On Sun, Oct 11, 2009 at 08:17:48PM -0400, Reid Barton wrote:
> It seems that the definition of split in System.Random is not really
> satisfactory.  

For the curious, the reason for the asymmetry between fst . split and
snd . split is that the RNG states produced by mkStdGen have varying
first component but second component (virtually) always equal to 1.
For full details see http://hackage.haskell.org/trac/ghc/ticket/3575

Regards,
Reid Barton
_______________________________________________
Haskell-Cafe mailing list
[email protected]
http://www.haskell.org/mailman/listinfo/haskell-cafe

Reply via email to