Logs: freenode/#haskell
| 2021-04-29 15:50:36 | <maerwald> | a) access to software developers, b) access to cheap software developers, c) bus factor |
| 2021-04-29 15:51:09 | <tdammers> | however d) access to non-shitty software developers |
| 2021-04-29 15:51:23 | <tdammers> | and e) guaranteeing an absolute minimum baseline of "documentation" |
| 2021-04-29 15:51:38 | <Taneb> | f) fun |
| 2021-04-29 15:52:02 | <tdammers> | that's actually a valid business concern |
| 2021-04-29 15:52:02 | <kuribas> | a) 1 we have good developers, b) we pay them a good salary c) I'll train them |
| 2021-04-29 15:52:20 | <maerwald> | ok, so you're the only experienced haskell dev? |
| 2021-04-29 15:52:46 | <kuribas> | maerwald: nah, there are many, the guys here, well-typed, etc... |
| 2021-04-29 15:52:58 | <maerwald> | kuribas: I mean in your team |
| 2021-04-29 15:53:16 | <kuribas> | currently, yes |
| 2021-04-29 15:53:40 | <tdammers> | fwiw, if you screw up and hire well-typed to mop up the pieces, I won't object |
| 2021-04-29 15:53:43 | <maerwald> | what happens if you get sick or leave the company, because well-typed made you an offer? ;) |
| 2021-04-29 15:54:02 | <kuribas> | maerwald: depends how well the service does? |
| 2021-04-29 15:54:06 | × | acidjnk_new2 quits (~acidjnk@p5487d90a.dip0.t-ipconnect.de) (Ping timeout: 240 seconds) |
| 2021-04-29 15:54:17 | <tomsmeding> | given "data Dir = Fwd | Rev ; type family Flip (a :: Dir) = r | r -> a where Flip 'Fwd = 'Rev ; Flip 'Rev = 'Fwd", is there a nice way to prove to ghc that "a ~ Flip (Flip a)" ? |
| 2021-04-29 15:54:31 | <kuribas> | maerwald: if it happens in the next month (unlikely) it'll get rewritten in clojure. |
| 2021-04-29 15:54:32 | <tomsmeding> | for a :: Dir |
| 2021-04-29 15:54:43 | × | merijn quits (~merijn@83-160-49-249.ip.xs4all.nl) (Ping timeout: 252 seconds) |
| 2021-04-29 15:55:10 | × | Qwerky_ quits (~qwerky@37.170.132.154) (Remote host closed the connection) |
| 2021-04-29 15:55:12 | <maerwald> | so in 3 years, you are gone, the service uses an ancient stackage LTS that you can't build with current stack anymore and the CTO orders to rewrite it in Go, because they can't compete with blockchain companies for haskell dev salaries :p |
| 2021-04-29 15:55:12 | <kuribas> | maerwald: if it happens next year, I suppose the other guys should know enough to get shit done. |
| 2021-04-29 15:56:07 | × | conal quits (~conal@64.71.133.70) (Ping timeout: 265 seconds) |
| 2021-04-29 15:56:14 | <kuribas> | I doubt it. Not go for sure, maybe clojure. |
| 2021-04-29 15:56:18 | × | zva quits (~zava@ip5f5bdf0f.dynamic.kabel-deutschland.de) (Ping timeout: 260 seconds) |
| 2021-04-29 15:56:22 | <minoru_shiraeesh> | write everything in pseudo-code and transpile to actual code |
| 2021-04-29 15:56:51 | <kuribas> | maerwald: or more likely, when I leave they hire another guy who knows haskell well. |
| 2021-04-29 15:57:14 | <maerwald> | tbh, I've always said "no" if someone asked me if haskell is a good idea for some commercial project |
| 2021-04-29 15:57:31 | <maerwald> | Either CTO and CEO know the deal and are all-in on it, or don't do it |
| 2021-04-29 15:57:31 | <kuribas> | maerwald: I guess you are in the wrong channel then? |
| 2021-04-29 15:57:32 | × | Pickchea quits (~private@unaffiliated/pickchea) (Ping timeout: 240 seconds) |
| 2021-04-29 15:57:36 | → | heatsink joins (~heatsink@108-201-191-115.lightspeed.sntcca.sbcglobal.net) |
| 2021-04-29 15:57:38 | <maerwald> | why is that? |
| 2021-04-29 15:58:00 | × | mikoto-chan quits (~mikoto-ch@gateway/tor-sasl/mikoto-chan) (Quit: mikoto-chan) |
| 2021-04-29 15:58:06 | <maerwald> | I'm not talking about open source or ppl who know what they are doing. I'm talking about someone asking what they should do |
| 2021-04-29 15:58:09 | <kuribas> | maerwald: if you don't like haskell, this isn't the place? |
| 2021-04-29 15:58:18 | <maerwald> | kuribas: sorry what? |
| 2021-04-29 15:58:26 | <tomsmeding> | liking haskell is something else than thinking it's good for task X |
| 2021-04-29 15:58:36 | × | coot quits (~coot@37.30.58.122.nat.umts.dynamic.t-mobile.pl) (Quit: coot) |
| 2021-04-29 15:58:42 | <tomsmeding> | where here, task X is quite broad: using it as a company |
| 2021-04-29 15:58:47 | <kuribas> | tomsmeding: eh, "haskell is wrong for commercial use" is very generic. |
| 2021-04-29 15:58:53 | <maerwald> | that's not what I said either |
| 2021-04-29 15:59:10 | <kuribas> | tomsmeding: that's very different from saying, haskell isn't suited for this particular task. |
| 2021-04-29 15:59:18 | <tomsmeding> | true |
| 2021-04-29 15:59:48 | <maerwald> | If someone asks me "We will use haskell in our commercial project, will you help us?" I will likely say yes ;) |
| 2021-04-29 15:59:50 | → | conal joins (~conal@64.71.133.70) |
| 2021-04-29 15:59:58 | <maerwald> | but if they ask me if they should use haskell, I say "no" |
| 2021-04-29 16:00:03 | × | adziahel[m] quits (adziahelma@gateway/shell/matrix.org/x-qwxmblcrsukkaonc) (Quit: Idle for 30+ days) |
| 2021-04-29 16:00:17 | <kuribas> | maerwald: it depends. |
| 2021-04-29 16:00:20 | × | L29Ah quits (~L29Ah@unaffiliated/l29ah) (Quit: Gateway shutdown) |
| 2021-04-29 16:00:32 | <kuribas> | maerwald: "yes" isn't a good generic answer, neither is "no". |
| 2021-04-29 16:01:33 | <kuribas> | maerwald: it depends on company culture, existing practices, expectations, scope of project, etc... |
| 2021-04-29 16:01:47 | <maerwald> | it depends on them asking me the question :p |
| 2021-04-29 16:03:47 | <maerwald> | if you ask a Buddhist, if you should become a Buddhist, they'll either say "no" or say nothing |
| 2021-04-29 16:03:59 | <maerwald> | saying nothing is kinda rude, so... |
| 2021-04-29 16:04:27 | <kuribas> | maerwald: it's biased against haskell, so I wonder why you are here. |
| 2021-04-29 16:04:28 | <tdammers> | a Zen Buddhist might answer "mu." |
| 2021-04-29 16:04:33 | <maerwald> | hehe |
| 2021-04-29 16:04:33 | × | geekosaur quits (930099da@rrcs-147-0-153-218.central.biz.rr.com) (Quit: Connection closed) |
| 2021-04-29 16:04:39 | → | Qwerky joins (~qwerky@37.170.132.154) |
| 2021-04-29 16:04:45 | <maerwald> | kuribas: no |
| 2021-04-29 16:04:55 | <kuribas> | maerwald: I don't find you are either enlightend, nor objective. |
| 2021-04-29 16:05:02 | <kuribas> | not that I am... |
| 2021-04-29 16:05:12 | <tdammers> | alternatively: "There is no 'Buddhism'. There is no 'should'. There is no 'I'." |
| 2021-04-29 16:05:24 | <maerwald> | kuribas: I don't engage in ad-hominemisms :p |
| 2021-04-29 16:05:39 | <maerwald> | tdammers: that's too abstract for me :> |
| 2021-04-29 16:05:42 | → | howdoi joins (uid224@gateway/web/irccloud.com/x-qncwqauoexuittvy) |
| 2021-04-29 16:06:04 | <maerwald> | but in general, I despise envangelism in tech |
| 2021-04-29 16:06:06 | → | geekosaur joins (930099da@rrcs-147-0-153-218.central.biz.rr.com) |
| 2021-04-29 16:06:17 | <kuribas> | maerwald: I am not saying you should be biased for haskell, it's just strange that you are here... |
| 2021-04-29 16:06:26 | × | hypercube quits (~hypercube@2603-6011-f901-9e5b-0000-0000-0000-08cf.res6.spectrum.com) (Quit: WeeChat 3.1) |
| 2021-04-29 16:06:36 | <maerwald> | kuribas: I find it strange too... world is full of miracles :> |
| 2021-04-29 16:06:54 | → | ddellac__ joins (ddellacost@gateway/vpn/mullvad/ddellacosta) |
| 2021-04-29 16:07:06 | <tdammers> | kuribas: he's kind of right though. as fantastic Haskell is as a language, from a business perspective it has downsides that can prove crucial |
| 2021-04-29 16:07:28 | <maerwald> | once I recommended rust to a startup company... sometimes I have nightmares about how they're doing |
| 2021-04-29 16:08:05 | <maerwald> | but my data point was: Microsoft rewrote some of their stuff in rust and achieved the same velocity after a couple of weeks |
| 2021-04-29 16:08:08 | <kuribas> | tdammers: what about clojure then? |
| 2021-04-29 16:08:10 | → | electricityZZZZ joins (~electrici@157-131-250-97.fiber.dynamic.sonic.net) |
| 2021-04-29 16:08:19 | <kuribas> | tdammers: is clojure "better" from a business perspective? |
| 2021-04-29 16:08:43 | <maerwald> | with haskell, your velocity is harder to predict as well |
| 2021-04-29 16:08:45 | geekosaur | would like to point out that clojure is a possibility because of evangelism, which maerwald has rejected |
| 2021-04-29 16:08:49 | <maerwald> | sometime way faster, sometimes not |
| 2021-04-29 16:08:50 | <minoru_shiraeesh> | otoh graham hutton, iirc, described lisp as one of his advantages over competition |
| 2021-04-29 16:08:52 | <tdammers> | kuribas: clojure has some advantages that Haskell lacks, most notably running on JVM and having relatively painless Java interop |
| 2021-04-29 16:08:59 | → | L29Ah joins (~L29Ah@unaffiliated/l29ah) |
| 2021-04-29 16:09:12 | × | Qwerky quits (~qwerky@37.170.132.154) (Ping timeout: 240 seconds) |
| 2021-04-29 16:09:22 | <minoru_shiraeesh> | so, it's not always a matter of using a language with cheapest coders |
| 2021-04-29 16:09:23 | <maerwald> | tdammers: and not crashing, although your code is completely wrong |
| 2021-04-29 16:09:27 | <joncol> | I see the words Clojure and Haskell mentioned here. Anyone that did the switch from a Clojure job to a Haskell dito around here? Would be fun to hear something about the change in experience. |
| 2021-04-29 16:09:57 | <tdammers> | minoru_shiraeesh: I'm not saying the downsides are *always* show stoppers. Something like Lisp or Haskell can be your productivity superweapon that gives you an edge over the competition. But that only works if you know how to play its strength, and have a plan to deal with the downsides. |
| 2021-04-29 16:10:33 | × | conal quits (~conal@64.71.133.70) (Read error: Connection reset by peer) |
| 2021-04-29 16:10:35 | → | ep1ctetus joins (~epictetus@ip72-194-54-201.sb.sd.cox.net) |
| 2021-04-29 16:10:57 | → | conal joins (~conal@64.71.133.70) |
| 2021-04-29 16:11:03 | <kuribas> | tdammers: I find the quality of some clojure libraries quite bad actually. |
| 2021-04-29 16:11:08 | <tdammers> | joncol: technically, yes - one of the jobs I had before making the jump to "professional Haskeller" used clojure almost exclusively. in reality, however, I had been using Haskell for personal stuff long before that, so I really came to clojure from Haskell, and found the experience thoroughly underwhelming |
| 2021-04-29 16:11:23 | × | ddellac__ quits (ddellacost@gateway/vpn/mullvad/ddellacosta) (Ping timeout: 252 seconds) |
| 2021-04-29 16:11:26 | <kuribas> | tdammers: and the documentation is abysmal |
| 2021-04-29 16:12:01 | <maerwald> | tdammers: I worked on a pretty large clojurescript codebase and one day I deleted on bracket too many, the compiler didn't complain and the customer page would go blank |
| 2021-04-29 16:12:12 | <minoru_shiraeesh> | "otoh graham hutton, iirc, described lisp as one of his advantages over competition" or paul graham? not sure |
| 2021-04-29 16:12:14 | → | Guest87676 joins (~laudiacay@67.176.215.84) |
| 2021-04-29 16:12:18 | <maerwald> | I'm not sure how that passed the compiler, but it did |
All times are in UTC.