RE: behaviour of System.system?

[Simon Peyton-Jones]

| Obviously it would be useful if the expected behaviour were documented
| more tightly, and conformed across compilers.  Which result should we
| choose to be standard?  (And can the decision please be 
| recorded in the Haskell 98 errata?)

I would welcome a tightened definition, and would add it
to the errata; but I am not qualified to produce one.

One difficulty is that the definition tries not to be Unix-centric.

Simon