• Warl0k3@lemmy.world
      link
      fedilink
      arrow-up
      23
      ·
      vor 5 Stunden

      Fucking killed me how he started the presser by claiming that $13 Billion over 25 years wasn’t enough to fix this, but then at the end tried claiming that if we “just give him $12 Billion we’ll have it fixed immediately” or whatever that was. Oh, and the parting shot where he clarified this is an isolated network so it could not have been cybercrime but also the overloading of the network infrastructure had nothing to do with the AI they just dumped on that isolated network. Well fucking gee golly, if it wasn’t the massive network traffic generating machine, then what was it, Duffy? It’s an isolated network, Duffy. There’s not that many options.

    • Fmstrat@lemmy.world
      link
      fedilink
      arrow-up
      7
      ·
      vor 4 Stunden

      Wait until you learn the AI startup funneled a ton of money into the Transportation Secretary’s son-in-law’s campaign.

  • Archer@lemmy.world
    link
    fedilink
    arrow-up
    9
    arrow-down
    1
    ·
    vor 5 Stunden

    The implication that someone is saying positive things about Bill O’Reily is horrific

    • gex@lemmy.world
      link
      fedilink
      arrow-up
      7
      ·
      vor 3 Stunden

      All devs test in a dev instance, but only some of them have a separate production instance

  • Thebeardedsinglemalt@lemmy.world
    link
    fedilink
    arrow-up
    16
    arrow-down
    1
    ·
    vor 6 Stunden

    I work in IT. I refuse to move something to Production until it’s thoroughly tested and signed off on. Because every single time someone says “we’ll test it in Live” it leads to massive failures and issues they can’t fix. Every. Single. Time.

    • otacon239@lemmy.world
      link
      fedilink
      arrow-up
      16
      ·
      vor 5 Stunden

      The number of things that go wrong while testing is about the same, but the difference is they go wrong on your computer instead of millions of customer’s computers.

    • cantstopthesignal@sh.itjust.worksOP
      link
      fedilink
      arrow-up
      12
      ·
      vor 6 Stunden

      And you know, you can keep the old system as a back up in case it breaks when it goes live. I guess that’s too much to ask of the agency that keeps all of the airplanes from crashing.

    • shalafi@lemmy.world
      link
      fedilink
      English
      arrow-up
      2
      ·
      vor 4 Stunden

      They did test SMART and they weren’t trying to roll the whole system at once. It worked fine in DC and FL for over 2 months. It worked fine for most of this week.

  • shalafi@lemmy.world
    link
    fedilink
    English
    arrow-up
    4
    ·
    vor 4 Stunden

    The first stage of the SMART rollout had been working fine since late September in DC and parts of Florida. They deployed that stage to the rest of the country on the 10th and it worked fine for 3 days.

    Given that, I’ll withhold judgement until I hear more.

    • Triumph@fedia.io
      link
      fedilink
      arrow-up
      6
      ·
      vor 4 Stunden

      When you’re the FAA, you still don’t decommission the old system until the new one has been in production for a long time. They left themselves without a fallback position.