Luanti logo

IRC log for #luanti-dev, 2026-06-05

| Channels | #luanti-dev index | Today | | Google Search | Plaintext

All times shown according to UTC.

Time Nick Message
01:47 user333_ left #luanti-dev
01:48 user333_ joined #luanti-dev
04:00 MTDiscord joined #luanti-dev
07:27 MTDiscord1 joined #luanti-dev
17:13 turtleman joined #luanti-dev
17:20 mrcheese joined #luanti-dev
18:29 Krock Will merge #17221 #17223 #17225 #17233 #17234 in 20 minutes
18:29 ShadowBot https://github.com/luanti-org/luanti/issues/17221 -- Send detach update before player move by keilogic
18:29 ShadowBot https://github.com/luanti-org/luanti/issues/17223 -- IrrlichtMt: Fix MSVC and MinGW compiler warnings by SmallJoker
18:29 ShadowBot https://github.com/luanti-org/luanti/issues/17225 -- NetworkPacket: Unify 'getString' uses by SmallJoker
18:29 ShadowBot https://github.com/luanti-org/luanti/issues/17233 -- Correct `get_short_description()` documentation by OgelGames
18:29 ShadowBot https://github.com/luanti-org/luanti/issues/17234 -- Update stale FSF license addresses by dannyward630
18:53 Krock - done -
19:08 Krock sfan5: why is std::list evil? Too much memory overhead, or indirection?
19:09 Krock I assume it's only evil in cases where the element count is fixed
19:14 sfan5 bad cache locality / too much pointer chasing
19:15 Krock makes sense, yes.
19:16 sfan5 https://github.com/luanti-org/luanti/pull/17221 why are we merging undisclosed AI contributions?
19:16 sfan5 (I admit I didn't notice it either)
19:17 sfan5 no real contributor would say that they couldn't compile-test because cmake is not installed. they would simply go and install it
19:19 Krock I did realize it's mostly written by AI but found the solution acceptable. It is undisclosed, hence I'll issue a warning for them.
19:19 sfan5 https://github.com/luanti-org/luanti/pull/17230 they're doing it on purpose
19:19 sfan5 no reply here, only a -1
19:20 Krock where did we document the AI disclose requirement?
19:20 sfan5 https://github.com/luanti-org/luanti/blob/master/.github/PULL_REQUEST_TEMPLATE.md
19:20 sfan5 > If you have used an LLM/AI to help with code or assets, you must disclose this.
19:21 Krock ah. Neither in doc/ nor on the docs website. I see. Thanks.
19:22 sfan5 if I had to guess someone is running this account to deliberately target bug/feature bounties, making AI do the work (it's directly hooked up to github) and collecting the cash
19:26 Krock https://github.com/UnsafeLabs/Bounty-Hunters
19:27 Krock This repository is a honey pot to blacklist AI contributors on the real project. Might be interesting for inspiration.
19:28 sfan5 huh
19:29 Krock (In case we're ever overrun by bounty hunter AI)
19:29 sfan5 anyway time for more productive things
19:30 [MatrxMT] <Zughy> I'm not sure what I should do here
19:30 [MatrxMT] <Zughy> Should I send the money or, due to their behaviour, keep it for myself?
19:30 Krock the implementation was rather trivial so it does not matter much whether an AI was consulted or not
19:31 sfan5 that's up to you unfortunately.
19:32 [MatrxMT] <Zughy> "Consulted" seems a bit limited here though. Have I misunderstood or it's quite likely they just prompted an AI without basically doing nothing?
19:32 [MatrxMT] <Zughy> *anything
19:33 Krock I'd argue that they did implement the change requests well. Though I don't know whether this was all done by AI too.
19:34 sfan5 personally I get the impression that the AI is literally tasked with directly interacting with github. there isn't even a human in the loop copy-pasting responses or doing other stuff. this is what bothers me the most.
20:10 crazylad joined #luanti-dev
20:10 crazylad joined #luanti-dev
20:14 YuGiOhJCJ joined #luanti-dev
21:42 pgimeno I would like to think that the -1 was a human though
21:52 MTDiscord <dragonwrangler1> That is actually hilarious
22:34 panwolfram joined #luanti-dev

| Channels | #luanti-dev index | Today | | Google Search | Plaintext