Re:Therac-25
by aaronem on Tuesday February 18, 2014 @11:43AM
(#7259120)
I don't disagree in the slightest that the software had a deadly bug, or that that bug shouldn't have been allowed to make it into production. But I am surprised no one in this thread has yet identified the real source of the problem. From Wikipedia:
> The accidents occurred when the high-power electron beam was activated instead of the intended low power beam, and without the beam spreader plate rotated into place. Previous models had hardware interlocks in place to prevent this, but Therac-25 had removed them, depending instead on software interlocks for safety.
I am just a self-taught programmer, and certainly not any kind of trained engineer. But when there's something your machine must not be able to do, and it's possible to design it such that it physically cannot do that thing, what possible excuse can there be for not doing so?
> The accidents occurred when the high-power electron beam was activated instead of the intended low power beam, and without the beam spreader plate rotated into place. Previous models had hardware interlocks in place to prevent this, but Therac-25 had removed them, depending instead on software interlocks for safety.
I am just a self-taught programmer, and certainly not any kind of trained engineer. But when there's something your machine must not be able to do, and it's possible to design it such that it physically cannot do that thing, what possible excuse can there be for not doing so?