Handling branch cuts for hypergeometric functions



> Actually, upon reading the thread I have the impression that
> most people don't mind sensible interactive queries.

What I personally would like to see is "standardized" behavior (i.e., all
queries behaving the same way, perhaps routed through the same query
procedure) and a means to override the behavior in user programs. (In other
words, I don't necessarily mind interactive queries: I do mind it that
there's no easy way to prevent them from popping up when it's
inappropriate.)

I haven't volunteered to do anything on this because it did not appear that
there's a consensus.


Viktor