• vinsombra@lemmy.world
    link
    fedilink
    English
    arrow-up
    2
    ·
    36 minutes ago
    As democracy is perfected, the office of president represents, more and more closely, the inner soul of the people. On some great and glorious day the plain folks of the land will reach their heart's desire at last and the White House will be adorned by a downright moron.
    

    H.L. Mencken, On Politics: A Carnival of Buncombe

  • SabinStargem@lemmy.today
    link
    fedilink
    English
    arrow-up
    6
    ·
    2 hours ago

    If it were me, I would have started with the least used routes for rolling out, confirm that they are in good order after a couple months, slowly expand the rollout locations over time. All the while being administered and monitored by experts whose sole purpose is to assess everything - cost, tradeoffs, problems, improvements, personnel training, the whole enchilada.

    My approach isn’t nearly good enough, yet Duffy is even below that questionable bar.

    🤨

    • NotAnOtter@lemmus.org
      link
      fedilink
      English
      arrow-up
      4
      ·
      2 hours ago

      There are so many ways to migrate. Migrating systems is very standard practice in software and there are many good options.

      “Make all traffic use the new system in the middle of the day” is among the worst available options.

  • Blackmist@feddit.uk
    link
    fedilink
    English
    arrow-up
    2
    ·
    2 hours ago

    I’m sure Claude thought it was a good idea. Everyone who knew better has likely been fired or “retired” long ago.

    • Corhen@lemmy.world
      link
      fedilink
      arrow-up
      3
      ·
      2 hours ago

      Grok i bet, but yes.

      ‘yes, the new system needs a stress test, so completing it while its busy allows you to complete both at the same time’

  • Hemingways_Shotgun@lemmy.ca
    link
    fedilink
    English
    arrow-up
    8
    ·
    4 hours ago

    Because every aspect of the current American system is a clownshow entirely peopled by unqualified boot-lickers…

    Does this make me Sherlock?

  • matlag@sh.itjust.works
    link
    fedilink
    arrow-up
    1
    ·
    2 hours ago

    It’s the exact same answer as to the previous question: “why deploy a data hungry system (SMART) on an infrastructure that was guaranteed not able to support the load (old copper network)?”

    Because incompetent fucking idiots make loads of stupid decision all the time.

  • Gpennyhigh@lemmy.world
    link
    fedilink
    arrow-up
    3
    ·
    5 hours ago

    He’s another one of the know-nothings maga loves, since they sound as misinformed and clueless as they are, very tribal.

  • melsaskca@lemmy.ca
    link
    fedilink
    arrow-up
    13
    ·
    11 hours ago

    "I’m just surface lying here. I never gave it too much thought, as I don’t really know the patterns of “working for a living”. ".

  • TheHiddenCatboy@lemmy.world
    link
    fedilink
    English
    arrow-up
    118
    ·
    edit-2
    1 day ago

    So. One of the things DOGE got rid of was the CAB - Change Advisory Board. I’m pretty sure there were established Change Windows with approvals required and a documented test plan and rollback plan. Don’t need those! That just gets in the way of business! Let’s get rid of it and see how that goes!

    Yeah. We saw how it went, alright.

    Edit: I should have said it’s apparent that DOGE got rid of CAB. Because this isn’t the kind of thing you do with an Information System if you have a CAB.

    • BrianTheeBiscuiteer@lemmy.world
      link
      fedilink
      arrow-up
      28
      ·
      1 day ago

      I work at a Fortune 500 company and we have CAB meetings all the time. It’s a bit of a pain but a necessary part of maintaining stability in Production systems.

    • Rhaedas@fedia.io
      link
      fedilink
      arrow-up
      21
      ·
      1 day ago

      It seems blowing up things for progress doesn’t work all so well outside the rocket industry. I guess military, but only certain things should blow up there.

      • TheHiddenCatboy@lemmy.world
        link
        fedilink
        English
        arrow-up
        11
        ·
        1 day ago

        It doesn’t even work that well in the Rocket Industry, honestly. And Elon knows that! The Falcon 9 rocket is a stable ‘release’ that’s not experimented on. Sure, Starship isn’t, but Starship isn’t yet a production rocket. He can ask Vulcan Centaur what happens when you have a glitch in a production model rocket. That’s gonna be grounded until they find out what broke on the launch yesterday and how to fix it going forward!

        • Rhaedas@fedia.io
          link
          fedilink
          arrow-up
          7
          ·
          1 day ago

          The thing missing was that with new rocket designs you expect some booms, but you take that data and you find out why so the next boom isn’t from the same thing. Which in the case of anything they’ve touched in the FAA or other places, is not what’s happening. It’s a free for all sponsored by the prevailing administration. There was not a goal to fix and improve, but to make a show of it (at a minimum, I’m sure conspiracies abound on the real purposes).

          I often see SpaceX as a separate thing from Musk, mainly because he’s been smart enough to let them run themselves more than the other companies. And it shows.

          • Mirshe@lemmy.world
            link
            fedilink
            arrow-up
            2
            ·
            4 hours ago

            You actually want to do component-level testing instead of an all-up, so theoretically you find the thing that makes it go boom without it going boom.

            Failure to Launch is doing a large multi-part on how NASA insisting on doing all-ups with the shuttle (because it’s easier to explain to media and budget dudes than component testing) was part of what led to Challenger happening. The issue with an all-up in rocketry without the component testing before means that when something goes boom, you’re now doing a huge deep dive on production data from subcontractors of subcontractors of contractors, and whatever you can piece together from the exploded components.

            • AA5B@lemmy.world
              link
              fedilink
              arrow-up
              2
              ·
              3 hours ago

              With the shuttle, you had less than half a dozen ships, all hand built. Component testing would not have been as likely

              The SpaceeX approach is all about “mass production”. Everything is componentized, everything is made to scale. They can easily component test an engine, for example, because there are always more and it is somewhat independent of the ship

            • Rhaedas@fedia.io
              link
              fedilink
              arrow-up
              3
              ·
              3 hours ago

              Good points, full building to fill isn’t great, but the assumption is that heavy testing is done before that point, so the failure isn’t a known one, but to learn a new factor to design for. The advantage now is the data retrieval, everything has a sensor or camera so it’s far easier (still difficult) to find the cause. Whereas a big part of the Challenger aftermath was going off of media footage to find the spot that led to looking closer at the o-rings. There is also the issue that some were already telling them to not launch, but that was quieted down because of the focus and they didn’t want to drag it out more (without looking, it had been delayed a few times I think). So that was more a production run that pushed past recommended limits.

              • AA5B@lemmy.world
                link
                fedilink
                arrow-up
                2
                ·
                3 hours ago

                There’s always pressure to ship! And it’s not like there’s a sudden cutoff between safe and not safe, nor was the o-ring failure an expected scenario. Do you really hold a critical launch because someone is being paranoid about temperature?

                Being somewhat facetious with that but it’s easy to see how the wrong decision was made and I know in would not have done better

    • santa@sh.itjust.works
      link
      fedilink
      arrow-up
      5
      ·
      1 day ago

      They got rid of CAB?! We have that for technical stuff at work that would not kill anyone if there were an issue. Wow.

    • NABDad@lemmy.world
      link
      fedilink
      English
      arrow-up
      4
      ·
      1 day ago

      Do you have references for that? I can’t find anything that says DOGE eliminated CABs.

      • TheHiddenCatboy@lemmy.world
        link
        fedilink
        English
        arrow-up
        11
        ·
        1 day ago

        Oh you got me. I don’t have any proof that CAB was gotten rid of…

        …except the fact that they made an impactful change during working hours and spent hours trying to back it out with the system being down. I should have said that one of the things DOGE apparently got rid of was the CAB. Because CAB would NOT have let this happen without LOUD protestations.

  • JeeBaiChow@lemmy.world
    link
    fedilink
    arrow-up
    6
    arrow-down
    1
    ·
    17 hours ago

    Given the global nature of the industry, is it even realistic to wait for a ‘down’ period? I suppose for local traffic, which would be the majority at international hubs. Would be interesting to see the graphs.

    • AA5B@lemmy.world
      link
      fedilink
      arrow-up
      1
      ·
      3 hours ago

      If you don’t run your change on Friday, at the very least you have people available to deal with any issues, without paying overtime.

      But yes, I’m sure sometimes are quieter than ithers

    • hyhachi@sh.itjust.works
      link
      fedilink
      arrow-up
      13
      arrow-down
      1
      ·
      17 hours ago

      im no stoner, but id put money on 4:20am est being a better time to run that than when they did. /s jk am stoner

  • ikidd@lemmy.dbzer0.com
    link
    fedilink
    English
    arrow-up
    21
    ·
    23 hours ago

    Wasn’t it even on a Friday afternoon?

    Definitely FNGs. Nobody that’s been in IT for 10 minutes updates shit on a Friday afternoon.

    • wampus@lemmy.ca
      link
      fedilink
      English
      arrow-up
      5
      arrow-down
      3
      ·
      10 hours ago

      Depends. I’d say that logic applies to the old IT people, ones who came in / had high salaries and valued their personal time with family etc. The people who implemented things like read only fridays.

      Younger workers, or people chasing money due to lower salaries/job insecurity though… an issue translates as OT, and as an easy marker to point at for “what you do here” to people trying to cut jobs. Even the older workers know, when shit works smoothly people wonder what you do – but when it breaks, they’re all thankful you fix it.

      Similar mentality as I’ve heard with cops. They’ll bust a kid on an inflated minor charge near the end of their shift, because the arresting officer is required to fill in the paper work, which translates as hours of OT pay.

      The whole point of read only Fridays, is that you know shit can go wrong, and you don’t generally want to have a bunch of people doing unplanned OT on the weekend. It’s based on a shared understanding that personal time is important, and that no one desperately needs/wants unplanned OT. So if enough people in the mix don’t care so much about personal time, and/or want that extra cash… so long as you can deflect the problem as a fault with “decades old routers”…

    • deeprlyeh@lemmus.org
      link
      fedilink
      English
      arrow-up
      14
      ·
      1 day ago

      This is all planned in P2025. It seems like incompetence and it is, but it is also their plan.

    • dan1101@lemmy.world
      link
      fedilink
      English
      arrow-up
      10
      ·
      1 day ago

      Perhaps, but it also fits the playbook of screwing something up and then asking for billions of taxpayer money to fix it.