Logs: freenode/#haskell
| 2021-02-28 09:32:11 | <edwardk> | boxscape: the problem with producing clones of me is that i have to feed them, and we still only have the one tax id, so it raises some serious eyebrows when we're forced to explain things to the IRS. GPT-3 me seems easier on that front, but seems even more dubious from an AI safety and alignment perspective. |
| 2021-02-28 09:32:43 | <boxscape> | those are some good points |
| 2021-02-28 09:34:47 | → | zaquest joins (~notzaques@5.128.210.178) |
| 2021-02-28 09:34:47 | <edwardk> | a friend of mine posted this to facebook https://www.irccloud.com/pastebin/dZzR4f7u/clones.txt |
| 2021-02-28 09:36:06 | <edwardk> | My response was that the number I'd pick would be very large to the point where I'll be calling on government assistance to deal with the refugee problem that I'd be creating, but that I didn't suspect it be quite to the point of overwhelming the western seaboard. |
| 2021-02-28 09:36:26 | <boxscape> | tempted to choose a number high enough that they don't fit into the observable universe to check if I can break the system |
| 2021-02-28 09:37:08 | × | mputz quits (~Thunderbi@dslb-088-064-063-125.088.064.pools.vodafone-ip.de) (Ping timeout: 245 seconds) |
| 2021-02-28 09:37:09 | × | Guest77245 quits (~textual@zrcout.mskcc.org) (Ping timeout: 264 seconds) |
| 2021-02-28 09:37:27 | × | danso quits (~dan@2001:1970:52e7:d000:96b8:6dff:feb3:c009) (Ping timeout: 272 seconds) |
| 2021-02-28 09:37:49 | <edwardk> | from the standpoint of many decision theories that seems pretty suboptimal as you'd be condemning a near total fraction of you to die. |
| 2021-02-28 09:38:08 | <boxscape> | I admit it's more of a "fun" choice than a smart choice |
| 2021-02-28 09:38:51 | × | inerkick quits (~rohit@2402:8100:2828:ff35:c014:d64c:1ae3:c0a0) (Quit: Leaving) |
| 2021-02-28 09:40:48 | <boxscape> | edwardk though I'm still working out on whether or not I should care about versions of myself ceasing to exist |
| 2021-02-28 09:40:57 | <boxscape> | s/on// |
| 2021-02-28 09:41:50 | <boxscape> | s/versions/instantiations |
| 2021-02-28 09:42:15 | → | m0rphism1 joins (~m0rphism@HSI-KBW-085-216-104-059.hsi.kabelbw.de) |
| 2021-02-28 09:42:21 | → | danso joins (~dan@2001:1970:52e7:d000:96b8:6dff:feb3:c009) |
| 2021-02-28 09:42:41 | → | merijn joins (~merijn@83-160-49-249.ip.xs4all.nl) |
| 2021-02-28 09:43:29 | <edwardk> | well, if you want to play a simple first order version of the game, assume you could instantiate another copy of yourself, but one of you gets executed, but you have a 50% chance of "you" being the copy, do you care? delay the execution 10 years so it takes with it 10 years of memories, do you care then? |
| 2021-02-28 09:44:12 | <boxscape> | My current best guess is I wouldn't care in the first case, but would in the second case |
| 2021-02-28 09:44:42 | <edwardk> | caring about this tends to bite folks who are looking for consistent ethical frameworks harder, folks who care about utilitarianism and variants and the like. |
| 2021-02-28 09:45:18 | <boxscape> | I've been meaning to read Bostrom's paper about infinities in ethics because that seems where this tends to break down, but haven't gotten around to it |
| 2021-02-28 09:45:25 | <boxscape> | s/seems/seems to be |
| 2021-02-28 09:45:27 | → | ixlun joins (~user@213.205.241.12) |
| 2021-02-28 09:45:30 | <edwardk> | fair |
| 2021-02-28 09:45:39 | → | catt joins (~r@31.127.31.99) |
| 2021-02-28 09:46:05 | → | Tuplanolla joins (~Tuplanoll@91-159-68-239.elisa-laajakaista.fi) |
| 2021-02-28 09:47:00 | <edwardk> | i don't personally tend to think it takes infinities for that to come up, a "Heat Signature" grade teleporter malfunction seems to hit all the same salient points |
| 2021-02-28 09:47:32 | <boxscape> | hm I'm not sure what that means |
| 2021-02-28 09:47:47 | <edwardk> | (Heat Signature the game has an npc who is a clone of herself created by a teleporter accident, who was contractually/culturally obligated to off either her original self or the cloned copy, but then reneged.) |
| 2021-02-28 09:48:04 | <boxscape> | I see |
| 2021-02-28 09:48:49 | × | mananamenos_ quits (~mananamen@193.red-88-11-66.dynamicip.rima-tde.net) (Ping timeout: 260 seconds) |
| 2021-02-28 09:49:29 | edwardk | should go to bed or write more code. not sure which |
| 2021-02-28 09:49:51 | <ixlun> | Hi all, if I've got `data Foo = Foo {num :: String, val :: Int, map :: Map String Int}' and I want to insert a new value into the map, is there a short hand I can use for constructing a new Foo with the new map returned from Map.insert? |
| 2021-02-28 09:50:19 | → | heatsink joins (~heatsink@2600:1700:bef1:5e10:692f:34e4:c65a:92f2) |
| 2021-02-28 09:50:41 | → | nineonine joins (~nineonine@2604:3d08:7785:9600:8c3e:8d1a:de68:76d3) |
| 2021-02-28 09:51:04 | <edwardk> | ixlun: foo = Foo "hello" mempty; bar = foo { map = Map.insert 1 2 (map foo) } -- is the common way to do this. you can shave a little bit by using RecordWildcards but they aren't popular. lenses is another way to make it more concise |
| 2021-02-28 09:51:36 | <edwardk> | er foo = Foo "hello" 14 mempty -- missed your 'val' member |
| 2021-02-28 09:52:31 | <ixlun> | Ah, didn't know about the `bar = foo {map = ... }' syntax |
| 2021-02-28 09:52:50 | <edwardk> | with lenses you can get that down to something like foo & map.at 1 ?~ 2 |
| 2021-02-28 09:53:22 | <ixlun> | I remember starting to take a look at lenses and I got scared! |
| 2021-02-28 09:53:48 | <edwardk> | there are some pretty gentle introductions, and a bunch of folks here and over in #haskell-lens happy to answer questions about them |
| 2021-02-28 09:54:18 | → | knupfer joins (~Thunderbi@200116b82cc9eb004cd887aa2955feba.dip.versatel-1u1.de) |
| 2021-02-28 09:54:26 | <edwardk> | unlike most haskell i recommend trying to get a feel for them from examples rather than studying the types super hard. then once you have the intuion the types peel themselves apart much easier when you go to try to figure out how they are actually implemented |
| 2021-02-28 09:54:37 | <edwardk> | otherwise you spend a lot of time staring a trees and not seeing the forest |
| 2021-02-28 09:55:09 | → | mananamenos joins (~mananamen@193.red-88-11-66.dynamicip.rima-tde.net) |
| 2021-02-28 09:55:14 | × | heatsink quits (~heatsink@2600:1700:bef1:5e10:692f:34e4:c65a:92f2) (Ping timeout: 264 seconds) |
| 2021-02-28 09:55:37 | → | bergey joins (~user@pool-74-108-99-127.nycmny.fios.verizon.net) |
| 2021-02-28 09:55:41 | <ixlun> | to be honest, that's kind of how I got to grips with applicative syntax. I brushed up on the types but it wasn't until I used parser combinators that I really got it. |
| 2021-02-28 09:56:03 | × | nineonine quits (~nineonine@2604:3d08:7785:9600:8c3e:8d1a:de68:76d3) (Ping timeout: 258 seconds) |
| 2021-02-28 09:56:40 | → | _ht joins (~quassel@82-169-194-8.biz.kpn.net) |
| 2021-02-28 09:57:03 | <ixlun> | Just one question, the brace syntax that you used above, do I need a language extension for that? |
| 2021-02-28 09:57:49 | <tomsmeding> | that syntax is ancient haskell :) |
| 2021-02-28 09:58:52 | <ixlun> | Cool, thanks :) |
| 2021-02-28 09:58:53 | <edwardk> | ixlun: that has been around since dirt |
| 2021-02-28 09:59:20 | <hyiltiz> | cd |
| 2021-02-28 09:59:53 | tomsmeding | thinks somebody went home |
| 2021-02-28 10:00:07 | <edwardk> | there's no place like $HOME |
| 2021-02-28 10:00:16 | × | bergey quits (~user@pool-74-108-99-127.nycmny.fios.verizon.net) (Ping timeout: 240 seconds) |
| 2021-02-28 10:02:45 | → | mayleesia joins (4d0db908@dynamic-077-013-185-008.77.13.pool.telefonica.de) |
| 2021-02-28 10:03:41 | → | gehmehgeh joins (~ircuser1@gateway/tor-sasl/gehmehgeh) |
| 2021-02-28 10:04:52 | <edwardk> | hyiltiz: i could maybe scrape the content code out of comonad.com, but i think john wiegley already has a script for doing that. |
| 2021-02-28 10:05:09 | × | mananamenos quits (~mananamen@193.red-88-11-66.dynamicip.rima-tde.net) (Ping timeout: 260 seconds) |
| 2021-02-28 10:05:40 | <hyiltiz> | yeah i accidentally tried to go home in #haskell |
| 2021-02-28 10:05:48 | <hyiltiz> | but #haskell is home so that is good |
| 2021-02-28 10:06:22 | <hyiltiz> | edwardk: don't wanna bother u with that; if u dont have anything ready to dump at me, I can look up john wiegley's script |
| 2021-02-28 10:07:04 | × | Tuplanolla quits (~Tuplanoll@91-159-68-239.elisa-laajakaista.fi) (Read error: Connection timed out) |
| 2021-02-28 10:07:34 | → | Tuplanolla joins (~Tuplanoll@91-159-68-239.elisa-laajakaista.fi) |
| 2021-02-28 10:07:36 | <edwardk> | ai dungeon's dragon model managed to perform a pretty convincing impression of me a few times |
| 2021-02-28 10:09:07 | × | da39a3ee5e6b4b0d quits (~da39a3ee5@2403:6200:8876:d719:a529:2602:758:74a6) (Quit: Textual IRC Client: www.textualapp.com) |
| 2021-02-28 10:10:03 | ← | m4lvin parts (~m4lvin@w4eg.de) () |
| 2021-02-28 10:12:05 | <edwardk> | https://gist.github.com/ekmett/38513c7894e2d28863f8202acb1a3447 https://gist.github.com/ekmett/1b9ad5c361dfd1f23887ce1ab11a23fe -- it did a better job of staying in genre in the former |
| 2021-02-28 10:12:44 | <kuribas> | ixlun: you can also write `set (map . at 1) 2 foo`, which I find more readable that the operators. |
| 2021-02-28 10:13:56 | <kuribas> | There is a logic to the way lens operators are constructed, but I still find them ugly... |
| 2021-02-28 10:14:11 | × | danso quits (~dan@2001:1970:52e7:d000:96b8:6dff:feb3:c009) (Ping timeout: 272 seconds) |
| 2021-02-28 10:14:36 | <hyiltiz> | those conversations look really fun |
| 2021-02-28 10:14:56 | × | Svet_ quits (~svet@90.200.185.163) (Ping timeout: 240 seconds) |
| 2021-02-28 10:16:04 | <kuribas> | ixlun: I also find the documentation for micro-lens (which is lens-compatible) a bit more beginner-friendly. |
| 2021-02-28 10:16:43 | × | merijn quits (~merijn@83-160-49-249.ip.xs4all.nl) (Ping timeout: 245 seconds) |
| 2021-02-28 10:17:00 | <hyiltiz-M> | Or optics |
| 2021-02-28 10:17:11 | <olligobber> | when I do `cabal repl' it loads one of my 16 module's functions, is there a way of not loading any? |
| 2021-02-28 10:17:35 | × | Sgeo quits (~Sgeo@ool-18b98aa4.dyn.optonline.net) (Read error: Connection reset by peer) |
| 2021-02-28 10:18:25 | <boxscape> | edwardk were the "Edward:" parts in the first one generated as well or only the "Eliezer:" parts? |
| 2021-02-28 10:20:18 | <olligobber> | ah, found this: https://github.com/haskell/cabal/issues/5374 |
| 2021-02-28 10:20:52 | → | danso joins (~dan@2001:1970:52e7:d000:96b8:6dff:feb3:c009) |
| 2021-02-28 10:21:12 | <edwardk> | everything below the -- End Prompt is generated |
| 2021-02-28 10:21:20 | <boxscape> | Huh, in that case I'm really impressed it came up with "I thought you didn't believe in ascribing zero or one hundred percent odds to anything?" and "So how many significant digits are there in that one hundred percent estimate?" |
| 2021-02-28 10:21:31 | <edwardk> | both of them shocked me |
| 2021-02-28 10:21:46 | <edwardk> | it does seem to have read a lot of yudkowsky though |
| 2021-02-28 10:21:55 | <edwardk> | so i'm not surprised it put some of his style of words into my mouth |
| 2021-02-28 10:22:03 | <boxscape> | hm, makes sense |
| 2021-02-28 10:22:54 | <edwardk> | another example of confusing personalities between characters is this one https://gist.github.com/ekmett/6cd785180659e3c2f9284e5128c11025 it makes him out to be much more of a bad-ass than he is in real life ;) |
| 2021-02-28 10:24:46 | <olligobber> | maybe I'll just use stack instead of cabal |
| 2021-02-28 10:25:32 | <edwardk> | olligobber: in theory you can set you a .ghci file, but iirc cabal repl loads the first (or is it the last) module in the library |
| 2021-02-28 10:26:25 | <edwardk> | olligobber: i guess you could add a second sublibrary with no modules and load it instead? |
| 2021-02-28 10:27:28 | <olligobber> | it loads the first |
| 2021-02-28 10:27:34 | <olligobber> | that seems hacky |
| 2021-02-28 10:29:10 | <edwardk> | olligobber: i think the mindset is if you list things in alphabetical order you'll tend to place Foo at the top of the list before Foo.Bar Foo.Baz, etc. |
| 2021-02-28 10:29:22 | <edwardk> | so it'll load a nice top level module |
| 2021-02-28 10:29:36 | <olligobber> | I don't want it to load any module... |
All times are in UTC.