• BassTurd@lemmy.world
    link
    fedilink
    arrow-up
    48
    arrow-down
    1
    ·
    1 day ago

    I barely proof read anything I type on my phone, and my comment history is a testament to that. I deploy code or system changes most days, but I proof read the shit out of those on top of the QC they goes through. Any company worth anything will have a process for reviewing and approving anything being deployed, or probably destroyed for that matter.

    • almost1337@lemmy.zip
      link
      fedilink
      arrow-up
      14
      ·
      edit-2
      7 hours ago

      Up until recently I worked for a company worth anything, and you would be surprised at how many major outages were caused by either skipping the process or gaps in the process.

      You know that adage: “the safety rules are written in blood”? The same is true for change processes, just with a cost measured in dollars instead of human injury/worse.

      • Yggstyle@lemmy.world
        link
        fedilink
        arrow-up
        1
        ·
        3 hours ago

        So we called this meeting to discuss a change in procedures. HR would like us to remind you that this is entirely unrelated to last week. Also has anyone seen Joe? We’d like him to be here for this. No reason why.

      • Sergio@piefed.social
        link
        fedilink
        English
        arrow-up
        10
        ·
        21 hours ago

        Or sometimes there are just multiple failures. That’s what I learned from reading Admiral Cloudberg about air disasters: even if you have n safety measures, there’s still the chance that there’ll be n+1 failures.

    • Earthwormjim91@lemmy.world
      link
      fedilink
      arrow-up
      4
      ·
      18 hours ago

      Code should ideally be going through tests before prod anyway. There should be no code changed from successful test to prod. Proofreading shouldn’t matter at that point. Just scheduling the actual deploy.

      • BassTurd@lemmy.world
        link
        fedilink
        arrow-up
        4
        ·
        18 hours ago

        I mostly mean proof what I’ve written prior to having someone else test. I often will comment out lines when trying different things so I just make sure I clean up what I’ve done. We have a few human checks as well as some automated checks between each stage of deployment for each environment.

        • Earthwormjim91@lemmy.world
          link
          fedilink
          arrow-up
          1
          ·
          18 hours ago

          Yeah that’s what the MRs are supposed to be for. To catch those and proofread.

          There shouldn’t be any changes at all from the last test to going to production though. Even cleaning up comments.

          • BassTurd@lemmy.world
            link
            fedilink
            arrow-up
            1
            ·
            18 hours ago

            Correct. I’m just saying that I proof read my work, that I deploy things, and that I don’t proof read my texts.

            • Earthwormjim91@lemmy.world
              link
              fedilink
              arrow-up
              2
              ·
              18 hours ago

              No I got that. I’m saying that by the time the prod deploy comes around, there’s no proofreading left to do anyway.

              Not proofreading texts should have zero bearing on being able to write and deploy software because it should be proofread several times before the actual prod deployment. Hell it very likely isn’t even the same person doing the deployment that wrote the code.