On the surface, it’s one for the crypto geeks. But with a bit of thought can apply to every field. “We have a neat security precaution, but we want only our friends to be totally secure,” goes the line of thinking. “So let’s set things up so that most people hearing of this will do something subtly wrong that we can then use.”
(I should add that I don’t think this is a good solution.)
More useful is the opposite perspective: as defenders relying on the fruits of public research, the prospect of finding ourselves on the business end of such tactics is very real. Defending against them takes one of my favorite mantras of late — only trusting what we can understand.
“Lets say that you have an unbreakable cipher (or its closest approximation) and that you, Eve, have the ability to break all other known ciphers. There is a risk that if you use or deploy your unbreakable cipher it may be captured by your enemy Alice and thus prevent you from eavesdropping on Alices communication. What should you do?[…]
It is not unimaginable that a cipher could exist which is secure only under some of its keys2. For example consider a cipher that is secure when the first bit of the key is $0$ but totally insecure when the first bit of the key is $1$. […] Under such a system Eve could choose to only use the strong keys and Alice would on the average use a weak key half the time.”