they almost certainly have a hardware spare, or at the very least, an accurately simulated version of it, because again, this is 50 year old hardware. So it’s pretty easy to just simulate it.
But yeah they are almost certainly pulling some really fucked QA on this shit.
I read someplace a while back that the average beginner dev has an error for every 10 lines of code, a working dev, 100, the (I think) US Air Force 1000. NASA (& company )was at a massive single error per 100000 lines of code. I wish I could find that article.
Keep in mind too these guys are writing and reading in like assembly or some precursor to it.
I can only imagine the number of checks and rechecks they probably go through before they press the “send” button. Especially now.
This is nothing like my loosey goosey programming where I just hit compile or download and just wait to see if my change works the way I expect…
they almost certainly have a hardware spare, or at the very least, an accurately simulated version of it, because again, this is 50 year old hardware. So it’s pretty easy to just simulate it.
But yeah they are almost certainly pulling some really fucked QA on this shit.
I read someplace a while back that the average beginner dev has an error for every 10 lines of code, a working dev, 100, the (I think) US Air Force 1000. NASA (& company )was at a massive single error per 100000 lines of code. I wish I could find that article.
NASA has claimed to have never written a bug in a shipped piece of code from what i can recall off the top of my head.