On 28 Apr 2014, at 8:09 , Víctor López Juan <vic...@lopezjuan.com
> Is this behaviour intentional, or is it a bug?
I guess it's a bug. I've only ever used foldAllP with commutative operators, so the order didn't matter.
> In any case, I wouldn't suggest changing the implementation, as it might
> break existing programs, and the alternative behaviour can be achieved
> by applying flip to the operator.
> However, perhaps it can be documented? I find the current behaviour
> surprising, and it might be so for other users. I'm willing to submit a
> patch to that effect.
I'd be fine with a patch that both flipped the order and documented it. I don't see a reason to preserve broken behaviour between versions -- that's what version numbers are for.