Re: "Perfect" or "Provable" security both crypto and non-crypto?

From: Andrew Swallow (am.swallow_at_eatspam.btinternet.com)
Date: 09/13/04 

Date: Mon, 13 Sep 2004 12:09:32 +0000 (UTC)

"Rob Warnock" <rpw3@rpw3.org> wrote in message
news:ubGdnUdf_6gE7NjcRVn-hQ@speakeasy.net...
[snip]
>
> The problem is that lots of people have known for decades *how*
> to construct programs from their proofs [or, as Dijkstra & Gries
> (and Gwyn, as above) suggest, construct the proof and the program
> simultaneously], but very few companies (or individual developers)
> are willing to pay the *cost* of such as style of coding, which
> does cost several times what "just hacking away until it mostly
> works" costs.

Before program proofs go far the proof system needs to:
Only use ASCII characters. If more operators are needed use words.
Replace predicate calculus with an algebraic logic - reverse polish
calculators did not sell well either.
Arrange a system for automatic checking of the logic.

Andrew Swallow

Loading