A regression in Maxima 5.17.1 with limits and assumptions?

On Tue, Dec 23, 2008 at 11:35 AM, Robert Dodier <robert.dodier at gmail.com>wrote:

> On 12/23/08, Stavros Macrakis <macrakis at alum.mit.edu> wrote:
> > ...And what about [if x>0 then a else b, asksign(x)]? Should that
> re-simplify the conditional?
>
> Since Maxima is mostly or entirely dominated by dynamic as opposed to
> lexical constructs, it seems inappropriate to go back and reevaluate or
> resimplify previously evaluated results.
>

I don't think this is a dynamic vs. lexical issue -- the two x's are the
same object. And I certainly would not want Maxima throwing in extra
*evaluations* as opposed to simplifications.

I wasn't actually suggesting that we re-simplify in this case.  What I *was*
suggesting is that there are probably order dependencies like this in the
internal code as well as in user code.

          -s