4 ms·
Is it me or certification is a very powerful way to keep control on the open sourced code ? I mean it's open so you can modify it but as soon as you touch one l
by wiz21c 3y ago
Is it me or certification is a very powerful way to keep control on the open sourced code ? I mean it's open so you can modify it but as soon as you touch one line, you can't claim the certification anymore... Therefore only MSFT has the ability to change the code (because I suppose they don't have to re-do the whole certification); until another big player gets the certification.
Right ?
- eschneider 3y agoNot really. It's pretty rare to use an RTOS without making _some_ modifications and it's intertwined with your application anyways. If you need certification, you're going to have to get it for the combined app/RTOS/RTOS-modifications anyways. And you're in better shape starting from a base that you know passes certification.
- krylon 3y agoMy understanding is you need to re-certify in some form upon changing anything about the code. I'm not sure if it's the same process each time or if you can do some kind of incremental re-certification for minor updates. Either way, Microsoft can afford to do that if they want to. But I would (naively) expect the rules to apply to anyone who wants to modify the code base, I don't think Microsoft has any privileged position when it comes to that.
- j16sdiz 3y agoMicrosoft have access to the material they used in previous certification, this may make the process a lot cheaper.
- demondemidi 3y agoYes! There are some industries that will not tolerate seat-of-the-pants development. Functional safety requirements make a nice tall fence to keep out risky developers. Which is why china automakers have no such regulations.
- AlotOfReading 3y agoSorry to burst your bubble, but FuSa dev teams are often full of "risky developers" who frankly shouldn't be within mile of any codebase your life depends on. You can only hope the safety culture and tooling keeps them in check. There's genuinely good parts of the standards, but a lot of the rituals are just checklist items to assure certification authorities you have a defined process similar to what they've seen before, regardless of whether that's actually effective at issue reduction/detection. In many cases, they're actually counterproductive and encourage people to focus on minute, obscure error sources rather than addressing significantly more common issues like memory safety and undefined behavior.
- demondemidi 3y agoThanks for your insight! As someone who has lead FuSa certification on multiple automotive products, I can’t disagree with you more. But maybe you just had a bad experience.
- AlotOfReading 3y agoI've also led the process at multiple companies. It's a big enough space that there's definitely room for different experiences, even within the same companies. That said, I'm surprised you've never encountered issues with people e.g. not understanding how serious UB is vs other types of issues in C-family languages. That's fairly universal even among developers from what I've seen.
- demondemidi 3y agoI think you might be confusing software standards with functional safety standards. They are related, but one is required by the other. FuSa from my role has been predominantly architecture and process. Developers do need to follow coding and documentation guidelines. If they are fighting that then there are bigger problems with leadership.
- 3y ago
- Aurornis 3y ago> Therefore only MSFT has the ability to change the code (because I suppose they don't have to re-do the whole certification); This is incorrect. Microsoft does not have free license to change the code and retain certification. The certification is handled by a separate entity. The certification rules apply to contributions from everyone, including Microsoft.