5 ms·

> "But this isn't academically good code." As others have said, the only real metric of whether or not something is good is if it works in live production.

eternalban 7y ago
No, that's a metric for "something is working". Ease of maintaining, extending, or fixing bugs in the same software are not informed by that metric.
- chris_wot 7y ago
  Except in this case, where it was a one-shot, very specific objective. Once this thing was released, it could not have been easily updated.
  - huffmsa 7y ago
    It's a classic misunderstanding between "academically viable" and "operationally viable". The military doesn't need to get an A+. It needs to win. Anything else is a bonus. Which is why the A-10 is a better plane than the F-35. One shows up and BRRRTs the opposition into a fine red mist, when you need it to. the other makes it pilot motion sick as soon as they put the helmet on.
    - GhostVII 7y ago
      The A-10 and F-35 are completely different planes for completely different purposes. The F-35 attempts to be able to perform most of the things the A-10 does, but you can't just say one is better than the other overall. The A-10 is probably better at shooting ground targets at close range. The F-35 is better at bombing them from 5 miles away.
      - huffmsa 7y ago
        That's the problem though. The F-35 tries to be good at too many things. It has the latest sensors and stealth and armaments. But it's taking forever and a day to get the damn thing out the door because it's too academically excellent.
  - irishcoffee 7y ago
    The wired article linked in the top comment talks at some length about how the software was able to be updated in the wild.
- huffmsa 7y ago
  It is though. Bad products eventually fail under their expected operating conditions. That means you didn't actually make something production ready. It didn't meet the specified requirements. That's a bad product. If I have a car motor which runs, but throws a piston 20k miles before it should, I didn't build to the production spec.

eternalban 7y ago

No, that's a metric for "something is working". Ease of maintaining, extending, or fixing bugs in the same software are not informed by that metric.

chris_wot 7y ago
Except in this case, where it was a one-shot, very specific objective. Once this thing was released, it could not have been easily updated.
- huffmsa 7y ago
  It's a classic misunderstanding between "academically viable" and "operationally viable". The military doesn't need to get an A+. It needs to win. Anything else is a bonus. Which is why the A-10 is a better plane than the F-35. One shows up and BRRRTs the opposition into a fine red mist, when you need it to. the other makes it pilot motion sick as soon as they put the helmet on.
  - GhostVII 7y ago
    The A-10 and F-35 are completely different planes for completely different purposes. The F-35 attempts to be able to perform most of the things the A-10 does, but you can't just say one is better than the other overall. The A-10 is probably better at shooting ground targets at close range. The F-35 is better at bombing them from 5 miles away.
    - huffmsa 7y ago
      That's the problem though. The F-35 tries to be good at too many things. It has the latest sensors and stealth and armaments. But it's taking forever and a day to get the damn thing out the door because it's too academically excellent.
- irishcoffee 7y ago
  The wired article linked in the top comment talks at some length about how the software was able to be updated in the wild.
huffmsa 7y ago
It is though. Bad products eventually fail under their expected operating conditions. That means you didn't actually make something production ready. It didn't meet the specified requirements. That's a bad product. If I have a car motor which runs, but throws a piston 20k miles before it should, I didn't build to the production spec.