algorithms for 'invert', was: program works with maxima 5.29.1 but freezes with maxima 5.30.0



On 2013-04-20, Steve Haflich <shaflich at gmail.com> wrote:

> One way to approach issues like this -- in modern times when most machines
> have multiple processor cores, cycles ate cheap, and when there are even a
> few lisps that support multiprocessing -- is to do both. Accept the result
> from whichever completes first, and cancel the other(s).
>
> Unfortunately this could never work for Maxima. Computational side effects,
> isolation of state, and even potential user interaction would all have to
> be carefully isolated, and Maxima would have to be redesigned from scratch.
> That isn't going to happen.

Yes, that's absolutely a reasonable approach. It would be easy to make
it work, at least in Unixland -- fork a process to handle each method.
My guess is that it wouldn't be much work (on the order of days) to get
a proof-of-concept implementation going.

best

Robert Dodier