BeefGriller is antifa5d ago
Still reading through the #Emacs #Elisp intro. One line reads, "A list of the arguments: in this example, an empty list, ‘()’," However, in my #mathematics-oriented brain, I correct it to "*the* empty list," I guess it's because set theory calls a set with no members *the* empty set. 🤡
Show threadKlara ❤️5d ago

@BeefGriller

Honestly, you're not wrong. All lists end with '() or nil depending on what you want to call it, and it's always the same, so I tend to call it the empty list as well :)

Show threadmaya5d ago
@boo_ @BeefGriller do they all get represented as the nil ptr? i think in at least in scheme there is no separate nil value so it may be actually eq objects (not sure about emocs lisp or common lisp)
Show threadKlara ❤️5d ago

@zenmaya @BeefGriller

Pretty sure it is that way, and I've mostly seen '() referred to as the empty list in Scheme as well. Sure, Scheme doesn't have nil, but I'm pretty sure '() is always the same. The Guile Scheme info page for Equality only lists numbers and characters as exceptions to eq? testing reference equality. I don't know if that's how RNRS defines it, but Guile is the Scheme dialect I'm familiar with.

Show threadBeefGriller is antifa5d ago

@boo_ @zenmaya I'm still a neophyte when it comes to Lisp, whatever the dialect. So I don't know if () is the same as nil, or whatever. The math part of my brain says yes - set theory and all that. Also, according to set theory, the empty set is a member of all sets, but I suspect that probably doesn't translate to Lisp lists. Still learning, though, and enjoying myself, as well as these conversations.

Thanks to both of you for the replies! 😃

Show threadKlara ❤️

@BeefGriller @zenmaya

Oh yeah, it might be a little too much at once to involve a bunch of different lisps  

But yes, nil is the empty list, it's basically because the empty list is also false in Elisp, so the tendency is to use nil when you mean false, and '() when you mean the empty list, but they're the exact same. All lists contain nil at the end, so even if the rest of the list model doesn't exactly match set theory, that part does, and that connection was one of the big things that made me think Lisp was really cool when I started learning it  

Happy hacking !

Mar 27 at 5:50pmWeb
Show threadPhilip4d ago

@boo_ @BeefGriller @zenmaya Disjunct datatypes is something that Scheme and traditional Lisps differ on: nil (or "()" via reader syntax) is a symbol, a boolean and a list, while Scheme #(), #f and 'nil are respectivly not-eq.

But it is funny to see people have the reverse struggle I had when learning mathematics: Unless I ignored it, I was usually confused why mathematicians placed such emphasis on something being "the" X. I had to engage with maths philosophy to understand it better ^^

Show threadFlorian Guthmann4d ago
@pkal @boo_ @BeefGriller @zenmaya That's why one should generally replace "the" with "a selected" and describe things by their universal properties.
Show threadBeefGriller is antifa3d ago
@pkal Interesting how Lisp dialects differ in that regard. I’m curious how and why that came about. But I’m not sure I’d understand that with my current knowledge. And thank you for that information!
Show threadPhilip3d ago
@BeefGriller The first LISPs only had cons-cells and symbols (no numbers, vectors, ...). So to designate empty lists, and falsehood, they had to decide on a specific symbol, which turned out to be the same one for matters of convenience, but not necessity. I am not sure about the early history of Scheme or when they decided to have disjunct datatypes, you'd have to look through the RnRS specs to find that out.