Logs: liberachat/#haskell
| 2021-07-24 22:01:01 | <davean> | maerwald: I'm really upset anyone approved 1.2 |
| 2021-07-24 22:01:11 | <davean> | maerwald: I can't even see what was gained |
| 2021-07-24 22:01:24 | <davean> | I was litterly already doing everything with 1.1 1.2 has |
| 2021-07-24 22:01:40 | <maerwald> | I've had my own share of drama the past 2 months, so I won't argue against random 1.2 :p |
| 2021-07-24 22:01:46 | <davean> | maerwald: is there somewhere to file an objection? |
| 2021-07-24 22:01:51 | <dminuoso> | 23:59:48 davean | we went from a type class with some meaning to one where you can't even be sure you can call the functions in the type class |
| 2021-07-24 22:02:04 | <maerwald> | davean: for what? It's too late |
| 2021-07-24 22:02:07 | <davean> | I'm really worried about the CLC approving 1.2 |
| 2021-07-24 22:02:09 | <dminuoso> | Maybe Im missing something, but can you give some examples? Im not seeing anything immediately obvious that's wrong with it |
| 2021-07-24 22:02:16 | <davean> | maerwald: the CLC's leadersihp franly |
| 2021-07-24 22:02:25 | <davean> | dminuoso: split - is it defined or not? |
| 2021-07-24 22:02:36 | <davean> | "maybe" |
| 2021-07-24 22:02:47 | <davean> | How do you know? |
| 2021-07-24 22:02:50 | <davean> | Uh, call it and find out! |
| 2021-07-24 22:02:54 | → | Cajun joins (~Cajun@ip98-163-211-112.no.no.cox.net) |
| 2021-07-24 22:03:52 | <davean> | maerwald: Frankly I tihnk the only correct move is to mark 1.2 broken |
| 2021-07-24 22:04:32 | <dminuoso> | davean: Mmm. So why didn't they just make it.. a subclass? |
| 2021-07-24 22:04:47 | <dminuoso> | The usual way to further constrain a type by additional mechanisms.. |
| 2021-07-24 22:04:56 | <davean> | dminuoso: because that was the essence of the class |
| 2021-07-24 22:05:06 | <davean> | dminuoso: I have no idea what happened with 1.2 |
| 2021-07-24 22:05:20 | <davean> | I came across 1.2 a few days ago updating stuff and was horrified it ever was approved |
| 2021-07-24 22:05:31 | <dminuoso> | davean: No I meant, couldn't they just have added a RandomGenSplit typeclass with RandomGen as a superclass, and be done with it? |
| 2021-07-24 22:06:01 | <davean> | dminuoso: I mean if they'd tried, sure - though I'd suggest mots of those functions don't belong in class persay |
| 2021-07-24 22:06:43 | <davean> | One might use them for optimization |
| 2021-07-24 22:06:54 | <janus> | davean: have you previously voiced your concerns in a forum that isn't as ephemeral? (not rhetorical, just curious) |
| 2021-07-24 22:07:24 | <davean> | janus: what, email the CLC who approve this disaster? |
| 2021-07-24 22:07:37 | <davean> | No, because they're the ones who I have a problem with clearly |
| 2021-07-24 22:07:42 | <davean> | I have no diea who I'd voice this to |
| 2021-07-24 22:07:46 | geekosaur | subscribed tp the CLC list precisely to see what kind of things were in the pipeline |
| 2021-07-24 22:07:50 | <maerwald> | davean: they're not very responsive on emails either |
| 2021-07-24 22:07:57 | <davean> | maerwald: no they aren't |
| 2021-07-24 22:08:04 | <davean> | I'm subscribed and they don't reply to most tihngs |
| 2021-07-24 22:08:09 | <geekosaur> | but I missed this one |
| 2021-07-24 22:08:15 | <maerwald> | I'm not blaming ppl for having no time... but if you don't have time, ask for help |
| 2021-07-24 22:08:37 | <davean> | I have no idea who I'd complain to that I think the CLC has gone off the tracks |
| 2021-07-24 22:08:58 | <davean> | But I very much don't trust their leadership after seeing random 1.2 and they get to change a lot of important stuff |
| 2021-07-24 22:09:54 | <davean> | I'm worried about every other change they might make |
| 2021-07-24 22:09:55 | × | mikoto-chan quits (~mikoto-ch@ip-193-121-10-50.dsl.scarlet.be) (Ping timeout: 265 seconds) |
| 2021-07-24 22:10:06 | × | dunkeln quits (~dunkeln@94.129.69.87) (Ping timeout: 250 seconds) |
| 2021-07-24 22:12:04 | × | Null_A quits (~null_a@2601:645:8700:2290:809c:ae43:df71:6365) (Remote host closed the connection) |
| 2021-07-24 22:12:16 | <davean> | geekosaur: Did you see the thing about Data.List also? I completely didn't see any discussion of that and am rather worried |
| 2021-07-24 22:12:20 | × | Brumaire quits (~quassel@81-64-14-121.rev.numericable.fr) (Quit: ran away) |
| 2021-07-24 22:12:42 | <oso> | do you have a link or a term to search? just Data.List CLC? |
| 2021-07-24 22:12:44 | <davean> | As far as I can tell CLC is acting unilaterally without discussion or I'm bad at reading emmail |
| 2021-07-24 22:12:46 | → | Null_A joins (~null_a@2601:645:8700:2290:809c:ae43:df71:6365) |
| 2021-07-24 22:12:47 | <geekosaur> | I found out about it from #ghc |
| 2021-07-24 22:13:06 | <davean> | Has the CLC become a lose cannon out of control? |
| 2021-07-24 22:13:13 | <maerwald> | davean: there was a huge thread about random |
| 2021-07-24 22:13:19 | <maerwald> | libraries ML I think |
| 2021-07-24 22:13:30 | <maerwald> | lots of drama etc |
| 2021-07-24 22:14:57 | <davean> | maerwald: I guess I don't really care how it happened - the result is one I consider unacceptable |
| 2021-07-24 22:15:11 | <davean> | so I can't be ok with random 1.2 no matter what discussion happened |
| 2021-07-24 22:15:43 | <maerwald> | ok great, I can't fork process and use it in cabal.project, because of cabal not having any way to run autoreconf |
| 2021-07-24 22:16:14 | <davean> | maerwald: hum? |
| 2021-07-24 22:16:30 | <maerwald> | yes |
| 2021-07-24 22:16:50 | <davean> | can't you run autoreconfig manually? |
| 2021-07-24 22:17:18 | <maerwald> | how would you do that? Cabal clones the repo in some private dir and builds stuff, then fails |
| 2021-07-24 22:17:46 | <davean> | If you do it by reference, not if you do it by inclusion |
| 2021-07-24 22:17:53 | <davean> | make process a submodule |
| 2021-07-24 22:18:00 | <maerwald> | urg |
| 2021-07-24 22:18:30 | <davean> | I didn't say it was nice |
| 2021-07-24 22:18:38 | <maerwald> | time cabal gets hooks |
| 2021-07-24 22:18:42 | <jumper149> | maerwald: Use nix ;) |
| 2021-07-24 22:18:46 | → | dunkeln joins (~dunkeln@94.129.69.87) |
| 2021-07-24 22:18:48 | <maerwald> | jumper149: no |
| 2021-07-24 22:18:49 | <davean> | maerwald: it has cooks |
| 2021-07-24 22:18:55 | <maerwald> | what's a cook? |
| 2021-07-24 22:19:08 | <davean> | maerwald: You'd have to make it be in Setup.hs though |
| 2021-07-24 22:19:16 | <maerwald> | that's not a user hook |
| 2021-07-24 22:19:39 | <maerwald> | https://github.com/haskell/process/issues/51 |
| 2021-07-24 22:19:41 | <maerwald> | oops |
| 2021-07-24 22:19:46 | <maerwald> | my clipboard is broken |
| 2021-07-24 22:19:51 | <maerwald> | good nothing weird coming out of it |
| 2021-07-24 22:19:56 | <maerwald> | this https://github.com/haskell/cabal/issues/7394 |
| 2021-07-24 22:20:16 | <maerwald> | that would solve 2000 and 1 problems |
| 2021-07-24 22:20:23 | <maerwald> | seems it's WIP |
| 2021-07-24 22:20:31 | <davean> | maerwald: I mean your problem here is that the process library isn't a proper cabal package really |
| 2021-07-24 22:20:39 | <davean> | maerwald: I'm troubled by solving it externally for that reason |
| 2021-07-24 22:20:47 | <maerwald> | davean: yeah, let me argue with Snoyman real quick about it... |
| 2021-07-24 22:20:50 | <maerwald> | or... not |
| 2021-07-24 22:21:06 | × | neightchan quits (~nate@108-233-125-227.lightspeed.sntcca.sbcglobal.net) (Ping timeout: 240 seconds) |
| 2021-07-24 22:21:07 | <maerwald> | (because stack runs autoreconf) |
| 2021-07-24 22:21:07 | <davean> | maerwald: I know, I've never gotten even remote exploits fixed from that crowd so yah |
| 2021-07-24 22:21:19 | <maerwald> | (so I can infer his first 3 replies to my issue) |
| 2021-07-24 22:21:40 | <davean> | maerwald: But I'd prefer stuff be properly packaged than patch it ad-hoc everywhere else |
| 2021-07-24 22:22:05 | <maerwald> | autoreconf is generally a controversial thing, even in source distros |
| 2021-07-24 22:22:51 | <maerwald> | if you have old buggy macros, you get problems... if you run it unconditionally, you have roughly 10% of failures |
| 2021-07-24 22:23:16 | <davean> | yah, its a bad solution to the problem bu tthe best thats at hand a lot of hte time |
| 2021-07-24 22:23:27 | <maerwald> | and then you have the "don't ever commit generated files to git" people |
| 2021-07-24 22:24:05 | <davean> | anyway, I know why you want these hooks, and I tihnk you understand why I don't like them as a solution here - pragmatics might win out |
| 2021-07-24 22:24:18 | × | Null_A quits (~null_a@2601:645:8700:2290:809c:ae43:df71:6365) (Ping timeout: 240 seconds) |
| 2021-07-24 22:24:28 | <maerwald> | hooks are the *perfect* solution for this (and many other things) |
| 2021-07-24 22:24:34 | × | dunkeln quits (~dunkeln@94.129.69.87) (Ping timeout: 240 seconds) |
| 2021-07-24 22:24:44 | × | favonia quits (~favonia@user/favonia) (Ping timeout: 272 seconds) |
| 2021-07-24 22:24:47 | <davean> | I don't think I should have ot use hooks to depend on a package :/ |
| 2021-07-24 22:25:09 | <maerwald> | yes, this is dev foo |
| 2021-07-24 22:25:13 | <maerwald> | process needs to be fixed |
| 2021-07-24 22:25:33 | × | azeem quits (~azeem@176.201.2.247) (Read error: Connection reset by peer) |
| 2021-07-24 22:25:46 | → | azeem joins (~azeem@dynamic-adsl-84-220-213-65.clienti.tiscali.it) |
| 2021-07-24 22:25:47 | <maerwald> | but now I got so annoyed about it that I riled about it here instead of running autoreconf... |
All times are in UTC.