[Haskell-cafe] Re[2]: strict Haskell dialect

[Brian Hulley]

Tomasz Zielonka wrote:
> On Sun, Feb 05, 2006 at 05:18:55PM -0000, Brian Hulley wrote:
>> I must admit I'm a bit confused as to why the strictness annotations
>> in Haskell (and Clean) are only allowed in data declarations and not
>> function declarations
>
> Clean does allow strictness annotations in function types.

Thanks for pointing this out - I must admit I had only taken a very quick 
look at Clean (I was overwhelmed by the complicated type system) but now 
I've found the place in the Clean book that describes strictness annotations 
for function types so I must look into this a bit more.

If I wanted to write a 3d computer game in Haskell (or Clean), would lazy 
evaluation with strictness annotations lead to as fast a program as eager 
evaluation with lazy annotations for the same amount of programming effort? 
And would the result be as fast as an equivalent program in C++ or OCaml or 
MLton? If so, there would obviously be no point wasting time trying to 
develop an eager dialect of Haskell (or Clean).

I wonder if current compilation technology for lazy Haskell (or Clean) has 
reached the theoretical limits on what is possible for the compiler to 
optimize away, or if it is just that optimization has not received so much 
attention as work on the type system etc?

Regards, Brian.