[monocasa]

If only. I work in the same town as a big medical device manufacturer, and several co workers over the years had worked there and immediately nope-ed out on moral grounds. Stuff like panics on anything out of the order in a morphine pump that defaults to full on while resetting. Apparently those killed a few people.

Meeting the FDA guidelines is more about finding the cheapest way to technically meet the spec rather than trying to build something safe.

[gmueckl]

I do get that such a device panics on the smallest error. But then it is supposed to go into a safe mode. "Full on" does not seem safe to me. Full off and emitting an acoustic alarm until it is actively acknowledged would be the right thing to do. Whoever designed this thing to do what it did was frankly a morron.

But gaming the certification process is unfortunately also a thing. In the EU the certification is performed by private companies who are themselves certified by the government for this job. The kicker is that they are competing against each other on a free market. Potentially shopping around for the most lenient certification process could be a thing. I haven't witnessed it yet but it certainly is possible in that system. The thought alone scares me.

[agumonkey]

Oh yeah EU certifications are sadly super weak. Now to your original point, I think it's quite possible to have an open review system. Linux and the likes have shown great capabilities in finding and fixing issues fast. With a national effort to ensure paid engineers it's not science fiction.

[agumonkey]

You know this mindset is becoming the norm. Everything is about doing the minimum to be legal and avoid lawsuit.