[Haskell-cafe] Re: A suggestion for the next high profile Haskell project

[ls-haskell-developer-2006 at m-e-leypold.de]

"Sebastian Sylvan" <sebastian.sylvan at gmail.com> writes:

> On 12/18/06, Bulat Ziganshin <bulat.ziganshin at gmail.com> wrote:
>> Hello Simon,
>>
>> Monday, December 18, 2006, 12:08:49 PM, you wrote:
>>
>> > My view is that Haskell's performance is very seldom the limiting factor
>>
>> of course. when someone said about GPG i just mentioned that this project
>> may be hihgly speed-dependent and this case Haskell is definitely not the
>> solution
>
> I highly disagree. Why would you want to write 99% of your code in a
> tedious and error-prone way to get speed in 1% of your code? Isn't it
> better to write that 1% of the code in a slightly more tedious way
> (Haskell imitating C), or even call a foreign function for it?

I completely agree ...

<...>

> I happen to think Haskell would be very well suited for this since the
> current project seems to be plagued by problems typical for C, and
> higly atypical for Haskell (book-keeping bugs, sometimes also leading
> to security issues).

... but I wonder: GPG, AFAIK undertakes some special measures to
ensure that neither clear text nor private keys are paged out to the
disk (since it might be recovered from there by "the enemy"). How
would you lock data in memory in Haskell? Would that be possible? 

It seems to me that all participants in this thread have missed this
point so far.

Regards -- Markus