On 8/5/08, Harald Geyer <harald at lefant.net> wrote:
> While reviewing maxima-Bugs-2037993 (and the already quite dated
> maxima-Bugs-1251540) and reading solve.lisp I realized that there
> is a global flag solve_inconsistent_error which is true by default
> (and there even existes documentation about it).
>
> I think solve_inconsistent_error should be false by default because,
> - this is the behaviour I'd expect from a mathematical POV
> - the name suggests that it affects solve, but it seems it only
> affects linsolve but not algsys. - The documentation isn't really
> clear about this.
> - As users are used to the behaviour of algsys (returning empty
> solutions) they are confused if linsolve behaves differently.
>
> Are there any objections to this?
My recommendation is to kill off solve_inconsistent_error entirely
and change any code it affects to act as though it were false.
If you are interested, please post a patch to do that, and then
if nobody objects, please commit the patch.
best
Robert Dodier