4 ms·

Imagine how terrible it would be for your customers once that happened in the production....

• konschubert 9y ago
  He said it's a surgical training startup so I think it would have been fairly okay.
  • Piskvorrr 9y ago
    Those things aren't cheap for simulators, either - not to mention knock-on costs. "What do you mean - I got the doctors in, which alone took a month of herding cats, and now it won't work, just because?" How low has the SW development bar gone, if "it's okay" now means "at least it's not directly killing people"?
    • babuskov 9y ago
      > How low has the SW development bar gone, if "it's okay" now means "at least it's not directly killing people"? I though that was the way ever since OS/2 failed. Getting stuff out to customers has priority over quality control.
      • johnchristopher 9y ago
        Disrupt, innovate, first-mover advantage, growth marketing, yadaa-yadaa.
    • FLUX-YOU 9y ago
      >How low has the SW development bar gone, if "it's okay" now means "at least it's not directly killing people"? It gets that low every time a hospital underfunds IT staff and makes horrible project management decisions and product buying decisions. I've seen that first hand. There's a bunch of corpses at the IT entrance of people who've tried to turn that around.
      • Piskvorrr 9y ago
        Hospital...corpses...I have a hard time distinguishing the literal and figurative context here.
    • jdietrich 9y ago
      There has always been a tradeoff between reliability and development time. There wouldn't be a games industry if every video game had the same level of software assurance as a mars lander, because Tetris would cost $200m to develop. A medical simulator lies somewhere between a mars lander and a video game - it needs to provide accurate simulation, but the odd crash isn't a complete dealbreaker. The bug is now patched, so the downtime appears to be less than 24 hours from discovery to fix. The original error is clearly a major blunder, but Oculus have responded properly.
    • konschubert 9y ago
      The GP was suggesting that this could kill people. I simply implied that it wouldn't, and compared to killing people, I would say a lost day is "okay".
      • Piskvorrr 9y ago
        I'll try that for my next programming blunder: "Sure, I've set back hundreds of people one day, but hey, didn't kill them! No big deal, they should even be grateful!" In other words, comparing to the worst possible outcome is, by definition, not a very high bar.
  • ams6110 9y ago
    Yes, but it's an interesting question to ponder. The last decade or two government and Military procurement has been leaning more heavily towards COTS (commercial off-the-shelf ) hardware and software. The thought was it's cheaper and possibly more reliable than the bespoke solutions that vendors were delivering in the past. Now we see that it isn't a guarantee of anything. Though still probably cheaper, at least initially. Something like this probably will happen with computer assisted surgery or medical procedures, or an aircraft in flight. Just a matter of time.