genBits: small addition to random API + largely new implementation that needs code review

[Scott Turner]

On 2011-06-28 14:58, Ryan Newton wrote:
> These new changes require a small API change to RandomGen.  The issue is
> that the legacy RandomGen API isn't very good at generating
> random */bits/*.  Rather, it next creates numbers in an arbitrary
> range: genRange.  
...
> I propose that we add a "genBits" function that reports how many bits of
> randomness are created by a generator.
> 
>   class RandomGen g where 
>     ...
>     genBits :: g -> Int
...
> What about generators that have a genRange which is not a power of two?

My own feeling is against this change. That's not based on a lot. I once
coded up a mechanism that could convert a stream of random numbers from
a given sequence of ranges, to another stream of random numbers for
ranges as demanded. No randomness was wasted. Nice, clean fun. I don't
claim it was particularly efficient.

Having worked with that, using bits to organize randomness seems arbitrary.