Jean-Christophe Filliatre wrote:
> There's a joke around in the formal methods community: ``would you
> prefer to get on a plane whose software has been proved correct or
> has been tested?''
>   
I heard, from someone working on an automated train system, that 
everyone working on the system was required to be on the inaugural run 
of said train.

I would trust such "people methods" even more than proof and testing!

Jacques