Re: Thoughts on Shellshock and beyond

[John Haxby <john.haxby () oracle com>]

On 12 Oct 2014, at 12:24, Florian Weimer <fw () deneb enyo de> wrote:

> I don't think Haskell is a magic bullet.  I do think type-rich
> languages (and languages with memory safety) have a lot to offer, but
> writing secure software in them is still hard.

I’d definitely agree with that.

Recently I was dealing with a problem where a developer had gone to a lot of trouble to design and implement an 
insecure authentication mechanism.   He thought he was doing the right thing but he just couldn’t see the flaws in what 
he’d done.  

The problem wasn’t the choice of programming language (python, as it happens) it was simply that getting the design and 
implementation right hard even though it looks easy.   Haskell (or Ada or CLU) would not have helped; a mathematically 
rigorous approach to the problem would have helped a lot, but it would not have made it easy.  To paraphrase Gödel 
somewhat: any non-trivial system has is not provably secure.

jch