4 ms·

Seems like the implants (hardware design and any software) need to be the equivalent of open source (in the GNU sense). Failing that, hardware designs, and the

Seems like the implants (hardware design and any software) need to be the equivalent of open source (in the GNU sense). Failing that, hardware designs, and the software - along with all necessary instructions - at least have to be put in escrow in the event that the company goes under. That would not immediately help, but at least there would be a fighting chance... As in this case to still be able to have someone fix/use the software to calibrate the implant. This should be standard regulation for approving such implants. And this can and should be paired with the other suggestions here such as mandatory insurance, etc. Edited for clarity

tracker1 4y ago
I would accept that all details are CCd to a govt office to be released if the company folds and it's obligations are not continued by anyone internally. And that any company buying said company needs to demonstrate intent, motivation and financial ability to maintain said products. Counter to the rest of technology where devices are literally designed to fail, this cannot be acceptable for embedded electronic devices.
gmueckl 4y ago
The design, documentation and especially source code should only be ever made available to entities with the knowledge and skill to maintain these implants safely and purely for the benefit of the users. If any random Joe Garage Hacker can get the source and devise patches for implant devices, the result will be an endless hellish nightmare for whoever has these implants in them. It'd be the IoT mess all over again, but with people's bodies, not just their homes.
- 63 4y ago
  That's just an argument for why it should be open source from the beginning: let bugs be spotted early and be sure the designs are as secure as we can make them before somebody gets saddled with one in their skin for life. Security through obscurity is not the answer.
  - gmueckl 4y ago
    That's exactly why the source code cannot be open to hacking by random people who suddenly wake up to a feeling of being qualified to work on something like that. There's a a general attitude of overconfidence among software developers that extends into other areas of expertise that they actually know nothing about. HN reflects this very well. And that's exactly why some things can never be allowed to be fully open. To whoever reads this comment, just ask yourself this: how much vetting and testing do you want for the code on a pacemaker or insulin pump? Do you think you can perform that level of testing at home? If you think that you can, you're exactly the kind of dangerous person that can't be allowed to mess with this stuff.
    - 63 4y ago
      I think we may be imagining different scenarios. I was operating under the idea of a company opening its code up to public scrutiny with enough time to accommodate any changes suggested by qualified members of the public. It seems you're imagining a world where medical devices are homebrew with no centralized authority?
bo1024 4y ago
Yes, strong agree. This is just the next generation of the FOSS and right to repair wars, with stakes raised (as was foreseeable).

tracker1 4y ago

I would accept that all details are CCd to a govt office to be released if the company folds and it's obligations are not continued by anyone internally. And that any company buying said company needs to demonstrate intent, motivation and financial ability to maintain said products. Counter to the rest of technology where devices are literally designed to fail, this cannot be acceptable for embedded electronic devices.

gmueckl 4y ago

The design, documentation and especially source code should only be ever made available to entities with the knowledge and skill to maintain these implants safely and purely for the benefit of the users. If any random Joe Garage Hacker can get the source and devise patches for implant devices, the result will be an endless hellish nightmare for whoever has these implants in them. It'd be the IoT mess all over again, but with people's bodies, not just their homes.

63 4y ago
That's just an argument for why it should be open source from the beginning: let bugs be spotted early and be sure the designs are as secure as we can make them before somebody gets saddled with one in their skin for life. Security through obscurity is not the answer.
- gmueckl 4y ago
  That's exactly why the source code cannot be open to hacking by random people who suddenly wake up to a feeling of being qualified to work on something like that. There's a a general attitude of overconfidence among software developers that extends into other areas of expertise that they actually know nothing about. HN reflects this very well. And that's exactly why some things can never be allowed to be fully open. To whoever reads this comment, just ask yourself this: how much vetting and testing do you want for the code on a pacemaker or insulin pump? Do you think you can perform that level of testing at home? If you think that you can, you're exactly the kind of dangerous person that can't be allowed to mess with this stuff.
  - 63 4y ago
    I think we may be imagining different scenarios. I was operating under the idea of a company opening its code up to public scrutiny with enough time to accommodate any changes suggested by qualified members of the public. It seems you're imagining a world where medical devices are homebrew with no centralized authority?

bo1024 4y ago

Yes, strong agree. This is just the next generation of the FOSS and right to repair wars, with stakes raised (as was foreseeable).