Newest at the top
| 2024-11-14 22:37:00 +0100 | <guy_> | seti+geti are abstractions extending around pattern matching on (:) where you get an extra piece of data |
| 2024-11-14 22:36:28 +0100 | <guy_> | its more like basically "because its foldable there is a toList" so its "linearizable" |
| 2024-11-14 22:36:10 +0100 | <guy_> | but yeah, totally different concept of linearity |
| 2024-11-14 22:36:02 +0100 | <guy_> | i think basically, because the state is updated each time, thats where the linearity comes in |
| 2024-11-14 22:35:34 +0100 | <guy_> | and then its weird because your functions are replaced by something that has concrete and variable data associated |
| 2024-11-14 22:35:16 +0100 | <guy_> | (s,s->a->(s,b)) actually, since you need the state aswell |
| 2024-11-14 22:34:29 +0100 | <guy_> | i think the "type theoretic linearity" can be used to great effect here, but i forget how! |
| 2024-11-14 22:34:13 +0100 | <guy_> | since you can always have s~() you can basically make a totally stateful language |
| 2024-11-14 22:33:53 +0100 | <guy_> | but now instead of a bivariate function a -> b, its the stateful function; s -> a -> (s,b) |
| 2024-11-14 22:33:29 +0100 | <guy_> | like how haskell has "all functions are bivariate functions" where partial application can return a new bivariate function |
| 2024-11-14 22:33:09 +0100 | <guy_> | that was something i think we were working on in terms of a "strictly stateful" functional programming language |
| 2024-11-14 22:32:56 +0100 | shapr | (~user@2601:19c:417e:5434:b5b7:a31:f560:51b7) shapr |
| 2024-11-14 22:32:52 +0100 | <guy_> | "type theoretic linearity". no thats about some kind of strict purity, right? like, variables used exactly once and then deleted? |
| 2024-11-14 22:31:53 +0100 | <guy_> | im always looking for keen people that can offer their services. better than being completely invisabailised imo |
| 2024-11-14 22:31:33 +0100 | CrunchyFlakes | (~CrunchyFl@ip1f13e94e.dynamic.kabel-deutschland.de) (Quit: ZNC 1.8.2 - https://znc.in) |
| 2024-11-14 22:31:17 +0100 | <guy_> | i was hopeing the advances in transparency and open society would maybe percolate through |
| 2024-11-14 22:31:02 +0100 | <guy_> | are you sure its not the ol' "not the secret societies responsibility" argument! |
| 2024-11-14 22:30:47 +0100 | <guy_> | too lazy!? |
| 2024-11-14 22:30:12 +0100 | dolio | (~dolio@130.44.140.168) dolio |
| 2024-11-14 22:28:29 +0100 | <haskellbridge> | <Bowuigi> I wanted to do that for my research but I got too lazy lol |
| 2024-11-14 22:28:10 +0100 | <haskellbridge> | <Bowuigi> Oh also make a blog and post the usual introductory stuff there, way more compact and easier to follow |
| 2024-11-14 22:26:55 +0100 | machinedgod | (~machinedg@d108-173-18-100.abhsia.telus.net) machinedgod |
| 2024-11-14 22:26:42 +0100 | manwithluck | (manwithluc@gateway/vpn/protonvpn/manwithluck) manwithluck |
| 2024-11-14 22:26:17 +0100 | manwithluck | (manwithluc@gateway/vpn/protonvpn/manwithluck) (Remote host closed the connection) |
| 2024-11-14 22:22:33 +0100 | <haskellbridge> | <Bowuigi> So linear here is not the type theoretic linearity right? |
| 2024-11-14 22:22:21 +0100 | dolio | (~dolio@130.44.140.168) (Quit: ZNC 1.9.1 - https://znc.in) |
| 2024-11-14 22:21:02 +0100 | machinedgod | (~machinedg@d108-173-18-100.abhsia.telus.net) (Ping timeout: 252 seconds) |
| 2024-11-14 22:20:17 +0100 | lxsameer | (~lxsameer@Serene/lxsameer) (Ping timeout: 252 seconds) |
| 2024-11-14 22:18:22 +0100 | <guy_> | also, anyone interested in the AGI stuff can shoot me a DM aswell |
| 2024-11-14 22:16:48 +0100 | peterbecich | (~Thunderbi@syn-047-229-123-186.res.spectrum.com) (Ping timeout: 246 seconds) |
| 2024-11-14 22:16:08 +0100 | <guy_> | if anyone has any specific questions i can answer then it would avoid the ire of the moderator! |
| 2024-11-14 22:15:43 +0100 | mange | (~user@user/mange) mange |
| 2024-11-14 22:14:16 +0100 | <guy_> | but i dont want to ramble... thats why i made the voice note! |
| 2024-11-14 22:14:02 +0100 | <guy_> | overwrites* (rewite is a protected word to do with dereferencing and the program monad) |
| 2024-11-14 22:13:13 +0100 | <guy_> | you end up getting something like seti+geti+rewrites |
| 2024-11-14 22:12:43 +0100 | <guy_> | this kind of overwrite / reuse of a restricted cache of variable names is basically amounting to some "nonliner" concept |
| 2024-11-14 22:12:42 +0100 | <guy_> | like "if the variable x corresponds to some Int label on the turing tape, what happens when you locally reasign x within a local scope" |
| 2024-11-14 22:11:12 +0100 | alphazone | (~alphazone@2.219.56.221) (Ping timeout: 246 seconds) |
| 2024-11-14 22:11:11 +0100 | <guy_> | (thats a pretty good tldr tbh) |
| 2024-11-14 22:10:29 +0100 | <guy_> | the turing completeness is basically to do with the "linear"ization of the graph, putting it into a turing machine on a 1d (liniarized) turing tape. when you have local scoping, its something like a violation of the 1-1 nature of the traversable laws. i thought it was super interesting! |
| 2024-11-14 22:08:53 +0100 | <guy_> | i didnt realise at the time there was this more complicated idea of the "graphically complete language" being *nonlinear* owing to local scoping considerations |
| 2024-11-14 22:08:22 +0100 | alphazone_ | (~alphazone@2.219.56.221) |
| 2024-11-14 22:08:19 +0100 | <guy_> | i was hoping that the term "graphically complete language" would become widespread so that haskell could be exeplary as such |
| 2024-11-14 22:08:14 +0100 | youthlic | (~Thunderbi@user/youthlic) youthlic |
| 2024-11-14 22:07:49 +0100 | youthlic | (~Thunderbi@user/youthlic) (Remote host closed the connection) |
| 2024-11-14 22:07:49 +0100 | youthlic | (~Thunderbi@user/youthlic) youthlic |
| 2024-11-14 22:07:25 +0100 | <haskellbridge> | <Bowuigi> I think it's fine though, my knowledge of graph theory is pretty much just the definitions of a graph and a DAG |
| 2024-11-14 22:07:23 +0100 | youthlic | (~Thunderbi@user/youthlic) (Remote host closed the connection) |
| 2024-11-14 22:05:47 +0100 | <guy_> | darn. i was hoping it was accessible |
| 2024-11-14 22:05:01 +0100 | <haskellbridge> | <Bowuigi> Definitely interesting but I don't know enough graph theory to understand it lol |