This is a dead horse, but anyway:
So I can believe that there might have been some sloppy code, but
1. A random bit flip is a massively unlikely scenario. Modern circuitry has such rapid rise/fall times on the output flops that getting one to go the "wrong way" is incredibly unlikely. Once in a blue moon.
2. A random bit flip that just happens to affect the exact memory location or range of memory locations that just happens to cause the accelerator to go full blast without also shutting down fuel injection, timing, etc.? Once in a blue moon squared.
3. Even if we get past both of the above highly unlikely conditions, we're also supposed to believe that this condition persisted for so long that some sort of "auto reset" did not kick in? Keep in mind firmware runs on a millisecond or nanosecond timescale. Two seconds is an eternity in this world - a well designed piece should have detected that the thing is going bonkers (probably due to the locking/death of various worker threads). But as its possible that the code was terrible, we can let this one slide.
4. And now if we get past all of 1, 2, and 3, we're also supposed to believe that at the same time, the brakes, a system which while partially electronic can be operated in a fully mechanical manner, also happened to fail at the same time?
EDIT: keep in mind that the alternative explanation is that the driver just
up. You tell me which is more likely.
