Newest at the top
| 2025-05-16 20:45:49 +0200 | <__monty__> | wbrawner: Do you actually care about transitive dependencies or do you care about how long/much space it takes to install them? |
| 2025-05-16 20:45:35 +0200 | <sclv> | wbrawner: some of that is imho “javascript disease” where people chop up libraries too small. i know what induces this in js (versioning problems) but idk in cargo |
| 2025-05-16 20:44:33 +0200 | <sclv> | two other things are “batteries included” made more sense when some batteries were otherwise hard to install (since fixed), when there were fewer and more stable libs, and when the notion of “batteries” covered far less surface area |
| 2025-05-16 20:44:29 +0200 | j1n37 | (~j1n37@user/j1n37) (Ping timeout: 252 seconds) |
| 2025-05-16 20:44:16 +0200 | <wbrawner> | I think trying to ship everything in stdlib is untenable but I really dislike working with npm/cargo that pull in hundreds of transitive dependencies when I ask for one |
| 2025-05-16 20:43:35 +0200 | haskellbridge | (~hackager@syn-096-028-227-029.res.spectrum.com) (Remote host closed the connection) |
| 2025-05-16 20:43:26 +0200 | j1n37- | (~j1n37@user/j1n37) j1n37 |
| 2025-05-16 20:42:49 +0200 | <sclv> | a big step |
| 2025-05-16 20:42:40 +0200 | <sclv> | the best part of platform was it was the first time we had a standard installer story across multiple platforms. ghcup, which replaced that, is leagues better now, but it was still |
| 2025-05-16 20:41:13 +0200 | <monochrom> | For example, with a compiler with insane cross-module even cross-package inlining, batteries included backfires badly (cf "cabal hell" back then). |
| 2025-05-16 20:38:56 +0200 | <__monty__> | I actually like JavaScript radically tiny libraries model (at least in theory). Makes it realistic for things to achieve perfection and not require unnecessary version constraint churn. |
| 2025-05-16 20:38:17 +0200 | <monochrom> | (Fortunately, we later found out that while those worked for Python---good for them---they didn't work for Haskell.) |
| 2025-05-16 20:36:49 +0200 | econo_ | (uid147250@id-147250.tinside.irccloud.com) |
| 2025-05-16 20:36:15 +0200 | <monochrom> | My theory is that there was a time our opinion leaders envied Python popularity and decided we should parrot them, for example "batteries included", for example making snake-oil claims like "increased productivity" with no data. |
| 2025-05-16 20:33:17 +0200 | <monochrom> | Plus outside GHC there was also "OMG Haskell Platform is the best idea ever" and "OMG Haskell Platform is the worst idea ever". (Haskell Platform was an effort to be a secondary "standard" library to include everything and batteries and the kitchen sink.) |
| 2025-05-16 20:32:19 +0200 | dibblego | (~dibblego@haskell/developer/dibblego) dibblego |
| 2025-05-16 20:32:19 +0200 | dibblego | (~dibblego@116.255.1.119) (Changing host) |
| 2025-05-16 20:32:19 +0200 | dibblego | (~dibblego@116.255.1.119) |
| 2025-05-16 20:31:18 +0200 | <monochrom> | Oh, historical GHC has gone through the mood swing of "OMG GHC must come with everything and batteries and the kitchen sink" and "OMG GHC must come with as little as possible". Twice. |
| 2025-05-16 20:30:06 +0200 | peterbecich | (~Thunderbi@syn-047-229-123-186.res.spectrum.com) peterbecich |
| 2025-05-16 20:29:13 +0200 | monochrom | reaches for libraries that come with GHC for everything. :) |
| 2025-05-16 20:29:04 +0200 | ljdarj | (~Thunderbi@user/ljdarj) ljdarj |
| 2025-05-16 20:27:42 +0200 | sabathan2 | (~sabathan@amarseille-159-1-12-107.w86-203.abo.wanadoo.fr) |
| 2025-05-16 20:27:21 +0200 | sabathan2 | (~sabathan@amarseille-159-1-12-107.w86-203.abo.wanadoo.fr) (Remote host closed the connection) |
| 2025-05-16 20:25:37 +0200 | sprotte24 | (~sprotte24@p200300d16f1bca002ca617d70fcb63bb.dip0.t-ipconnect.de) |
| 2025-05-16 20:22:21 +0200 | sabathan2 | (~sabathan@amarseille-159-1-12-107.w86-203.abo.wanadoo.fr) |
| 2025-05-16 20:22:20 +0200 | <wbrawner> | Personally I prefer the Go approach of only using third-party stuff when I really need it but old habits die hard in unfamiliar settings |
| 2025-05-16 20:21:57 +0200 | sabathan2 | (~sabathan@amarseille-159-1-12-107.w86-203.abo.wanadoo.fr) (Remote host closed the connection) |
| 2025-05-16 20:21:41 +0200 | <wbrawner> | I'm coming from Android dev where we reach for libraries for pretty much anything, and in my free time I often write Go where I can just use things from the stdlib and almost never reach for third-party dependencies |
| 2025-05-16 20:21:04 +0200 | machinedgod | (~machinedg@d108-173-18-100.abhsia.telus.net) machinedgod |
| 2025-05-16 20:21:00 +0200 | <wbrawner> | as an aside: is http-client commonly used? |
| 2025-05-16 20:20:42 +0200 | <wbrawner> | ski: that did the trick, thanks for the help! |
| 2025-05-16 20:20:04 +0200 | <monochrom> | Hrm, I made the wrong bet. :) |
| 2025-05-16 20:16:50 +0200 | <EvanR> | if you tried reading in an unknown amount using strict bytestring you'd have to "reallocate" when your guess at total size keeps being wrong |
| 2025-05-16 20:16:23 +0200 | <EvanR> | I see. By using lazy I/O you can allocate chunks in order and they're already chained together |
| 2025-05-16 20:13:29 +0200 | tromp | (~textual@2001:1c00:3487:1b00:a44a:50e6:3df5:3b66) (Quit: My iMac has gone to sleep. ZZZzzz…) |
| 2025-05-16 20:13:26 +0200 | tzh | (~tzh@c-76-115-131-146.hsd1.or.comcast.net) |
| 2025-05-16 20:13:16 +0200 | vanishingideal | (~vanishing@user/vanishingideal) (Ping timeout: 276 seconds) |
| 2025-05-16 20:12:05 +0200 | acidjnk | (~acidjnk@p200300d6e71c4f410c8650aa1a5c1c11.dip0.t-ipconnect.de) acidjnk |
| 2025-05-16 20:11:52 +0200 | dibblego | (~dibblego@haskell/developer/dibblego) (Ping timeout: 252 seconds) |
| 2025-05-16 20:09:39 +0200 | <ski> | but since wbrawner already had a `Response a' here, no need to wrap in `Right' and pass to `getResponseBody', better to just extract the `responseBody' field |
| 2025-05-16 20:09:04 +0200 | <EvanR> | classy |
| 2025-05-16 20:08:46 +0200 | <ski> | (throwing exception in case of `Left', apparently) |
| 2025-05-16 20:08:25 +0200 | <ski> | where `Result = Either Whatever' |
| 2025-05-16 20:08:21 +0200 | sabathan2 | (~sabathan@amarseille-159-1-12-107.w86-203.abo.wanadoo.fr) |
| 2025-05-16 20:08:18 +0200 | acarrico | (~acarrico@pppoe-209-99-221-107.greenmountainaccess.net) (Quit: Leaving.) |
| 2025-05-16 20:08:12 +0200 | <ski> | EvanR : it is `IO', but it takes `Result (Response a)', not `Response a' |
| 2025-05-16 20:08:03 +0200 | <EvanR> | hax |
| 2025-05-16 20:07:52 +0200 | sabathan2 | (~sabathan@amarseille-159-1-12-107.w86-203.abo.wanadoo.fr) (Remote host closed the connection) |
| 2025-05-16 20:05:45 +0200 | <c_wraith> | length is sufficient |