> He went on to point out that they had calculated the
amount of memory the application would leak in the total possible flight time
for the missile and then doubled that number.
I half wonder if the calculation cost more than just fixing the code.
When I worked in embedded the world was sadly most often: Design and order hardware first, adapt software and requirements later.
Appeared backwards to me but in the world of mass products of extremely high quantities it makes sense to get the best hardware deals you can get month and sometimes years in advance.
Buggy software getting re-used in another hardware configuration (with no physical safety interlocks) was one of the root causes of the Therac-25 disaster.
Or an adversary steals the code and refuses it. Actually pretty sure that's a real thing in the defense industry.
A friend of mine works on F16's. He says they don't really have half the plans to re-create or fix them and another third are just outright wrong, but instead they have to reverse engineer them. I exaggerate a bit. Still it's enough so even if another country stole all the plans they'd not be able to build their own F16s without spending massive amounts to re-engineer half of it and avoid the known issues.
That's long been America's main advantage. We can apparently spend insane amounts of money on this stuff, which would bankrupt other nations.
It's an odd but seemingly effective strategy. I'm not sure if it's planned or really just a happenstance of Washington politics. Also explains the F35 being absurdly expensive, maybe its a "feature" not a bug.
I half wonder if the calculation cost more than just fixing the code.
reply