2025/06/13

Newest at the top

2025-06-13 14:54:58 +0200sabathan2(~sabathan@amarseille-159-1-12-107.w86-203.abo.wanadoo.fr)
2025-06-13 14:54:34 +0200sabathan2(~sabathan@amarseille-159-1-12-107.w86-203.abo.wanadoo.fr) (Remote host closed the connection)
2025-06-13 14:53:37 +0200target_i(~target_i@user/target-i/x-6023099) target_i
2025-06-13 14:50:06 +0200sabathan2(~sabathan@amarseille-159-1-12-107.w86-203.abo.wanadoo.fr)
2025-06-13 14:49:58 +0200 <tomsmeding> but there can be value in discussing how to make low-effort documentation as useful as possible, because realistically, a decent fraction of code on hackage is going to have low-effort documentation only :p
2025-06-13 14:49:40 +0200sabathan2(~sabathan@amarseille-159-1-12-107.w86-203.abo.wanadoo.fr) (Remote host closed the connection)
2025-06-13 14:49:19 +0200 <tomsmeding> ideally you have various different kinds of documentation for various different audiences and consumption styles, yes
2025-06-13 14:48:55 +0200AlexZenon(~alzenon@178.34.163.76)
2025-06-13 14:48:41 +0200 <kuribas`> tomsmeding: my take is that you need both, but also different forms of documentation, based on usecase, references, guides, overviews, getting-started, etc...
2025-06-13 14:47:07 +0200acidjnk(~acidjnk@p200300d6e71c4f374cafb09973b7f579.dip0.t-ipconnect.de) acidjnk
2025-06-13 14:46:13 +0200tmciver(~tim@syn-198-255-177-240.res.spectrum.com) tmciver
2025-06-13 14:46:12 +0200sajenim(~sajenim@user/sajenim) sajenim
2025-06-13 14:46:06 +0200mari-estel<3 doctests
2025-06-13 14:45:45 +0200 <tomsmeding> kuribas`: I guess, yes :p
2025-06-13 14:45:39 +0200tmciver(~tim@syn-198-255-177-240.res.spectrum.com) (Ping timeout: 268 seconds)
2025-06-13 14:45:25 +0200 <kuribas`> tomsmeding: you can reverse that: "it's rare documentation is simultaneously so precise and so general that without any types you can fully understand what a function does"
2025-06-13 14:44:17 +0200AlexZenon(~alzenon@178.34.163.76) (Ping timeout: 244 seconds)
2025-06-13 14:43:34 +0200mari-estel(~mari-este@user/mari-estel) mari-estel
2025-06-13 14:42:13 +0200sabathan2(~sabathan@amarseille-159-1-12-107.w86-203.abo.wanadoo.fr)
2025-06-13 14:41:51 +0200sabathan2(~sabathan@amarseille-159-1-12-107.w86-203.abo.wanadoo.fr) (Remote host closed the connection)
2025-06-13 14:41:42 +0200jespada(~jespada@r179-25-11-207.dialup.adsl.anteldata.net.uy) jespada
2025-06-13 14:41:18 +0200 <__monty__> Automatic edge-case covering examples à la Quickcheck could be useful. Highlighting unexpected behavior with memptys or such.
2025-06-13 14:38:32 +0200trickard_trickard
2025-06-13 14:38:28 +0200jespada(~jespada@r179-25-11-207.dialup.adsl.anteldata.net.uy) (Quit: My Mac has gone to sleep. ZZZzzz…)
2025-06-13 14:37:51 +0200Frostillicus(~Frostilli@pool-71-174-119-69.bstnma.fios.verizon.net)
2025-06-13 14:37:04 +0200 <tomsmeding> though more complicated libraries or functions don't really lend themselves to such insightful examples
2025-06-13 14:36:28 +0200 <tomsmeding> and actually, for something like Data.List.map, an example `map (*2) [1,2,3,4] == [2,4,6,8]` may actually be quicker to read, understand and generalise for a reader than some prose description of what 'map' does
2025-06-13 14:35:39 +0200Frostillicus(~Frostilli@pool-71-174-119-69.bstnma.fios.verizon.net) (Ping timeout: 260 seconds)
2025-06-13 14:35:20 +0200 <tomsmeding> right, this depends on what kind of library it is
2025-06-13 14:35:03 +0200 <Leary> True; if you're writing something that should be accessible to less experienced Haskellers then the translation from Haskell types to English would have its merits.
2025-06-13 14:33:57 +0200sabathan2(~sabathan@amarseille-159-1-12-107.w86-203.abo.wanadoo.fr)
2025-06-13 14:33:35 +0200sabathan2(~sabathan@amarseille-159-1-12-107.w86-203.abo.wanadoo.fr) (Remote host closed the connection)
2025-06-13 14:28:48 +0200bitdex(~bitdex@gateway/tor-sasl/bitdex) (Quit: = "")
2025-06-13 14:27:23 +0200wbooze(~inline@ip-005-146-196-202.um05.pools.vodafone-ip.de) Inline
2025-06-13 14:27:08 +0200 <tomsmeding> Leary: ^
2025-06-13 14:27:02 +0200 <tomsmeding> so what may seem like trivial documentation to you (and me) may actually help some readers
2025-06-13 14:26:41 +0200 <tomsmeding> for e.g. functions like Data.List.map
2025-06-13 14:26:27 +0200 <tomsmeding> counterpoint, I guess: a friend of mine who is learning haskell with a Java background significantly prefers reading documentation and looking at examples over studying types of library functions
2025-06-13 14:24:17 +0200xff0x(~xff0x@2405:6580:b080:900:a4eb:10d1:2f25:4397)
2025-06-13 14:23:48 +0200sabathan2(~sabathan@amarseille-159-1-12-107.w86-203.abo.wanadoo.fr)
2025-06-13 14:23:24 +0200sabathan2(~sabathan@amarseille-159-1-12-107.w86-203.abo.wanadoo.fr) (Remote host closed the connection)
2025-06-13 14:18:29 +0200sabathan2(~sabathan@amarseille-159-1-12-107.w86-203.abo.wanadoo.fr)
2025-06-13 14:18:08 +0200sabathan2(~sabathan@amarseille-159-1-12-107.w86-203.abo.wanadoo.fr) (Read error: Connection reset by peer)
2025-06-13 14:17:54 +0200raym(~ray@user/raym) (Ping timeout: 268 seconds)
2025-06-13 14:17:13 +0200constxd(~constxd@user/constxd) constxd
2025-06-13 14:16:43 +0200wbooze(~inline@ip-005-146-196-202.um05.pools.vodafone-ip.de) (Quit: Leaving)
2025-06-13 14:15:14 +0200 <tomsmeding> now, people yelling at you is something different, sure
2025-06-13 14:14:59 +0200 <tomsmeding> haddock yelling at you is just software yelling at you; it's annoying because it decreases the usefulness of warnings, but so be it
2025-06-13 14:14:32 +0200 <Leary> I also lean towards that just because I hate useless words, but I guess the real issue is the stigma of undocumented functions and haddock yelling at me. >.>
2025-06-13 14:13:38 +0200Frostillicus(~Frostilli@pool-71-174-119-69.bstnma.fios.verizon.net)