On Mon, Oct 11, 2010 at 09:41, Brandon Moore wrote:

While I can see your point about potentially introducing new security holes, and producing much less trusted code, I feel having tidy, pure libraries that we can all integrate into our Haskell is a benefit that far outweighs this.  Especially when we have nice things like the type system, which can be used to alleviate many of the security worries.

I agree in general, for code like servers and file formats, but I worry in particular about cryptographic primitives. Some side channel attacks seem to call for a very low-level language, to make it easier to verify that e.g. execution time and the memory access pattern does not depend on the key.

I personally think we have to draw the line somewhere regarding what we care about when it comes to security. (Provable) correctness, maintainability through well-structured code are things we are more likely to gain through using high-level languages like Haskell. That is actually a lot of security bundled up in those things. What we lose is low-level control, which would be required to thwart side-channel attacks. On the other hand, I'm not convinced openssl or gnutls deal with side-channel attacks very effectively either. In any case, there is nothing that says we must have only *one* SSL library, based on this discussion there seems to be people in the community who still would prefer a binding to openssl/gnutls. /M -- Magnus Therning                        (OpenPGP: 0xAB4DFBA4) magnus＠therning．org          Jabber: magnus＠therning．org http://therning.org/magnus         identi.ca|twitter: magthe