[Openmcl-devel] process-run-function and mach ports usage

Robert Goldman rpgoldman at sift.info
Thu Feb 24 15:48:03 UTC 2011


On 2/24/11 Feb 24 -8:59 AM, Raffael Cavallaro wrote:
> 
> On Feb 24, 2011, at 8:28 AM, Robert P. Goldman wrote:
> 
>> I was peripherally involved with some people trying to do high reliability os development for avionics, and as far as I can tell no one wants to pay what this kind of software engineering really costs. Not even in industries that are safety critical.
> 
> Hold on, do you mean to say that the wisdom of the market won't make the right decision here without some sort of regulatory intervention?
> ;^)

Actually, even regulatory intervention doesn't necessarily solve this
problem.  In many cases, high reliability software simply becomes
stagnant.  In many cases, high reliability applications use patched
software (and even computing platforms) from the 1970s because it's
prohibitively expensive to do a clean-slate rewrite and get certification.

If you impose a solution through regulation, and the market isn't
willing to pay the cost of complying with regulation, then you just get
stuck with the stuff that's already certified.

It's quite a challenge.  The "wait until my customers find the bugs for
me" philosophy has spread like a plague.

best,
r



More information about the Openmcl-devel mailing list