[Haskell-cafe] Re: Write Haskell as fast as C. [Was: Re: GHC predictability]

[Andrew Coppin]

Bryan O'Sullivan wrote:
> Andrew Coppin wrote:
>
>   
>> On the other hand, this is the anti-theisis of Haskell. We start with a
>> high-level, declarative program, which performs horribly, and end up
>> with a manually hand-optimised blob that's much harder to read but goes
>> way faster.
>>     
>
> Buh?  This is hard to read?
>   

Look closer: it's hardER to read.

  mean xs = sum xs / fromIntegral (length xs)

  mean = go 0 0 n
    where
      go s l x
        | x > m = s / fromIntegra l
        | otherwise = go (s+x) (l+1) (x+1

One version makes it instantly clear, at a glance, what is happening. 
The other requires you to mentally walk round a look, imperative style, 
to figure out what's happening. It's not a *big* deal, but it's unfortunate.

I'm more worried about what happens in less trivial examples. [Let's 
face it, who wants to compute the sum of the numbers from 1 to N?]