« first day (2802 days earlier)      last day (78 days later) » 

06:58
@Adám I'll take your suggestion and raise you a blog post! blog.wilsonb.com/posts/2024-09-27-brainfuck.html
@Silas Oooo. That's really nice.
@B.Wilson @Silas Should ⍸⍣¯1 (or maybe better {⍵∊⍨⍳⊃⌈/⍵}) be a primitive? Monadic might make sense. A shape left argument, i.e. {⍺←⊃⌈/⍵ ⋄ ⍵∊⍨⍳⍺} might be nice, but ↑∘⍷ isn't exactly rocket science.
@B.Wilson Added to apl.news
07:23
@Adám I'd almost want a primitive ⍣¯1 operator instead, a la Kap, BQN, and Uiua. Though, perhaps it's more just that I don't reach for ⍸⍣¯1 enough. Need to stew on this a bit.
I do like the choice of .
@Adám That was quick. Cheers.
@B.Wilson isn't very often needed. These are the only instances on APLcart:
⊥⍣¯1
+\⍣¯1
≠\⍣¯1
×∘*⍨⍣¯1
⍸⍣¯1
And ⊥⍣¯1 is only needed because scalar⊤ is badly designed.
I might be more interested in ⍣¯1 if we supported more inverses, including some interesting (=arbitrary) ones like ×/⍣¯1 giving prime factors (padded with 1s), ⍴⍣¯1 gave ⊢⍴∘⍳×/ etc.
07:52
Haha. Of course someone coded up Lambert W. I wish there were an I-beam that listed all supported inverses, actually.
Considering how often you see ⊢⍴∘⍳×⌿, even when just playing around with data, that would be nice.
IIRC it is fairly simple to glean which inverses are supported from the source.
Have you looked at my "Inverses" and "Additional inverses" internal wiki pages?
Looking now!
Feel free to amend, of course.
 
2 hours later…
10:02
Inverse execute... That's cute. I guess there's already plenty of relevant code available from the fancier APLAN stuff.

« first day (2802 days earlier)      last day (78 days later) »