Hacker News Reader

Comment by xg15 | Hacker News Reader

But who would that "someone" be, and to what conditions? Or do you expect the people in the article to just write the drivers for their own implants themselves?

View on HN

> Or do you expect the people in the article to just write the drivers for their own implants themselves?

No, but a community of volunteers might. ThreadX [0] is proof enough that an OSS community can keep a piece of software certified under strict security standards. I don't see why the same couldn't be achieved with software for medical devices.

[0] https://threadx.io/faq/#what-happens-with-the-existing-safet...

A couple reasons come to mind. (1) you can't separate the software from the rest of the device: it has to be Approved as a whole. (2) certification isn't enough. e.g., you could submit the software to a quality agency who would certify that it meets IEC62304, but then the entire device and its clinical test data still has to be submitted to the FDA, along with its Quality Record and all that (submission process, maintaining the Design History File, etc.) costs Real Money.

And if you manage to get it approved for sale, then there are regulatory requirements around support and maintenance of the device that aren't going to happen for free.

It's a nice idea, but unless you get funding from a philanthropic billionaire, I don't see it happening

Sounds like a good reason for these theoretical grassroots programmers to circumvent the regulatory agencies that are more of a hindrance than a help. ‘We cant do that because bureaucracy’ is such a non-starter.

If you think that the FDA is more of a hindrance than an help, then I guess I have nothing else to say.

Anyone under whichever conditions they choose.

We, society, grant temporary monopolies in the form of copyright/patents "To promote the Progress of Science and useful Arts". They do not intend to further use that monopoly, so they lose it.

You’re missing the point. The chance of someone investing in the development of drivers for niche failed outdated niche product is nil.

Untrue. Even small communities have shown to be able to do maintenance on whatever piques their interest. 250 customers with a high stakes product is large enough for them (or their insurer) to collectively hire say one programmer and one mechanic. That would be more capacity than 50% of FOSS.

You're perhaps underestimating vision scientists. If the problem is a driver (or something else that doesn't require a surgeon), I'm not the only person who might volunteer to develop the driver, given the alternative of people returning to blindness.

Still strictly better than the current situation.

Public trustee. That’s how it works with other assets, like apartment buildings with people living in them. Often public trustees will outsource the management & disposal of these assets. Hopefully for the people depending on the assets, “disposal” means “sale to a new owner.”

Maybe another company? It seems this company was badly ran with nice tech. But at least if open, we have a chance. If closed, nothing happens and things get lost.

I wouldn't necessarily have any expectations - the point of FOSS is to have those options at all.

What do you think might be done to avoid this sort of thing?

Presumably James Weiland [0]. He's the (excellent) USC engineer that moved to Michigan in the fallout of the dean Puliafito scandal [1]. Puliafito was great at raising funds by conning rich people, but uninterested in supporting legitimate scientific research. His antics would make Hunter Biden blush.

[0] https://weilandresearch.bme.umich.edu/ [1] https://www.latimes.com/local/california/la-me-usc-doctor-20...

Reply on news.ycombinator.com