oh yeah, another thing that's actually sort of been driving my thinking about vectorization for Perhaps is ibeing able to intelligently vectorize over its right argument, which would require that it's some way or another able to tell when it wouldn't make sense to be looking for a list in a list versus several scalars in a list without compromising any of the cases where it would

@UnrelatedString Usually the problem I have with ⁺ is that when it "duplicates" the previous link, I think it just inserts the link into the program, ignoring precedence. For example Z⁺5¡ should be ZZ$5¡, but instead becomes ZZ5¡, which is different

@UnrelatedString iⱮ is the third most used digraph in Jelly answer, it would be a very useful byte reduction

@cairdcoinheringaahing wow, now I'm imagining how bad it was back when Ɱ was Ѐ

and on the topic of that list, $€ would be a sensible thing to abbreviate as well

ị“ would not necessarily feel right but it would be practical

if breaking hypothetical symmetry between the parentheses isn't an issue ( could be used, or the code page could just be adjusted to replace one of the unused symbols with something more fitting

@UnrelatedString Bit out there, but “...« is unimplemented. What if it were equal to “...”i@?

Yeah, I was about to suggest that

But I feel like there's also a lot of cases where you're indexing into a compressed string, or a list of strings, or a list of numbers

@UnrelatedString $€ -> ( would fit with the symmetry of µ€ -> )

true

$€ is really just a smaller scale )

...a bit further down that list, turning €Ẏ into a single concat-map quick would also make sense, and could even net two bytes of savings in the case where it's mapping a dyad within a monadic chain

speaking of quicks taking dyads

@cairdcoinheringaahing yeah one useful change might be to literally just remove line 1278 of interpreter.py if the plan is to fork rather than reimplement

I think that's intended behaviour

It fits with the behaviour of ¡ to update the left argument with the new result

...¡ does that too?

Yep

It's how 0+¡1 is n'th Fib IIRC

ahh

so that's the motivation lol

definitely makes more sense now

Try it online!

It's basically equivalent to rarg = ret, because larg == ret

yeah