Logs on 2024-10-17 (liberachat/#xmonad)
| 00:14:44 | <haskellbridge> | <galactic_starfish> @irc_libera.chat_foxonsocks:kf8nh.com: that sounds like a profile or env var oddity. |
| 00:21:14 | <geekosaur> | they left |
| 00:22:39 | <geekosaur> | I was too busy to look into it at the time but would be interested in the log |
| 00:27:26 | <geekosaur> | *session log |
| 01:45:41 | <haskellbridge> | <柱間 (no dm, not friendly)> geekosaur: thanks for the response, i was a bit late ig |
| 01:46:27 | <geekosaur> | np |
| 01:46:29 | <haskellbridge> | <柱間 (no dm, not friendly)> geekosaur: well, if we can pin haskell revisions like this, then it's quite cool |
| 01:53:24 | <haskellbridge> | <柱間 (no dm, not friendly)> geekosaur: does it mean that haskell2010 will always work? |
| 01:53:53 | <haskellbridge> | <柱間 (no dm, not friendly)> if i write a software today, can i rest assured knowing that it wont break ? |
| 01:55:29 | <geekosaur> | no, sadly. there have been library, as opposed to language, breakages |
| 01:56:44 | <geekosaur> | (Richard Bird was very grumoy about this IIRC) |
| 01:57:12 | <haskellbridge> | <柱間 (no dm, not friendly)> geekosaur: can you explain? because i saw haskell projects that are 20y old and still work |
| 01:57:48 | <geekosaur> | they get updated as needed, generally. there are also Haskell projects that haven't been updated |
| 01:58:02 | <geekosaur> | breaking library changes off the top of my head: |
| 01:58:15 | <geekosaur> | - Num lost Eq and Show "superclasses" |
| 01:58:32 | <geekosaur> | - Monad gained Applicative "superclass" (and thereby also Functor) |
| 01:58:48 | <geekosaur> | - "fail" was removed from Monad and moved to a separate MonadFail typeclass |
| 01:58:53 | <haskellbridge> | <柱間 (no dm, not friendly)> so cant someone just lock it in a specific revision |
| 01:59:22 | <geekosaur> | - pretty much everything that used to be specific to lists either uses Foldable or Traversable now |
| 01:59:53 | <haskellbridge> | <柱間 (no dm, not friendly)> geekosaur: but even tho some things got removed, they still in some older revisions, right? |
| 02:00:27 | <geekosaur> | yes. all f these could be gotten back with an older version of "base" |
| 02:00:43 | <haskellbridge> | <柱間 (no dm, not friendly)> locking libraries versions is a thing to keep in mind, if someone does that and uses a specific revision (let's say from 2021) then there's no issue , right? |
| 02:00:54 | <geekosaur> | the problem is ,the compiler has parts of "base" wired-in, so if you want some particular version of "base" you must use the compiler that comes with it |
| 02:01:23 | <haskellbridge> | <柱間 (no dm, not friendly)> geekosaur: oh, a specific version of GHC then? |
| 02:02:01 | <geekosaur> | yes |
| 02:02:09 | <geekosaur> | well, major version at least |
| 02:03:19 | <haskellbridge> | <柱間 (no dm, not friendly)> well, i dont see any problem then, the approach could be: |
| 02:03:19 | <haskellbridge> | ... long message truncated: https://kf8nh.com/_heisenbridge/media/kf8nh.com/MToxpDZZcKcKNdHmIBWsdCCM/yTxXZntkGKo (4 lines) |
| 02:03:24 | <haskellbridge> | <柱間 (no dm, not friendly)> well, i dont see any problem then, the approach could be: |
| 02:03:24 | <haskellbridge> | ... long message truncated: https://kf8nh.com/_heisenbridge/media/kf8nh.com/NiLaftbHSFgWxwudpczZaOTI/dkvteibTG_s (4 lines) |
| 02:03:29 | <haskellbridge> | <柱間 (no dm, not friendly)> * think |
| 02:05:07 | <geekosaur> | this si more or less what stack does |
| 02:05:44 | <geekosaur> | the downside is that eventually older compilers won't work on newer systems (7.10.x crashes at startup on Ubuntu 22.04 and up, for example) |
| 02:06:16 | <haskellbridge> | <柱間 (no dm, not friendly)> why is this? |
| 02:06:54 | <geekosaur> | because it's not just ghc that is changing |
| 02:07:03 | <haskellbridge> | <柱間 (no dm, not friendly)> geekosaur: pretty cool then, i might look forward into how stack does it, and copy to a more cmake-like structure (im using c++ with haskell so...) |
| 02:07:40 | <haskellbridge> | <柱間 (no dm, not friendly)> geekosaur: i'm sorry to take your time, but if you could explain a bit more, is this somehow related to the kernel itself? like any operation not being supported anymore |
| 02:07:47 | <geekosaur> | with stack you develop with a snapshoot (see https://stackage.org) |
| 02:08:07 | <geekosaur> | it includes a compiler and a collection of libraries at versions known to work with the compiler and each other |
| 02:08:48 | <geekosaur> | it may be glibc, it may be other system dependencies, I don't think anyone's actually looked into why it's happening |
| 02:09:56 | <haskellbridge> | <柱間 (no dm, not friendly)> i see |
| 02:10:13 | <haskellbridge> | <柱間 (no dm, not friendly)> geekosaur: this is really cool, i wonder how they do it |
| 02:10:44 | <haskellbridge> | <柱間 (no dm, not friendly)> but like, someone can just compile the software in a stateless system and ship with all dependencies to other OSes |
| 02:11:15 | <geekosaur> | that was the idea behind MUSL |
| 02:11:36 | <geekosaur> | the problem there is that the kernel's interface is not fixed, so libcs must track how it changes |
| 02:11:42 | <haskellbridge> | <柱間 (no dm, not friendly)> at least here at guix i can use the command "guix pack" to do it, there might exist something similar in nix, and as i've already saw, nix is better when it comes to reproducibility, because there's no side effects (pure) |
| 02:12:07 | <geekosaur> | glibc and MUSL contain a lot of code to adapt to the kernel at runtime. if they detect an unsupported kernel, they crash |
| 02:13:12 | <haskellbridge> | <柱間 (no dm, not friendly)> damn |
| 02:13:43 | <haskellbridge> | <柱間 (no dm, not friendly)> but there might be a way around it |
| 02:18:38 | <haskellbridge> | <geekosaur (@geekosaur:matrix.org)> Let everyone know if you find it |
| 02:20:53 | <haskellbridge> | <柱間 (no dm, not friendly)> ofc =) |
| 02:25:57 | <haskellbridge> | <柱間 (no dm, not friendly)> damn, this is a lot of workarounds... |
| 02:25:57 | <haskellbridge> | https://git.savannah.gnu.org/cgit/guix.git/tree/gnu/packages/haskell.scm#n702 |
| 02:26:19 | <haskellbridge> | <柱間 (no dm, not friendly)> but like, it works |
| 02:26:19 | <haskellbridge> | but it stops every C-library from having trouble with it, lol |
| 02:27:23 | <haskellbridge> | <柱間 (no dm, not friendly)> wow, there's even ghc6.6 in there... take a look |
| 02:28:02 | <haskellbridge> | <柱間 (no dm, not friendly)> there's something to glibc if you read above too, maybe on ghc4.0 or smth, just on the top |
| 02:28:16 | <haskellbridge> | <柱間 (no dm, not friendly)> those are workarounds, but they're quite cool |
| 02:47:20 | × | td_ quits (~td@i53870917.versanet.de) (Ping timeout: 252 seconds) |
| 02:49:06 | → | td_ joins (~td@i53870903.versanet.de) |
| 04:48:24 | × | alp quits (~alp@2001:861:e3d6:8f80:cfc1:c60c:c177:1e26) (Ping timeout: 246 seconds) |
| 05:23:02 | → | alp joins (~alp@2001:861:e3d6:8f80:c1d0:38:3b03:da86) |
| 05:33:09 | × | alp quits (~alp@2001:861:e3d6:8f80:c1d0:38:3b03:da86) (Ping timeout: 248 seconds) |
| 05:38:53 | → | alp joins (~alp@2001:861:e3d6:8f80:689:f2c1:b3cb:79a1) |
| 06:48:32 | × | ft quits (~ft@p4fc2a216.dip0.t-ipconnect.de) (Quit: leaving) |
| 06:52:56 | → | ash3en joins (~Thunderbi@31.16.203.214) |
| 07:07:01 | × | ash3en quits (~Thunderbi@31.16.203.214) (Quit: ash3en) |
| 09:20:21 | → | ash3en joins (~Thunderbi@2a01:780:8000:115:27a0:cdec:aee3:14b8) |
| 09:21:07 | × | ash3en quits (~Thunderbi@2a01:780:8000:115:27a0:cdec:aee3:14b8) (Client Quit) |
| 09:28:27 | → | Digitteknohippie joins (~user@user/digit) |
| 09:28:54 | × | Digit quits (~user@user/digit) (Ping timeout: 260 seconds) |
| 09:56:48 | Digitteknohippie | is now known as Digit |
| 10:26:06 | → | yaslam joins (~yaslam@user/yaslam) |
| 11:50:11 | → | ash3en joins (~Thunderbi@149.222.128.226) |
| 11:58:10 | × | ash3en quits (~Thunderbi@149.222.128.226) (Quit: ash3en) |
| 12:01:09 | → | Digitteknohippie joins (~user@user/digit) |
| 12:01:22 | × | Digit quits (~user@user/digit) (Ping timeout: 252 seconds) |
| 12:05:56 | → | ash3en joins (~Thunderbi@149.222.128.226) |
| 12:11:25 | Digitteknohippie | is now known as Digit |
| 12:44:45 | × | alp quits (~alp@2001:861:e3d6:8f80:689:f2c1:b3cb:79a1) (Ping timeout: 246 seconds) |
| 12:50:31 | × | catman quits (~catman@user/catman) (Quit: WeeChat 4.4.2) |
| 13:14:51 | → | catman joins (~catman@user/catman) |
| 13:22:11 | × | ash3en quits (~Thunderbi@149.222.128.226) (Quit: ash3en) |
| 14:15:51 | → | Digitteknohippie joins (~user@user/digit) |
| 14:16:39 | × | Digit quits (~user@user/digit) (Ping timeout: 252 seconds) |
| 14:23:07 | Digitteknohippie | is now known as Digit |
| 14:27:34 | → | rekahsoft joins (~rekahsoft@76.69.85.220) |
| 14:43:31 | → | ash3en joins (~Thunderbi@193.32.248.167) |
| 14:50:01 | × | ash3en quits (~Thunderbi@193.32.248.167) (Quit: ash3en) |
| 15:11:03 | × | hiecaq`` quits (~hiecaq`@user/hiecaq) (Quit: ERC 5.5.0.29.1 (IRC client for GNU Emacs 29.4)) |
| 15:28:59 | → | alp joins (~alp@2001:861:e3d6:8f80:553e:4712:5919:60eb) |
| 15:50:32 | × | Digit quits (~user@user/digit) (Ping timeout: 255 seconds) |
| 16:00:21 | → | Digit joins (~user@user/digit) |
| 16:18:15 | × | alp quits (~alp@2001:861:e3d6:8f80:553e:4712:5919:60eb) (Ping timeout: 246 seconds) |
| 16:52:13 | → | alp joins (~alp@2001:861:e3d6:8f80:34e0:ead3:355f:5f57) |
| 17:10:37 | → | ft joins (~ft@p4fc2a216.dip0.t-ipconnect.de) |
| 18:43:34 | → | Digitteknohippie joins (~user@user/digit) |
| 18:44:55 | → | ircbrowse_tom joins (~ircbrowse@user/tomsmeding/bot/ircbrowse-tom) |
| 18:44:59 | Server | sets mode +cnt |
| 18:46:11 | × | bsima quits (~bsima@2604:a880:400:d0::19f1:7001) (Quit: ZNC 1.8.2 - https://znc.in) |
| 18:46:34 | → | bsima joins (~bsima@143.198.118.179) |
| 18:54:09 | × | alp quits (~alp@2001:861:e3d6:8f80:34e0:ead3:355f:5f57) (Ping timeout: 252 seconds) |
| 18:59:54 | Digitteknohippie | is now known as Digit |
| 19:12:52 | → | alp joins (~alp@2001:861:e3d6:8f80:892c:a656:e37a:e5b6) |
| 19:18:06 | × | alp quits (~alp@2001:861:e3d6:8f80:892c:a656:e37a:e5b6) (Ping timeout: 272 seconds) |
| 19:37:39 | → | alp joins (~alp@2001:861:e3d6:8f80:c611:3442:c342:856d) |
| 21:39:18 | → | rascasse joins (~rascasse@user/diep) |
| 22:22:26 | × | rascasse quits (~rascasse@user/diep) (Remote host closed the connection) |
All times are in UTC on 2024-10-17.