The stakes, of course, are the numbers that go in the little boxes of your game-theory pay-off matrix.  They should be chosen to represent reality.  

Amazing:  If you introduce a little clever, fancy, high-tech, everyone is so fascinated that they just don't want to fill in the boxes!  Payoff minus infinity?  But the chance is so low, why don't we just round it up to minus ten, and calculate with that instead?  

My mood here is a holdover from a different thread:  Science fundies are just as bad as religious fundies--they don't care about anything except their little toy theory/theology.  

A story:  A while ago I came across an entertaining book on a very dull subject--I still have it--called "Structures:  or why things don't fall down" by J.E.Gordon.  It's from the 1980s, but Gordon has been around a while--consulting on projects that had to work or investigating projects that didn't work.  (Oops!)  At one point he wanders onto a tangent about why architects don't use glass for structural (load bearing) members.

First he sets it up:  Glass is a great material--it is easy to work with, stronger than steel (at least by weight), and has many good properties.  Nevertheless, it is never used.  

The reason:  It's mode of failure.  When glass fails under stress, it fails without warning, all at once, and totally.  This is bad--it makes the structure untrustworthy.  Since all structures fail, someday, eventually, you need to use a material with a predictable life, and that gives a predictable warning--usually by exhibiting strain in advance of failure.  

The moral:  There are some technologies you cannot use, no matter how attractive they are, because their defects are too great.  

I am glad architects understand this.  It would be good if this technical principle were understood more generally by those who claim to be scientific.  

The Fates are kind.