2024/05/26

Newest at the top

2024-05-26 13:13:56 +0200BigKozlowski(~BigKozlow@194.5.60.133)
2024-05-26 13:09:56 +0200BigKozlowski(~BigKozlow@194.5.60.133) (Ping timeout: 252 seconds)
2024-05-26 13:09:12 +0200euphores(~SASL_euph@user/euphores) (Quit: Leaving.)
2024-05-26 13:09:09 +0200 <ski> (but it's not obvious to me if he wants something akin to that, or perhaps something else entirely)
2024-05-26 13:08:57 +0200 <tomsmeding> it's odd to say that a language did something wrong without suggesting any alternative, so I'm going to assume that he did intend an alternative :p
2024-05-26 13:08:44 +0200 <ski> some kind of take on (4) is what my "reflective syntax" is intended to cover
2024-05-26 13:07:56 +0200 <tomsmeding> ski: > The key reference is "Notions of computation and monads". The key idea is to give the semantics of a term (in an ML-like language) as a morphism in the Kleisli category of a monad. That's exactly what you do when you write a program in "monadic style". You keep a dog and bark yourself!
2024-05-26 13:07:32 +0200 <ncf> i think it's clear what is meant but unclear how to solve it
2024-05-26 13:07:28 +0200 <ski> it's not obvious to me that that's what he's suggesting
2024-05-26 13:07:11 +0200 <tomsmeding> ski: automatic idiom brackets around everything?
2024-05-26 13:07:07 +0200 <sprout> ask
2024-05-26 13:06:55 +0200 <ski> apart from (4), the criticism sounds reasonable. (4) is unclear what is meant, to me
2024-05-26 13:05:43 +0200 <sprout> weird, I cannot seem to follow him on mastodon
2024-05-26 13:05:16 +0200 <ski> (checking the source seems to work)
2024-05-26 13:04:41 +0200 <tomsmeding> https://tomsmeding.com/ss/get/tomsmeding/f97paG
2024-05-26 13:04:37 +0200 <ncf> maybe you need more javascript
2024-05-26 13:04:32 +0200 <int-e> oh. odd.
2024-05-26 13:04:25 +0200 <ski> int-e : it's enabled
2024-05-26 13:04:18 +0200 <ski> what ncf said
2024-05-26 13:04:13 +0200 <int-e> ski: Mastodon needs Javascript unfortunately.
2024-05-26 13:04:04 +0200 <ncf> int-e: probably the fact that instances have to be unique, and you "select" which instance to use by wrapping things in newtypes
2024-05-26 13:03:44 +0200 <safinaskar> ncf: toot i'm talking about was made earlier that this April 2024. thanks anyway
2024-05-26 13:03:33 +0200 <ski> (hm, that page is blank, for me)
2024-05-26 13:02:25 +0200 <int-e> What does he mean by "imagine that carrier determines structure"?
2024-05-26 13:02:23 +0200 <ski> i do agree that some notion of automatic Moggi could be useful, though
2024-05-26 13:01:52 +0200 <sprout> maybe they are, but that's up the eye of the beholder since most haskell programmers seem to do fine
2024-05-26 13:01:47 +0200 <tomsmeding> the toot is also a bit "from the category theory bubble", I'd say :p
2024-05-26 13:01:36 +0200mrmr1553343(~mrmr@user/mrmr) (Quit: Bye, See ya later!)
2024-05-26 13:01:32 +0200 <ski> (specifically "make side effects as painful as possible". not counting `unsafePerformIO' and similar, there are no side-effects, so therefore it both makes side-effects as easy as possible, and as hard as possible. (vacuously))
2024-05-26 13:01:28 +0200 <sprout> it's more a sentiment than anything else. haskell wasn't engineered to make side effects painful
2024-05-26 13:00:52 +0200wootehfoot(~wootehfoo@user/wootehfoot)
2024-05-26 13:00:15 +0200 <ncf> https://types.pl/@pigworker/112214856633998920
2024-05-26 13:00:06 +0200gmg(~user@user/gehmehgeh)
2024-05-26 13:00:01 +0200 <ski> that does sentiment does sound a bit confused
2024-05-26 12:59:45 +0200BigKozlowski(~BigKozlow@194.5.60.133)
2024-05-26 12:59:16 +0200 <ncf> sounds like something conor would say
2024-05-26 12:58:48 +0200 <tomsmeding> the retrofitting of that idea is idiom brackets, I think
2024-05-26 12:57:16 +0200 <sprout> yah, sounds important
2024-05-26 12:57:02 +0200 <safinaskar> i am unable to find it anymore
2024-05-26 12:56:56 +0200 <safinaskar> original toot
2024-05-26 12:56:53 +0200 <safinaskar> please, help me find this original too
2024-05-26 12:56:43 +0200 <safinaskar> i once found a toot in some mastodon server (maybe types.pl), which said something like this: "in haskell programmers are punished for using side effects. haskell was intentionally made to make side effects as painful as possible. so haskell programmers learnt that they should use side effects as little as possible. but such design is wrong. in fact, proper programming language should implement automatic Moggi transformation"
2024-05-26 12:45:35 +0200BigKozlowski(~BigKozlow@194.5.60.133) (Ping timeout: 264 seconds)
2024-05-26 12:40:51 +0200BigKozlowski(~BigKozlow@194.5.60.133)
2024-05-26 12:39:04 +0200destituion(~destituio@2a02:2121:6c3:b5b4:4960:5aad:77da:7e14)
2024-05-26 12:38:08 +0200 <tomsmeding> heh
2024-05-26 12:37:39 +0200 <ncf> https://commons.wikimedia.org/w/index.php?title=File_talk%3AMultivalued_functions_with_List_monad.…
2024-05-26 12:37:05 +0200L29Ah(~L29Ah@wikipedia/L29Ah) ()
2024-05-26 12:36:07 +0200BigKozlowski(~BigKozlow@194.5.60.133) (Ping timeout: 272 seconds)
2024-05-26 12:35:41 +0200 <tomsmeding> where they go from `(cbrt [8, -8])` to `(map cbrt) [8, -8]`