@Dennis I used this thingy right here: .d('Steve').d(' Ball').s('mer').d(' still').d(' do').s('es').d(' no').s('t').d(' know'). You just have to be careful with the words a bit.
@Dennis 05AB1E has a better dictionary in many cases. The words it has are more often than not pretty much enough, and they're each 2 bytes. So, you should start learning it...you regret that you didn't lowercase those words back then right?
@Dennis surely? I mean, I can do a pull request that lowercases the words while keeping them at the same index...of course then goes the competitiveness
@EriktheOutgolfer Well, if the first character of the literal decides which encoding to use, a trailing delimiter would not be required at the end of the program. That works out very rarely though.
@EriktheOutgolfer Got a lot of ideas floating in my head, but TIO is taking up all my free time at this point.
@EriktheOutgolfer That's unlikely to become a thing. I plan to cast characters to their code point when needed, so 3ẋ”A would repeat 3 65 times.
@EriktheOutgolfer I'm not sure what the point of this would be. Jelly is golfy because it's tacit.
@EriktheOutgolfer That already works.
I am considering altering the behavior of unparseable nilads though. The implicit print is rarely useful, and instead of printing these, these values could be stored on a stack, and pop'd to be used instead of the rightmost argument for coming dyads.
Jelly, 49 bytes
“ṬċḌ)⁹œḃṣ⁷Ṅḋ%W3Œƭ;ḷẓ“¡=~Ṃ©N|ȯ“¡ṭṂ“CṬṀWỌ»ḣJK;¥€”.Y
Try it online!
