Subject: Handling branch cuts for hypergeometric functions
From: Wolfgang Jenkner
Date: Thu, 24 Feb 2005 04:32:59 +0100
Raymond Toy writes:
> Raymond> What to do?
>
> Nothing.
>
> While everyone seemed to agree we should do something and more
> interactive queries was not really desired, no one volunteered to
> actually make interactive queries less interactive, so I'm not going
> to do anything about the branch cuts.
Actually, upon reading the thread I have the impression that most
people don't mind sensible interactive queries.
Personally, I find that interactive querying (and also Albert Reiner's
idea of providing `typical values' for such queries) is a good thing,
because doing so is both mathematically correct and in accordance with
heuristical practice.
IMHO, helping Maxima to pick the correct branch (or any other case
where mathematical correctness is at stake) is so important that it
should be done within the current design, which can be adjusted at
some later point.
Wolfgang (de La Palisse)