> You can prove programs are correct. The only possible flaw they can have is the specification is wrong.

So specifications are kind of like programs?

Have you heard of logical positivism?

> That's very much not what I said. There may be many reasons.

My point was that it always seems to be some external factor. That strikes me as being very convenient.

> Assuming some conclusion without actual research is braindead.

I didn't think I assumed anything. Like anybody else, I have many things that I need to assess in my day to day life, and often deal with considerable uncertainty.