Newest at the top
| 2026-01-26 17:39:54 +0100 | merijn | (~merijn@77.242.116.146) (Ping timeout: 260 seconds) |
| 2026-01-26 17:36:49 +0100 | comerijn | (~merijn@77.242.116.146) merijn |
| 2026-01-26 17:34:18 +0100 | shr\ke | (~shrike@user/shrke:31298) (Ping timeout: 256 seconds) |
| 2026-01-26 17:29:39 +0100 | lucabtz | (~lucabtz@user/lucabtz) (Quit: Lost terminal) |
| 2026-01-26 17:05:09 +0100 | GdeVolpiano | (~GdeVolpia@user/GdeVolpiano) GdeVolpiano |
| 2026-01-26 17:02:38 +0100 | GdeVolpiano | (~GdeVolpia@user/GdeVolpiano) (Quit: WeeChat 4.7.2) |
| 2026-01-26 16:53:34 +0100 | chele | (~chele@user/chele) chele |
| 2026-01-26 16:53:21 +0100 | chele | (~chele@user/chele) (Remote host closed the connection) |
| 2026-01-26 16:41:27 +0100 | <haskellbridge> | <Man of Letters (Mikolaj)> (and is not limited to type "T", but works for any type operator "k"; here it uses T and type pairs and it still fails even then) |
| 2026-01-26 16:40:25 +0100 | qqq | (~qqq@185.54.21.105) (Quit: Lost terminal) |
| 2026-01-26 16:38:33 +0100 | <haskellbridge> | <Man of Letters (Mikolaj)> that could probably be worked around by inserting something like "a_b ~ '(Fst a_b, Snd a_b)" into "h", but the point is, in general "h" comes from an external library and is not limited to type pairs and should not be changed for each type of arguments the users of the library may want to use it for |
| 2026-01-26 16:37:50 +0100 | YoungFrog | (~youngfrog@2a02:a03f:ca07:f900:b685:3825:8bb0:ad7d) youngfrog |
| 2026-01-26 16:37:30 +0100 | YoungFrog | (~youngfrog@2a02:a03f:ca07:f900:e244:f9d3:3752:93e5) (Quit: ZNC 1.7.x-git-3-96481995 - https://znc.in) |
| 2026-01-26 16:36:36 +0100 | <haskellbridge> | ... long message truncated: https://kf8nh.com/_heisenbridge/media/kf8nh.com/xIMBotwOvnctkswZfetBzvsv/igMLrx_ztsQ (14 lines) |
| 2026-01-26 16:36:34 +0100 | <haskellbridge> | <Man of Letters (Mikolaj)> unfortunately, this failed in the wild at use of higher-order functions with quantified arguments, similarly as in |
| 2026-01-26 16:36:18 +0100 | karenw | (~karenw@user/karenw) (Ping timeout: 260 seconds) |
| 2026-01-26 16:32:42 +0100 | pavonia | (~user@user/siracusa) (Quit: Bye!) |
| 2026-01-26 16:30:37 +0100 | trickard | (~trickard@cpe-80-98-47-163.wireline.com.au) |
| 2026-01-26 16:30:25 +0100 | trickard | (~trickard@cpe-80-98-47-163.wireline.com.au) (Read error: Connection reset by peer) |
| 2026-01-26 16:25:45 +0100 | merijn | (~merijn@77.242.116.146) merijn |
| 2026-01-26 16:23:39 +0100 | trickard_ | trickard |
| 2026-01-26 16:22:03 +0100 | merijn | (~merijn@77.242.116.146) (Quit: leaving) |
| 2026-01-26 16:20:19 +0100 | ec | (~ec@gateway/tor-sasl/ec) ec |
| 2026-01-26 16:19:55 +0100 | ec | (~ec@gateway/tor-sasl/ec) (Remote host closed the connection) |
| 2026-01-26 16:11:41 +0100 | Pozyomka | (~pyon@user/pyon) pyon |
| 2026-01-26 16:09:59 +0100 | humasect | (~humasect@dyn-192-249-132-90.nexicom.net) humasect |
| 2026-01-26 16:07:36 +0100 | trickard_ | (~trickard@cpe-80-98-47-163.wireline.com.au) |
| 2026-01-26 16:07:22 +0100 | trickard_ | (~trickard@cpe-80-98-47-163.wireline.com.au) (Read error: Connection reset by peer) |
| 2026-01-26 16:07:10 +0100 | merijn | (~merijn@77.242.116.146) merijn |
| 2026-01-26 16:07:05 +0100 | fp | (~Thunderbi@2001-14ba-6e24-3000--198.rev.dnainternet.fi) (Quit: fp) |
| 2026-01-26 16:04:05 +0100 | Pozyomka | (~pyon@user/pyon) (Quit: WeeChat 4.8.1) |
| 2026-01-26 15:54:00 +0100 | comerijn | (~merijn@77.242.116.146) (Ping timeout: 256 seconds) |
| 2026-01-26 15:52:42 +0100 | img | (~img@user/img) img |
| 2026-01-26 15:51:27 +0100 | img | (~img@user/img) (Quit: ZNC 1.10.1 - https://znc.in) |
| 2026-01-26 15:33:40 +0100 | <mjacob> | EvanR, monochrom: With “cycle” in FRP, I meant that, if the program is converted to a dataflow graph where the nodes represent computations and edges represent uses, a cycle in that graph. |
| 2026-01-26 15:32:11 +0100 | ttybitnik | (~ttybitnik@user/wolper) (Quit: Fading out...) |
| 2026-01-26 15:29:22 +0100 | <haskellbridge> | <Man of Letters (Mikolaj)> let me try my big project that uses https://hackage.haskell.org/package/dependent-enummap ... |
| 2026-01-26 15:28:32 +0100 | <haskellbridge> | ... long message truncated: https://kf8nh.com/_heisenbridge/media/kf8nh.com/NXWYnHcDHOAropDWOwElprDd/SxwevPmJWUA (7 lines) |
| 2026-01-26 15:28:31 +0100 | <haskellbridge> | <Man of Letters (Mikolaj)> yay, it works and in both directions: |
| 2026-01-26 15:26:43 +0100 | marinelli | (~weechat@gateway/tor-sasl/marinelli) marinelli |
| 2026-01-26 15:22:44 +0100 | <haskellbridge> | <Man of Letters (Mikolaj)> oh, so I _can_ define "Fst"; let me try in that case... |
| 2026-01-26 15:21:50 +0100 | machinedgod | (~machinedg@d75-159-126-101.abhsia.telus.net) machinedgod |
| 2026-01-26 15:21:01 +0100 | marinelli | (~weechat@gateway/tor-sasl/marinelli) (Remote host closed the connection) |
| 2026-01-26 15:20:55 +0100 | <Leary> | '(* |
| 2026-01-26 15:20:31 +0100 | <Leary> | Man of Letters (Mikolaj): What's unsound is `eta :: forall t. t :~: (Fst t, Snd t); eta = unsafeCoerce (Refl @t)`---so long as you aren't cooking with `unsafeCoerce` there won't be a problem. Re `Fst` and `Snd`, they're just ordinary closed type families. |
| 2026-01-26 15:16:14 +0100 | <ags> | What advantage does `Servant` have over `Yesod`? I understand that `Servant` is able to automatically generate a lot of stuff but it seems to me that `Yesod` is able to achieve the same type-safety via TemplateHaskell and QuasiQuotes. Am I missing something? |
| 2026-01-26 14:59:32 +0100 | <haskellbridge> | <Man of Letters (Mikolaj)> - '(a, b) |
| 2026-01-26 14:59:11 +0100 | <haskellbridge> | <Man of Letters (Mikolaj)> Leary: thank you; spot on, too bad, yeah, I guess I may try to use the internals of the libraries I use to write versions of some of their functions specifically for "(a, b), but it's a shame; if I had "Fst"and"Snd` maybe constraints in my own code would suffice, but probably these would trigger the same unsoundness that GHC assuming them does |
| 2026-01-26 14:58:12 +0100 | ags | (~ags@p200300dcb736f100c4bc8629e572e5b9.dip0.t-ipconnect.de) |
| 2026-01-26 14:56:02 +0100 | yin | (~zero@user/zero) zero |