Home freenode/#haskell: Logs Calendar

Logs: freenode/#haskell

←Prev  Next→
Page 1 .. 1431 1432 1433 1434 1435 1436 1437 1438 1439 1440 1441 .. 5022
— 502,152 events total
2020-11-21 20:20:06 <zyklotomic> yup fair, no shortcut
2020-11-21 20:20:09 <dminuoso> Without it, you'd just be a victim to do whatever random folks on freenode told you
2020-11-21 20:20:38 <geekosaur> that said, "a little knowledge is a dangerous thing"
2020-11-21 20:21:08 <merijn> So is a lot of knowledge ;)
2020-11-21 20:22:17 ski . o O ( <https://www.vex.net/~trebla/humour/tautologies.html#6> )
2020-11-21 20:22:34 → jonatanb joins (jonatanb@gateway/vpn/protonvpn/jonatanb)
2020-11-21 20:22:34 <merijn> I saw Vanessa's tweet over "being the Dickens of code" and my initial reaction was "I'm the H.P. Lovecraft of code", in that my code describes a slow descent into gibbering madness that unsettles the reader and makes them question their place in the universe... >.>
2020-11-21 20:24:35 <merijn> monochrom: I've got an addendum to your xkcd link on that page ;)
2020-11-21 20:24:44 × oish quits (~charlie@228.25.169.217.in-addr.arpa) (Ping timeout: 265 seconds)
2020-11-21 20:24:54 <merijn> monochrom: https://www.smbc-comics.com/comic/2014-09-02
2020-11-21 20:25:43 → falafel__ joins (~falafel@2601:547:1303:b30:7811:313f:d0f3:f9f4)
2020-11-21 20:26:27 × justanotheruser quits (~justanoth@unaffiliated/justanotheruser) (Ping timeout: 272 seconds)
2020-11-21 20:26:44 → guest29 joins (5f4965a4@95.73.101.164)
2020-11-21 20:28:42 → heatsink joins (~heatsink@107-136-5-69.lightspeed.sntcca.sbcglobal.net)
2020-11-21 20:29:19 <dsal> zyklotomic: I ask @pl for suggestions sometimes and if they're obviously better, I use them. hlint also gives good recommendations sometimes. Only rarely does it give really bad advice, IMO.
2020-11-21 20:29:42 <zyklotomic> yeah hlint has really helped
2020-11-21 20:29:58 <zyklotomic> i, just seconds ago, as a matter of fact, wondered
2020-11-21 20:30:07 <dsal> hlint got mad at me for using [Char] somewhere. I said what I meant.
2020-11-21 20:30:07 <merijn> hlint's advice is contentious :p
2020-11-21 20:30:09 <zyklotomic> how did I write such a clean loc, "oh wait hlint did"
2020-11-21 20:30:15 × danvet quits (~Daniel@2a02:168:57f4:0:efd0:b9e5:5ae6:c2fa) (Ping timeout: 272 seconds)
2020-11-21 20:30:23 <dsal> Yeah, all opinions are bad. Code formatters can be pretty awful.
2020-11-21 20:30:23 → Jesin joins (~Jesin@pool-72-66-101-18.washdc.fios.verizon.net)
2020-11-21 20:30:48 <dsal> But when I had no idea what I was doing, some opinions were better than my own much of the time.
2020-11-21 20:30:49 <dminuoso> I dont linters at all.
2020-11-21 20:31:14 <dminuoso> They tend to create noise for disagreeing on my style, and then I spend more time tweaking the linter rules than it improves my code
2020-11-21 20:31:25 <dminuoso> Because I refuse to blindly obey hlint
2020-11-21 20:31:51 <dsal> I run it on occasion and it sometimes points out a dumb thing I did in refactoring.
2020-11-21 20:31:58 <merijn> hlint is great
2020-11-21 20:32:03 <merijn> If you're Neil Mitchell :)
2020-11-21 20:32:16 <glguy> hlint is great for adding cpp noise to your code as you block out regions that it can't handle
2020-11-21 20:32:39 <dsal> I think it's pretty good if you're learning and you want some degree of code review without bugging humans constantly.
2020-11-21 20:32:40 <zyklotomic> as long as it's easy to tell the linter to shut up where you want to? i haven't gotten to the stage where I trust my style more
2020-11-21 20:32:46 <zyklotomic> yeah
2020-11-21 20:32:49 ski . o O ( "He is also known for crediting his computer \"Shalosh B. Ekhad\" as a co-author" <https://en.wikipedia.org/wiki/Shalosh_B._Ekhad> )
2020-11-21 20:33:03 <glguy> The problem with leaning on it as a new user is you don't know which suggestions are ones you should have avoided
2020-11-21 20:34:48 <dsal> I guess it depends on how you approach it. I never look at it to ask if my code is good. Just another opinion. Sometimes interesting ones.
2020-11-21 20:35:22 <hpc> the trouble with hlint is there's not much it can do to catch things that matter
2020-11-21 20:35:35 <hpc> compared to say, shellcheck which is almost compiler levels of valuable
2020-11-21 20:36:04 → cr3 joins (~cr3@192-222-143-195.qc.cable.ebox.net)
2020-11-21 20:36:04 <zyklotomic> i think at least as a very very novice, doing things like map (\x -> x+3), it was invaluable
2020-11-21 20:36:32 <zyklotomic> it should get recommended more
2020-11-21 20:36:41 → alp joins (~alp@2a01:e0a:58b:4920:c8af:eb98:65bc:27e4)
2020-11-21 20:36:52 <dminuoso> zyklotomic: As you can see there's a lot of disagreement in the past few minutes already.
2020-11-21 20:36:55 <zyklotomic> and well, you'll know when you don't need it
2020-11-21 20:36:56 × chkno quits (~chkno@75-7-2-127.lightspeed.sntcca.sbcglobal.net) (Read error: Connection reset by peer)
2020-11-21 20:37:03 → chkno joins (~chkno@75-7-2-127.lightspeed.sntcca.sbcglobal.net)
2020-11-21 20:37:31 <zyklotomic> dminuoso: :p, well specifically from the perspective of a complete noob though
2020-11-21 20:37:48 <[exa]> hlint is GREAT for newbies
2020-11-21 20:38:02 <monochrom> hlint is a very misleading name.
2020-11-21 20:38:12 <monochrom> hlint is hardly comparable to C lint.
2020-11-21 20:38:14 <[exa]> yeah, hhint
2020-11-21 20:38:21 <dminuoso> At work I have a colleague who runs his python linter in the CI. So if you submit a merge request, and you use an underscore in a place that the linter things is "terrible style", it is undeployable.
2020-11-21 20:38:34 <maerwald> hlint is annoying in CI
2020-11-21 20:38:41 <monochrom> Whereas C lint points out very real mistakes, hlint is merely style nitpicking.
2020-11-21 20:38:41 <dsal> Yeah, it never tells me my code is wrong, just stuff like, "Hey, `fmap f . fmap g` is `fmap (f . g)`
2020-11-21 20:38:43 <dminuoso> So Im fairly allergic to linters. Beginners trust them too much
2020-11-21 20:38:43 <dsal> "
2020-11-21 20:38:48 <sszark> has hackage always not had a SSL certificate? or is that a mistake
2020-11-21 20:39:09 <dminuoso> sszark: What do you mean? The website?
2020-11-21 20:39:13 <monochrom> The damage is that beginners look at the name "hlint" and don't know it's a blatant lie.
2020-11-21 20:39:17 <zyklotomic> yah i was going to say that double fmap rule was eye opening
2020-11-21 20:39:25 <glguy> I agree they're annoying in CI, but if you accept the assumption that it's a good idea to use it at all it doesn't seem like a bad idea to enforce people clean up the suggestions
2020-11-21 20:39:35 <hpc> sszark: it just doesn't redirect to https
2020-11-21 20:39:41 <zyklotomic> dsal: i notice that when i'm writing javascript too now, it's crazy
2020-11-21 20:39:46 <glguy> having long-running lint suggestions clouds seeing the new arguably useful ones, right?
2020-11-21 20:40:14 <sszark> oh i see, that makes sense i guess hpc
2020-11-21 20:40:21 <maerwald> well, the problem is if you disagree with most hlint suggestions and the ones that I don't disagree with are not worth the time
2020-11-21 20:40:30 <monochrom> Like, there is a reason I write "\x -> f x" when teaching beginners. Shut up, hlint.
2020-11-21 20:40:49 <merijn> sszark: hackage does have SSL, but there's no autoforward
2020-11-21 20:41:00 <monochrom> No, actually, I should speak like hlint does. Why don't you shut the f* up, hlint.
2020-11-21 20:41:13 <glguy> relying on autoforward isn't a great plan. explicitly use https if that's what you want
2020-11-21 20:41:13 <dminuoso> Also, you should name it TLS. SSL is usually used to refer to the old versions of the protocol suite. :)
2020-11-21 20:41:14 <merijn> sszark: There was an email about setting up an autoforward the past week
2020-11-21 20:41:25 <merijn> sszark: Anyway, hackage doesn't need SSL for security anyway
2020-11-21 20:41:38 <merijn> (well, assuming you got cabal-install from a trusted source :p)
2020-11-21 20:42:05 × guest29 quits (5f4965a4@95.73.101.164) (Remote host closed the connection)
2020-11-21 20:42:09 <maerwald> merijn: and how do you verify the hackage index?
2020-11-21 20:42:15 <merijn> maerwald: It's signed
2020-11-21 20:42:33 <maerwald> with a human key?
2020-11-21 20:42:33 × chkno quits (~chkno@75-7-2-127.lightspeed.sntcca.sbcglobal.net) (Read error: Connection reset by peer)
2020-11-21 20:42:38 <merijn> maerwald: You need an uncompromised cabal-install first, but you can't download that from Hackage anyway
2020-11-21 20:42:50 <dsal> Transport isn't going to verify your index
2020-11-21 20:42:50 → triteraflops joins (~triterafl@host-148-170-141-28.public.eastlink.ca)
2020-11-21 20:43:04 → chkno joins (~chkno@75-7-2-127.lightspeed.sntcca.sbcglobal.net)
2020-11-21 20:43:12 <merijn> maerwald: Hackage uses TUF, so, eh, look up the details yourself? https://theupdateframework.io/
2020-11-21 20:43:22 → justanotheruser joins (~justanoth@unaffiliated/justanotheruser)
2020-11-21 20:44:17 <triteraflops> I decided linear programming is a good idea, and would like to experiment further with it. How "ready" is linear haskell? Could it plausibly be used on a real world project without anything catching fire?
2020-11-21 20:44:38 <merijn> triteraflops: Considering it requires an unreleased GHC "no" :p
2020-11-21 20:44:55 <triteraflops> lol
2020-11-21 20:44:58 <triteraflops> irp
2020-11-21 20:44:59 <triteraflops> rip
2020-11-21 20:45:40 × Tario quits (~Tario@201.192.165.173) (Read error: Connection reset by peer)
2020-11-21 20:45:40 <merijn> You could, conceivably, build something non-trivial by grabbing a build from CI and using head-hackages, but it's very much a "DIY ductape and wire" kinda stage"
2020-11-21 20:46:28 <triteraflops> "
2020-11-21 20:47:30 <triteraflops> Well, what about Idris? I know I should ask #idris, but what the hey.
2020-11-21 20:47:41 <merijn> Does Idris even have linear types?
2020-11-21 20:47:48 <triteraflops> idris2 does
2020-11-21 20:50:24 × xsperry quits (~as@unaffiliated/xsperry) ()
2020-11-21 20:50:27 <MarcelineVQ> there's no no-fire gurantees on any software
←Prev  Next→
Page 1 .. 1431 1432 1433 1434 1435 1436 1437 1438 1439 1440 1441 .. 5022

All times are in UTC.