[Haskell-cafe] Thoughts about redesigning "Num" type class

[Jerzy Karczmarczuk]

Dominique Devriese comments my remark :
> Could you mention, please, a system whose measures are for sure positive
> reals, infinity included?
here:
> I think he's talking about the mathematical notion of measure:
> https://en.wikipedia.org/wiki/Measure_(mathematics)
Gosh, I should have read Olaf Klinke somewhat slower...
But I was a physicist too long. Sorry!

On the other hand, I would like to see where, in what kind of 
computations, the Rig type class would be useful (and how to squeeze 
infinity into it). I spent some time doing some computations using 
probabilistic measures and very standard programming standards, and I am 
conservatist. Of course, as Macchiavelli said in "Prince" (more or 
less):  "the innovator makes enemies of all those who have done well 
under the old conditions".
I am doubtful whether I "did well", but I wait for the innovators to 
show me their "new conditions". A good example, where a new Prelude with 
Rigs, Rngs, etc., permit to solve better some concrete, useful 
computational problems.

Thanks in advance.

Jerzy Karczmarczuk