Logs: freenode/#haskell
| 2020-10-09 19:46:56 | × | taurux quits (~taurux@net-188-152-69-171.cust.dsl.teletu.it) (Ping timeout: 272 seconds) |
| 2020-10-09 19:47:31 | <dsal> | Heh. Yeah. I used to write a lot of (small) scheme. It was fine. ocaml was a bigger step for me. But I went from "Haskell is really hard and only for elite snobs" to finding it the easiest language to work in for almost anything I want to do. |
| 2020-10-09 19:47:41 | → | taurux joins (~taurux@net-93-144-148-150.cust.dsl.teletu.it) |
| 2020-10-09 19:48:04 | <dolio> | Maybe the problem is that I was content messing around with homework exercises in Haskell, but it seems like a lot of people only want to 'mess around' if it involves a ton of web junk or something. |
| 2020-10-09 19:48:16 | <dwts> | I wish I'll reach that stage, so far it manages to make me feel like a complete idiot :) |
| 2020-10-09 19:48:35 | <dsal> | Yeah, that's why I like the approach of suspending knowledge and watching it all unfold. |
| 2020-10-09 19:48:53 | <dolio> | And all the web junk in Haskell is fancy. |
| 2020-10-09 19:48:55 | <dwts> | the irc community doesn't seem snob at all btw, lots of interesting conversations in here |
| 2020-10-09 19:48:55 | → | elliott_ joins (~elliott@pool-108-51-141-12.washdc.fios.verizon.net) |
| 2020-10-09 19:48:57 | <dsal> | Much of the really cool magic goes away when you become a magician, but then you can still get more done more eaisly. |
| 2020-10-09 19:49:21 | <dsal> | Haskell folks really like Haskell and want everyone to feel the same joy. |
| 2020-10-09 19:49:44 | <ski> | dwts : yea, often it seems newbies goes way overboard with `$' .. i much prefer seeing the brackets, much of the time. using `.' is fine, though |
| 2020-10-09 19:50:14 | <dsal> | If I have a lot of $ in code, it's because I'm shell scripting incrementally in Haskell. |
| 2020-10-09 19:50:36 | <dwts> | ski: I guess the haskell people tried a lot to avoid the parens. You can clearly see that in the funciton definitions |
| 2020-10-09 19:51:05 | <dminuoso> | Question, why exactly do we have no parens for function application? |
| 2020-10-09 19:51:08 | <monochrom> | What dolio said is true of me too. Also generally not just web junk, but overall any "real" "application". |
| 2020-10-09 19:51:18 | <dsal> | dminuoso: You can if you want, but why do you want? |
| 2020-10-09 19:51:26 | <ski> | dminuoso : curried style ? |
| 2020-10-09 19:51:33 | <ghoulguy> | dminuoso: with functions all taking a single argument it probably gets pretty noisy to do it with them |
| 2020-10-09 19:51:34 | <dwts> | I thought hat you can use parens? |
| 2020-10-09 19:51:37 | <ski> | possibly goes back to ISWIM, at least ? |
| 2020-10-09 19:51:57 | <dwts> | s/hat/that |
| 2020-10-09 19:52:04 | <nshepperd1> | too many parentheses spoil the curry |
| 2020-10-09 19:52:10 | <monochrom> | Me, I believe in "stronger foundation before higher ambition". More need to mess around with "toys" to more deeply understand what's really going on, before it's meaningful to try a useful project. |
| 2020-10-09 19:52:29 | <dolio> | Logicians were already avoiding parentheses when Church was around. |
| 2020-10-09 19:52:45 | <dsal> | > subtract (7) (5) |
| 2020-10-09 19:52:46 | <ski> | yea, combinatory logic |
| 2020-10-09 19:52:47 | <lambdabot> | -2 |
| 2020-10-09 19:52:48 | <dminuoso> | Im just wondering whether someone knows what the reason was |
| 2020-10-09 19:52:58 | <dminuoso> | Id say the immediate reason was that Miranda was that way I guess |
| 2020-10-09 19:53:06 | <ski> | Church did curried stylw |
| 2020-10-09 19:53:27 | <ski> | yea |
| 2020-10-09 19:53:30 | <dminuoso> | But, perhaps the origins of parens-less function application are just lambda calculus? |
| 2020-10-09 19:53:35 | <dwts> | I assume the lambda bot is written in haskell too? |
| 2020-10-09 19:53:39 | <ski> | yes |
| 2020-10-09 19:53:41 | <dwts> | nice |
| 2020-10-09 19:53:47 | <monochrom> | If you point at Miranda then you still have the question "so why did they do it this way?" |
| 2020-10-09 19:53:51 | <dminuoso> | Right. :) |
| 2020-10-09 19:53:52 | × | Sososasa quits (~textual@2601:643:8000:a570:9513:b921:a468:618c) (Quit: Textual IRC Client: www.textualapp.com) |
| 2020-10-09 19:53:56 | <ski> | @version |
| 2020-10-09 19:53:56 | <lambdabot> | lambdabot 5.3.0.1 |
| 2020-10-09 19:53:57 | <lambdabot> | git clone https://github.com/lambdabot/lambdabot |
| 2020-10-09 19:54:02 | <dminuoso> | My instinct says it was from lambda calculus. |
| 2020-10-09 19:54:16 | <monochrom> | But really you look at eg Landin's papers "f(x)(y)(z)(t)" after a while you would invent "f x y z" too. |
| 2020-10-09 19:54:26 | <ski> | monochrom : the "why ?"s never end :) |
| 2020-10-09 19:54:33 | <dsal> | It also kind of overloads the meaning of () |
| 2020-10-09 19:54:42 | <ski> | math, yes |
| 2020-10-09 19:54:47 | × | wz1000 quits (~wz1000@static.11.113.47.78.clients.your-server.de) (Ping timeout: 240 seconds) |
| 2020-10-09 19:54:49 | <ski> | (heavily) |
| 2020-10-09 19:54:54 | <monochrom> | Disgust is the mother of syntax invention. :) |
| 2020-10-09 19:55:01 | <dminuoso> | Well, I was hoping for someone to just have a citation ready "XYZ explained the decision in publication ABC" |
| 2020-10-09 19:55:03 | <dminuoso> | :p |
| 2020-10-09 19:55:12 | <dminuoso> | monochrom: Disgust is sometimes also the child of it. |
| 2020-10-09 19:55:24 | <dwts> | 22:54 < monochrom> Disgust is the mother of syntax invention. :) |
| 2020-10-09 19:55:25 | × | elliott_ quits (~elliott@pool-108-51-141-12.washdc.fios.verizon.net) (Ping timeout: 265 seconds) |
| 2020-10-09 19:55:28 | <dwts> | that quote is gold ^ |
| 2020-10-09 19:55:42 | <ski> | there's `@remember', if you think it's memorable enough |
| 2020-10-09 19:55:53 | <dolio> | Back in the 30s and such they were doing `∃ x . P` to avoid the parentheses in `(∃x)(P)` and such. :) |
| 2020-10-09 19:56:09 | → | isovector1 joins (~isovector@172.103.216.166) |
| 2020-10-09 19:56:13 | <dminuoso> | dolio: hold on, what did the spaces add exactly there? |
| 2020-10-09 19:56:24 | <dminuoso> | `∃x.P` seems clear as well |
| 2020-10-09 19:56:24 | <dolio> | The spaces aren't important. The dot is. |
| 2020-10-09 19:56:27 | <dminuoso> | Ah |
| 2020-10-09 19:56:29 | <ski> | before ⌜(∀x)Px⌝ it was ⌜(x)Px⌝ |
| 2020-10-09 19:56:38 | <dolio> | Dot is kind of like $. Low-precedence. |
| 2020-10-09 19:56:42 | <monochrom> | dolio, then I'm really disappointed that many 1960s logic textbooks still go "(∃x)(P)" |
| 2020-10-09 19:56:44 | <ski> | @help remember |
| 2020-10-09 19:56:45 | <lambdabot> | remember <nick> <quote>: Remember that <nick> said <quote>. |
| 2020-10-09 19:57:03 | <dminuoso> | See, the one thing mathematical publications is missing, is fixity declarations of these notations. |
| 2020-10-09 19:57:05 | <dwts> | ah, I see |
| 2020-10-09 19:57:29 | <ski> | the brackets in ⌜(x)Px⌝ comes from the "rounded dent" in Frege's two-dimensional notation for graphical formulae |
| 2020-10-09 19:57:32 | <dwts> | thanks ski |
| 2020-10-09 19:57:39 | <dolio> | Also you write `P ⊃. Q ⊃ R` to avoid `P ⊃ (Q ⊃ R)` |
| 2020-10-09 19:58:02 | <ski> | yes, there should be a language that picks up the dot notation, instead of brackets |
| 2020-10-09 19:58:03 | <monochrom> | Yikes. |
| 2020-10-09 19:58:04 | <dolio> | Because making it associate one way wasn't a thing yet. |
| 2020-10-09 19:58:08 | → | heatsink joins (~heatsink@107-136-5-69.lightspeed.sntcca.sbcglobal.net) |
| 2020-10-09 19:58:27 | <monochrom> | Well there is now a language that picks up the $ notation. It's called Haskell. :) |
| 2020-10-09 19:58:47 | × | isovector1 quits (~isovector@172.103.216.166) (Client Quit) |
| 2020-10-09 19:59:01 | <ski> | it annoyes me when people write `foo $ bar $ ...' or `blah (f $ ...)' |
| 2020-10-09 19:59:18 | <ski> | usually the code is clearer, reinserting the brackets |
| 2020-10-09 19:59:45 | <dminuoso> | my rule of thumb is, ($) is for non-BlockArguments code and liftIO/lift :p |
| 2020-10-09 20:00:11 | <ski> | yea, `lift $ ...',`liftIO $ ...',`return $ ...',`pure $ ...' can be okay |
| 2020-10-09 20:01:09 | <ski> | (but if `...' is just a simple expression (on a single line), with no brackets, i'd usually replace the `$' by brackets, anyway) |
| 2020-10-09 20:01:43 | <dsal> | $ is for the last thing in a chain of . |
| 2020-10-09 20:01:56 | <Cale> | I'm perfectly happy writing stuff like liftIO . forM blah $ \x -> ... |
| 2020-10-09 20:02:16 | <dolio> | Yeah, putting a $ at the end of a composition is what I usually do. |
| 2020-10-09 20:02:25 | × | heatsink quits (~heatsink@107-136-5-69.lightspeed.sntcca.sbcglobal.net) (Ping timeout: 240 seconds) |
| 2020-10-09 20:04:28 | nshepperd1 | can't stop thinking about that idea of abusing BlockArguments to substitute indentation for parentheses |
| 2020-10-09 20:04:36 | <Cale> | P ⊃: Q ⊃. R ⊃ S for P ⊃ (Q ⊃ (R ⊃ S)) |
| 2020-10-09 20:04:42 | <dolio> | nshepperd1: Yeah, can't say I'm a fan of that. |
| 2020-10-09 20:04:42 | <Cale> | moar dots |
| 2020-10-09 20:05:00 | <Cale> | That'll be a 50 dkp minus for parens |
| 2020-10-09 20:05:04 | × | conal quits (~conal@64.71.133.70) (Quit: Computer has gone to sleep.) |
| 2020-10-09 20:05:07 | <dolio> | Cale: You're not thinking creatively enough. |
| 2020-10-09 20:05:10 | <dminuoso> | nshepperd1: can you elaborate? |
| 2020-10-09 20:05:40 | <ski> | nshepperd1 : yea, i'll need to think about that |
| 2020-10-09 20:05:40 | <dolio> | P ⊃ Q .⊃. P ⊃ R |
| 2020-10-09 20:06:17 | <ski> | that's ⌜(P ⊃ Q) ⊃ (P ⊃ R)⌝ |
| 2020-10-09 20:06:28 | <dolio> | Yeah, obviously. :) |
All times are in UTC.