Robert Dodier

A slightly simpler way to write the original example. (%i10) simp: false; (%o10) false (%i11) 1 + 1 - 1*h; (%o11) 1 + 1 + (- 1) h It's interesting to me that was parsed essentially as 1 plus 1 plus ((minus 1) times h): (%i12) :lisp $_ ((MPLUS) 1 1 ((MTIMES) ((MMINUS) 1) $H)) I wonder why it wasn't parsed as 1 plus 1 plus (minus (1 times h)). I suppose the two forms are equivalent, but a different input does produce what I expected: (%i13) 1 - 1*h; (%o13) 1 - 1 h (%i14) :lisp $_ ((MPLUS) 1 ((MMINUS)...

For the record, I also think the DWIM inferred subscript stuff is terrible. That said, I think (1) is better than what we have now (and also preferable to (2)).

About declaring . and sconcat as n-ary, I seem to recall that Maxima's notion of n-ary-ness is equivalent to being associative in addition to being defined for arbitrary lists of arguments (maybe this coincides with a conventional understanding of the term, I dunno). I guess that implies such functions are defined for the no-arguments case, although I don't know if all n-ary functions behave that way in Maxima. Sounds like . (with suitable flags) and sconcat have those properties; given that, it...

Pretty-printer slightly incorrect output for 'product(...)

Fixed by commit [ b83d81f ].

Agreed that user-defined and built-in operators should act the same. I see that lreduce, rreduce, and tree_reduce also take an initial value as an argument. Maybe while we're at it, we can ensure that those have the same behavior as xreduce.

@jgmbenoit I notice that Debian bugreport #1131495 is marked "closed". From your point of view, is there any further action needed on the part of the Maxima project?

I agree the existing built-in Boolean operators are more than a bit of a mess, although given that they are widely used and therefore baked into a lot of existing code, we would have to talk about the consequences of changing them. I am going to guess that the crucial aspect, from a user's point of view, is that and and or are noncommutative and short-circuiting. (I am also guessing that simplifying versus evaluating is less of an issue.) I will assume for now that we can't change the existing noncommutative,...