J already shares some of mathematica's strengths (interactivity, integrated
visualisation, data analysis). And I think a high quality symbolic
manipulator is well within reach. But
https://reference.wolfram.com/language/tutorial/SomeNotesOnInternalImplementation.html
gives some idea of its sheer scale; I don't think approaching it is at all
viable.
(That said, it is probably possible to do better in some respects.
Mathematica tends to be unsoundly optimistic and assume that edge cases, such
as division by zero, will not occur.)
On Fri, 30 Sep 2022, Henry Rich wrote:
The implementation of D. was buggy. I hope we can do better with the
addon. I thought there was a rudimantary D.-replacement in the addon
but my memory may be faulty.
With enough Jers contributing maybe this could be a competitor to
Mathematica. Time will tell.
Henry Rich
On 9/30/2022 12:45 AM, Raul Miller wrote:
Yes, rank 0 a valid assumption. Sort of:
J used to have two derivative words, d. (rank 0) and D. (jacobian).
As near as I can tell, the library only replaces d. and does not yet
provide a D. implementation.
That said... I would not think of this as a mathematica competitor --
more like an early version of macsyma.
----------------------------------------------------------------------
For information about J forums see http://www.jsoftware.com/forums.htm
----------------------------------------------------------------------
For information about J forums see http://www.jsoftware.com/forums.htm
