before the solution is installed at all locations.

  • brokenneon@kbin.social
    link
    fedilink
    arrow-up
    4
    ·
    1 year ago

    I always say I write throwaway code that never dies. I shutter to think how many pieces of code I wrote 10+ years ago are still buried deep in systems running today. Shutter.

  • Emi@lemmy.blahaj.zone
    link
    fedilink
    English
    arrow-up
    2
    ·
    1 year ago

    I have a mental image of 50+ lines that could be replaced with 15 if someone just used a loop.

  • celipon@kbin.social
    link
    fedilink
    arrow-up
    2
    ·
    1 year ago

    It is always the worst code you wrote that survives. There’s a terrible university dorm management software I wrote eight years ago as a student. They still use it. The crazy complicated test framework wrappers for some hardware I wrote five years ago. They still use it. The godawful and crazy complicated communication protocol I whipped up four years ago, still used in medical equipment today.

    • masterspace@kbin.social
      link
      fedilink
      arrow-up
      2
      ·
      1 year ago

      The crappy scripts that I wrote while teaching myself to code at an electrical engineering / architecture firm are used more often than the professional software I’ve built for FAANG and Fortune 500 companies since.

  • alp@kbin.social
    link
    fedilink
    arrow-up
    2
    ·
    1 year ago

    I really think that optimization courses should have a special lecture in the optimization of optimization…

  • nobodyspecial@kbin.social
    link
    fedilink
    arrow-up
    1
    ·
    1 year ago

    As with relationships, if you make a mistake you’ll wind up supporting it for life.

    I kid, I kid. We all know to job hop every 2 years for better compensation. It’ll be someone else’s problem after that.

    • mfz@kbin.social
      link
      fedilink
      arrow-up
      1
      ·
      1 year ago

      Yes, yes, and someone else’s problem will be your problem after the job hop! :)