[svlug] mission critical computing and air safety

[ian maclure]

On Wed, 22 Sep 2004 08:43:02 -0700, Rick Kwan <kwanrj03 at comcast.net
wrote:

> Which leads me to the conclusion... we're in for more of the same.
> We were simply lucky this time that no one died.

	By definition this was not a system critical failure since there
	was not an immediate possibility of death or injury due to
	the problem and there was a work-around.

	However, the thought that a commodity O/S with known security
	and reliability problems is being used for something like this
	is more than a little disturbing.

	Manned airline flight simulators for instance do not run 
	Windows. Some of CAE's latest products are PC driven but I
	believe they use an RTOS of some sort to do the actual real-time
	stuff. Standard Linux kernel ( + low latency patch ) 	implementations
have been driving the same class of simulator 
	for about the last decade with outstanding reliability.

	Wonder if there are any assault lawyers interested in this.
	M$ is after all sitting on a pile o' cash.

	IBM