Re: += in ada

["Robert I. Eachus" <rieachus@comcast.net>]

Hyman Rosen wrote:

> I was simply pointing out that here was a real-life Ada software
> project that devoted plenty of energy to counting cycles, rather
> than leaving it to the compiler.

And my point was that they didn't.  They reused the code from the Ariane 
4 without change, assuming that way they didn't have to repeat the cycle 
counting.  After all the same code on the same hardware would produce 
exactly the same results.

Of course, the problem was that the guidance hardware was identical, but 
the trajectory, inertial moments for the stack, and maximum safe engine 
deflection were not.  That made all of their software reuse assumptions 
invalid.  The next two Ariane 5 failures were hardware failures, but the 
root cause was identical.  You can't change part of the system, and then 
assume that the requirements don't change elsewhere.  (The most recent 
failure flight 157 in December 2002 dumped two satellites in the 
Atlantic Ocean.  Preliminary data indicated a cooling problem in the 
uprated Vulcan 2 engines, but a final report has yet to be published.)

-- 
                                          Robert I. Eachus

"Quality is the Buddha. Quality is scientific reality. Quality is the 
goal of Art. It remains to work these concepts into a practical, 
down-to-earth context, and for this there is nothing more practical or 
down-to-earth than what I have been talking about all along...the repair 
of an old motorcycle."  -- from Zen and the Art of Motorcycle 
Maintenance by Robert Pirsig