Home freenode/#haskell: Logs Calendar

Logs: freenode/#haskell

←Prev  Next→
Page 1 .. 491 492 493 494 495 496 497 498 499 500 501 .. 5022
— 502,152 events total
2020-10-07 17:01:24 → chenshen joins (~chenshen@2620:10d:c090:400::5:7709)
2020-10-07 17:01:25 <monochrom> This requires a compromise in existing type inference, i.e., breaks backward compatibility. The compromise will appear in 9.0, is already in the release candidate.
2020-10-07 17:01:53 → isovector1 joins (~isovector@172.103.216.166.cable.tpia.cipherkey.com)
2020-10-07 17:01:57 <monochrom> Then the impredicativity itself will be added later.
2020-10-07 17:02:02 → nineonine joins (~nineonine@216.81.48.202)
2020-10-07 17:02:30 <zincy_> monochrom: What do you mean by too easy?
2020-10-07 17:02:53 <monochrom> Having first-class I/O actions.
2020-10-07 17:03:10 ← isovector1 parts (~isovector@172.103.216.166.cable.tpia.cipherkey.com) ()
2020-10-07 17:03:47 <monochrom> If you have first-class procedures then you have first-class I/O actions as a special case.
2020-10-07 17:04:46 <monochrom> The way Haskell ups the game is not merely having first-class I/O actions, that's too easy. The way Haskell ups the game is that even the sequential composition connectives >>= and <*> are first class.
2020-10-07 17:04:58 → geekosaur joins (ac3a8b66@172.58.139.102)
2020-10-07 17:05:30 <monochrom> Dumbing down to mainstream speak, Haskell has first-class and programmable semicolon, other languages don't.
2020-10-07 17:06:58 <zincy_> monochrom: Ah thanks
2020-10-07 17:07:13 <monochrom> For example look at how even Rust has only managed to allow semicolon for two effect types, normal I/O and exception-throwing I/O. You can't make semicolon also work for your own homemade effect type. Well Haskell can.
2020-10-07 17:08:28 hackage silkscreen 0.0.0.2 - Prettyprinting transformers. https://hackage.haskell.org/package/silkscreen-0.0.0.2 (robrix)
2020-10-07 17:08:52 → jneira_ joins (~jneira@80.30.101.206)
2020-10-07 17:09:09 → fendor_ joins (~fendor@t204-126.demo.tuwien.ac.at)
2020-10-07 17:10:38 <zincy_> monochrom: What is a useful definition of the term "computational effect"
2020-10-07 17:10:52 <zincy_> Just an environment modification in a computation?
2020-10-07 17:10:55 → jpcooper joins (~user@unaffiliated/jpcooper)
2020-10-07 17:11:00 → Gurkenglas_ joins (~Gurkengla@unaffiliated/gurkenglas)
2020-10-07 17:11:20 <monochrom> I don't think you should aim for a mathematically checkable definition. Rather, expect a social definition.
2020-10-07 17:11:36 → Saukk joins (~Saukk@2001:998:f1:3963:1c59:9bb5:b94c:2)
2020-10-07 17:11:48 <monochrom> But you can always just say "monad" :)
2020-10-07 17:11:56 × jpcooper quits (~user@unaffiliated/jpcooper) (Client Quit)
2020-10-07 17:12:00 × fendor quits (~fendor@e237-037.eduroam.tuwien.ac.at) (Ping timeout: 272 seconds)
2020-10-07 17:12:14 <zincy_> It is odd that so many things in programming languages rely on social definitions
2020-10-07 17:12:34 <zincy_> so monad <==> computational effect
2020-10-07 17:12:45 <monochrom> Ahem, programming is a social construct. >:)
2020-10-07 17:13:15 × thunderrd quits (~thunderrd@183.182.110.178) (Read error: Connection reset by peer)
2020-10-07 17:13:22 <monochrom> I have settled with http://www.cs.utoronto.ca/~trebla/CSCC24-2020-Summer/06-fam.html#effect
2020-10-07 17:13:27 × jrqc quits (~rofl@96.78.87.197) (Ping timeout: 240 seconds)
2020-10-07 17:13:46 → thunderrd joins (~thunderrd@183.182.110.178)
2020-10-07 17:13:57 <monochrom> e.g., "things a mathematical function cannot do"
2020-10-07 17:14:17 → avdb joins (~avdb@ip-62-235-189-65.dsl.scarlet.be)
2020-10-07 17:14:27 hackage sequence-formats 1.5.1.4 - A package with basic parsing utilities for several Bioinformatic data formats. https://hackage.haskell.org/package/sequence-formats-1.5.1.4 (stephan_schiffels)
2020-10-07 17:14:50 <zincy_> Is Maybe not a deterministic effect?
2020-10-07 17:15:53 <monochrom> It is very much like how "computation" itself doesn't have a mathematical definition either, only a social one. We socially use the Church-Turing thesis, and this means we anticipate that it is possible that one day we may find and change to something even better.
2020-10-07 17:16:06 → shatriff joins (~vitaliish@88.155.140.206)
2020-10-07 17:16:46 <monochrom> But X->Y cannot do what X -> Maybe Y does
2020-10-07 17:17:05 <zincy_> yeah that always bothers me about the fact we cannot agree what computation means
2020-10-07 17:17:11 <monochrom> Declaring failure is considered to be effectful.
2020-10-07 17:17:30 <monochrom> Computing is a social construct. Change my mind. >:)
2020-10-07 17:17:43 <geekosaur> "failure" is a social construct, tbh
2020-10-07 17:17:48 <monochrom> haha
2020-10-07 17:17:51 <zincy_> So deterministic in a narrow sense
2020-10-07 17:18:00 <monochrom> I love this "social construct" meme :)
2020-10-07 17:18:04 <zincy_> As in 2 + 2 never relies on context
2020-10-07 17:18:17 <koz_> Being deterministic and being context-free are not equivalent.
2020-10-07 17:18:22 → thir joins (~thir@p200300f27f0fc60038c1b16891cbfa03.dip0.t-ipconnect.de)
2020-10-07 17:18:23 → jrqc joins (~rofl@96.78.87.197)
2020-10-07 17:18:56 <zincy_> koz_: Right depends on the context
2020-10-07 17:19:17 × emmanuel` quits (~user@2604:2000:1382:ce03:25fa:9e0d:1140:beff) (Remote host closed the connection)
2020-10-07 17:19:25 → emmanuel` joins (~user@2604:2000:1382:ce03:6c74:afa7:7b91:b469)
2020-10-07 17:19:58 <zincy_> monochrom: What is a social construct
2020-10-07 17:20:24 ← emmanuel` parts (~user@2604:2000:1382:ce03:6c74:afa7:7b91:b469) ()
2020-10-07 17:20:36 <monochrom> "artificial" on steroid
2020-10-07 17:20:42 <zincy_> What about biological computation at the level of proteins
2020-10-07 17:20:47 → emmanuel_erc joins (~user@2604:2000:1382:ce03:6c74:afa7:7b91:b469)
2020-10-07 17:20:56 <zincy_> Or should i say "computation"
2020-10-07 17:21:31 <zincy_> If humans weren't around these cellular computations would still occur in other animals
2020-10-07 17:21:46 <zincy_> This begs the question
2020-10-07 17:21:52 → wroathe_ joins (~wroathe@c-68-54-25-135.hsd1.mn.comcast.net)
2020-10-07 17:22:23 × livvy quits (~livvy@gateway/tor-sasl/livvy) (Ping timeout: 240 seconds)
2020-10-07 17:22:24 <geekosaur> anything not captured by mathematical laws is a construct over them. social construct is because we define what programming means, so it's our construct. yours is a biological construct
2020-10-07 17:22:46 × wroathe quits (~wroathe@c-73-24-27-54.hsd1.mn.comcast.net) (Ping timeout: 272 seconds)
2020-10-07 17:22:47 × thir quits (~thir@p200300f27f0fc60038c1b16891cbfa03.dip0.t-ipconnect.de) (Ping timeout: 240 seconds)
2020-10-07 17:22:50 → howdoi joins (uid224@gateway/web/irccloud.com/x-sgkwzvljtciaifbw)
2020-10-07 17:22:58 hackage Shpadoinkle-lens 0.0.0.1 - https://hackage.haskell.org/package/Shpadoinkle-lens-0.0.0.1 (fresheyeball)
2020-10-07 17:23:17 <zincy_> Right but biological construct isn't in the same class as social constructs
2020-10-07 17:23:55 <geekosaur> that depends on your definitions :)
2020-10-07 17:23:55 <zincy_> If you take a photo of a paint splatter and it happens to be a valid perl program, is it a program?
2020-10-07 17:24:19 <zincy_> I would say no according to any conceptually useful definition of program
2020-10-07 17:24:20 <monochrom> I think I don't care.
2020-10-07 17:24:33 × alp quits (~alp@2a01:e0a:58b:4920:88ef:76a6:5d73:1be6) (Ping timeout: 272 seconds)
2020-10-07 17:24:37 <zincy_> How can you not care about definitions
2020-10-07 17:24:42 <geekosaur> it is a program. it may not be a _useful_ program, but that's where the "social" part comes in
2020-10-07 17:25:03 <zincy_> If someone asked you to do a job as a programmer and then handed you a paintbrush would you care then?
2020-10-07 17:25:15 <monochrom> I care about definitions. It's why I use different definitions for different purposes.
2020-10-07 17:25:47 × macrover quits (~macrover@ip70-189-231-35.lv.lv.cox.net) (Remote host closed the connection)
2020-10-07 17:25:48 <zincy_> Does that work
2020-10-07 17:26:34 × taurux quits (~taurux@net-130-25-101-135.cust.vodafonedsl.it) (Ping timeout: 272 seconds)
2020-10-07 17:26:55 → alp joins (~alp@2a01:e0a:58b:4920:78a9:bf4d:b06b:60d7)
2020-10-07 17:27:14 → taurux joins (~taurux@net-130-25-101-135.cust.vodafonedsl.it)
2020-10-07 17:27:16 × brandly quits (~brandly@c-73-68-15-46.hsd1.ma.comcast.net) (Ping timeout: 246 seconds)
2020-10-07 17:27:27 hackage Shpadoinkle-lens 0.0.0.2 - Lens combinators for Shpadoinkle applications. https://hackage.haskell.org/package/Shpadoinkle-lens-0.0.0.2 (fresheyeball)
2020-10-07 17:28:07 <monochrom> A student handing in a paint splatter that happens to be a valid perl program that fulfills my assignment requirements, I'll accept as a submission that fulfills my assignment requirements.
2020-10-07 17:28:28 × stree quits (~stree@50-108-75-26.adr01.mskg.mi.frontiernet.net) (Quit: Caught exception)
2020-10-07 17:28:43 → mav1 joins (~mav@i5E86B7C0.versanet.de)
2020-10-07 17:28:45 → stree joins (~stree@50-108-75-26.adr01.mskg.mi.frontiernet.net)
2020-10-07 17:28:47 <monochrom> This is independent of that I won't take a programmer job that gives me a paintbrush, not a keyboard and monitor, as the tool they make me use for the job.
2020-10-07 17:28:58 → abhixec joins (~abhixec@c-67-169-141-95.hsd1.ca.comcast.net)
2020-10-07 17:29:30 <monochrom> And that is yet independent of that I'm fine if some other programmer takes that job.
2020-10-07 17:29:56 <zincy_> this is far too nuanced
2020-10-07 17:30:05 <geekosaur> why?
2020-10-07 17:30:16 <monochrom> Compared to what?
2020-10-07 17:30:18 <zincy_> sarcasm
2020-10-07 17:30:37 → Jesin joins (~Jesin@pool-72-66-101-18.washdc.fios.verizon.net)
2020-10-07 17:30:52 <zincy_> I guess definitions should be useful and usefulness changes depending on circumstance
2020-10-07 17:31:08 <monochrom> It has always been that way.
←Prev  Next→
Page 1 .. 491 492 493 494 495 496 497 498 499 500 501 .. 5022

All times are in UTC.