Update on alternative hashing for String keys with hash-based Maps

[Alan Bateman]

On 14/07/2012 12:47, Rémi Forax wrote:
> The other solution is for 7 to use murmur hash instead of classical 
> hash but to always with the same seed, in that case it will not change 
> from run to run.
That would restore predictability but I assume opens up the possibility 
of collisions again (albeit much harder to do).

In any case, I think Doug's work in the latest CHM is very promising, 
lots of advantages.

-Alan.