| Time |
Nick |
Message |
| 00:02 |
|
appguru joined #minetest-dev |
| 01:14 |
|
fluxionary joined #minetest-dev |
| 01:20 |
|
fluxionary joined #minetest-dev |
| 02:20 |
|
Desour joined #minetest-dev |
| 02:31 |
|
diceLibrarian joined #minetest-dev |
| 03:33 |
|
diceLibrarian joined #minetest-dev |
| 05:00 |
|
MTDiscord joined #minetest-dev |
| 05:17 |
|
[MTMatrix] joined #minetest-dev |
| 05:17 |
|
[MTMatrix] joined #minetest-dev |
| 05:19 |
|
[MTMatrix] joined #minetest-dev |
| 05:22 |
|
[MTMatrix] joined #minetest-dev |
| 05:24 |
|
[MTMatrix] joined #minetest-dev |
| 05:27 |
|
[MTMatrix] joined #minetest-dev |
| 05:27 |
|
[MTMatrix] joined #minetest-dev |
| 05:30 |
|
[MTMatrix] joined #minetest-dev |
| 05:30 |
|
[MTMatrix] joined #minetest-dev |
| 05:30 |
|
[MTMatrix] joined #minetest-dev |
| 05:37 |
|
[MTMatrix] joined #minetest-dev |
| 05:37 |
|
[MTMatrix] joined #minetest-dev |
| 05:38 |
|
[MTMatrix] joined #minetest-dev |
| 05:40 |
|
[MTMatrix] joined #minetest-dev |
| 05:43 |
|
[MTMatrix] joined #minetest-dev |
| 07:29 |
|
YuGiOhJCJ joined #minetest-dev |
| 07:38 |
|
calcul0n_ joined #minetest-dev |
| 10:11 |
sfan5 |
don't worry |
| 13:31 |
|
json joined #minetest-dev |
| 13:32 |
|
json68 joined #minetest-dev |
| 13:42 |
|
json joined #minetest-dev |
| 13:43 |
|
fluxionary joined #minetest-dev |
| 13:43 |
|
imi joined #minetest-dev |
| 19:18 |
rubenwardy |
Fedora are unsure about removing Minetest Game as it may violate Fedora policy https://bodhi.fedoraproject.org/updates/FEDORA-2023-877949f206 |
| 19:21 |
|
Desour joined #minetest-dev |
| 19:23 |
rubenwardy |
ah nevermind, already resolved |
| 19:23 |
rubenwardy |
It's kind of a pain to navigate all the different distros' issue trackers to see what they are doing |
| 19:24 |
|
fkgkzxcv joined #minetest-dev |
| 19:28 |
rubenwardy |
arch is still including MTG |
| 19:28 |
rubenwardy |
debian is still on 5.6.0 |
| 19:41 |
MTDiscord |
<warr1024> > It's better to have Minetest with Minetest Game than not have it at all ... controversial. It's ironic to see FOSS projects complaining about us removing the feature (i.e. bundling) that Microsoft was convicted of in EU and was forced to implement anti-trust features into their OS. |
| 19:59 |
rubenwardy |
see https://gist.github.com/rubenwardy/eb481b51a5e0574e5219b6967752bdb0 |
| 19:59 |
rubenwardy |
!title |
| 19:59 |
ShadowBot |
Distro Minetest Game status · GitHub |
| 20:09 |
MTDiscord |
<warr1024> You probably shouldn't be using ✔️ for the status for Fedora if the removal was only partially successful. |
| 20:10 |
MTDiscord |
<warr1024> What are the "debian specific customizations" to MTG that they're talking about? |
| 20:11 |
rubenwardy |
no idea |
| 20:13 |
MTDiscord |
<warr1024> I can't even imagine what kind of customizations they'd feel the need to have. Like, iceweaselifying it or something? |
| 20:13 |
sfan5 |
where is that mentioned? |
| 20:13 |
sfan5 |
ah |
| 20:14 |
MTDiscord |
<warr1024> They ship a C++ compiler but they don't have ways of applying Debian-specific customizations to C++ code that people download from the internet that would get compiled by it. Similarly, wget and curl are available but they have no way to ensure that whatever is downloaded with them is trademark-sanitized. |
| 20:14 |
rubenwardy |
can't see any MTG changes here https://sources.debian.org/src/minetest/5.6.1%2Bdfsg%2B~1.9.0mt8%2Bdfsg-4/debian/ |
| 20:15 |
MTDiscord |
<warr1024> This backlash is all kinda stupid, but it also serves to highlight the very reason why we removed MTG from MT, because of this misconception that MTE is just an engine for playing MTG. |
| 20:15 |
rubenwardy |
anyone know what issue tracker to use to contact the arch maintainers? |
| 20:16 |
ROllerozxa |
they use their own gitlab instance you need to request an account for |
| 20:16 |
MTDiscord |
<wsor4035> there gitlab, but you have to request an account |
| 20:16 |
rubenwardy |
oh lame |
| 20:16 |
ROllerozxa |
alternatively email the packager I guess |
| 20:16 |
MTDiscord |
<wsor4035> https://gitlab.archlinux.org/archlinux/packaging/packages/minetest |
| 20:16 |
rubenwardy |
I tried using Arch SSO with my AUR login but that didn't work |
| 20:16 |
rubenwardy |
guess it's not so single |
| 20:17 |
rubenwardy |
anyone have an account there? |
| 20:17 |
MTDiscord |
<warr1024> I keep seeing "Flatpak: removed" and thinking it means "flatpak just removed the entire minetest package rather than dealing with this" instead of "successfully removed MTG from MT" |
| 20:18 |
rubenwardy |
do any major distros remove ContentDB support? ie: of those listed there |
| 20:18 |
sfan5 |
I'm not aware of any |
| 20:28 |
[MTMatrix] |
<fgaz> rubenwardy: by the way here's a more complete list of distros that package minetest https://repology.org/project/minetest/versions |
| 20:34 |
MTDiscord |
<warr1024> That has a lot of old versions of MT being shipped with old versions of distros, which kind of makes sense to ignore and makes that list hard to read. |
| 20:35 |
rubenwardy |
updated |
| 20:38 |
MTDiscord |
<warr1024> Heh, you've got FreeBSD but missing OpenBSD and NetBSD. I don't know about Net but Open is outdated at 5.6.1, looks like. |
| 20:38 |
MTDiscord |
<warr1024> er, wait, 5.6.0 |
| 20:38 |
rubenwardy |
neither of those are on the downloads page |
| 20:40 |
[MTMatrix] |
<fgaz> warr1024: yeah it's a bit difficult to read for widely packaged stuff: https://github.com/repology/repology-webapp/issues/62 https://github.com/repology/repology-webapp/issues/126 |
| 20:43 |
rubenwardy |
anyway, done with this for now |
| 20:43 |
rubenwardy |
thanks for the link fgas |
| 20:43 |
rubenwardy |
-s+z |
| 20:43 |
MTDiscord |
<warr1024> "neither of those are on the downloads page" ... lol, what does it take to actually get on there then? |
| 20:44 |
sfan5 |
half on topic: should I be including MTG in my "daily" builds? |
| 20:46 |
MTDiscord |
<warr1024> only if it's "builds" of MTG... |
| 20:48 |
sfan5 |
I guess no then |
| 20:49 |
MTDiscord |
<warr1024> I think we should be consistent about not shipping MTG with MT unless we're specifically making a separate "bundle product" or something, and those should only be necessary for special cases like deployment somewhere without network access. |
| 20:49 |
sfan5 |
well I used to bundle voxelgarden with them |
| 20:52 |
MTDiscord |
<warr1024> It's not bad to offer bundles but it would be best to make it clear that that's not the "standard" MT experience, which is now "pick a game first". See NodeCore on Google Play, which is a bundle that Roller made, but it's also pretty clearly a separate "brand" from MT. |
| 20:53 |
MTDiscord |
<warr1024> Technically you can still do everything with Android NodeCore that you can with MT (though I think it has different defaults so you have to explicitly enable stuff) but the rebrand helps avoid confusing players into thinking it's representative of the whole MT ecosystem. |
| 20:53 |
MTDiscord |
<luatic> Call it "Minetest + Bloat" and do a fisheye transform on the Minetest logo |
| 20:53 |
MTDiscord |
<warr1024> whatever foats your bloat. |
| 20:53 |
MTDiscord |
<wsor4035> proposal to rename minetest game to bloat game? 👀 |
| 20:54 |
MTDiscord |
<wsor4035> luatic: but then you would have to include dreambuilder |
| 20:55 |
MTDiscord |
<warr1024> The irony of considering MTG "bloat" even though it's one of the more feature-baren games ... but I guess it's "bring your own bloat game" |
| 20:55 |
MTDiscord |
<wsor4035> changing gears slightly, looking at https://github.com/minetest/minetest/issues/14190, wondering if it would make sense to just go all in on luajit? |
| 20:57 |
MTDiscord |
<warr1024> Do we still have some platforms that can't do LuaJIT? IIRC Apple Silicon was a problem at one point... |
| 20:57 |
celeron55 |
i'd say feel free to bundle MTG or any game you like, BUT the end result then has to be called a bundle of some kind. brand it however you like, but not just "minetest". bundles have their uses, there's nothing inherently wrong in them |
| 20:57 |
celeron55 |
with them* |
| 20:58 |
MTDiscord |
<wsor4035> warr1024: https://love2d.org/forums/viewtopic.php?t=94760 would imply yes, but use newer versions |
| 20:58 |
MTDiscord |
<paradust> emscripten/wasm can't do luajit, but i can always bundle regular lua as long as it is compatible |
| 20:59 |
celeron55 |
even just "sfan5's daily build" is a distinctive brand compared to "here's minetest.zip" |
| 20:59 |
sfan5 |
to be clear I was referring to https://forum.minetest.net/viewtopic.php?p=18903#p18903 IMO there's close to zero chance that it is mistaken |
| 20:59 |
sfan5 |
+for an official thing |
| 20:59 |
MTDiscord |
<wsor4035> paradust, can it not do luajit at all (including with jit off?) |
| 21:00 |
MTDiscord |
<warr1024> Yeah, it seems to me like the only things that we lose going all in on LuaJIT are (1) the possibility that some platforms transiently don't support it, i.e. we have to wait for JIT ports to a platform before we can do MT ports, and (2) we may lose any hope of using a different JIT, e.g. in the case where WASM outperforms LuaJIT and there are ways to run Lua on it or something. |
| 21:01 |
pgimeno |
ISTR having read the LJ interpreter was written in assembler. Maybe I'm wrong. |
| 21:01 |
MTDiscord |
<paradust> wsor: never tested it |
| 21:01 |
sfan5 |
wsor4035: increasing library dependency always has negative effects regarding portability or debuggability and there's no strong case for it |
| 21:02 |
celeron55 |
sfan5: if you see confusion about the builds regarding to what you bundle in them, then consider how to make it clearer. otherwise, it's not a big deal |
| 21:02 |
MTDiscord |
<warr1024> I'm not saying that it's not okay for sfan5 to define his builds as an MT+MTG bundle, but I think not including MTG in the bundle would be a quality boost that shouldn't just be ignored. At the very least, MTG's change schedule is not going to match up with MTE's so it's not very efficient to always download them both each time. |
| 21:04 |
MTDiscord |
<wsor4035> sfan5: more so was suggesting it because its already an optional dependency that pretty popular/commonly used. but given the issues with platform support, seems not a good option at this time |
| 21:05 |
sfan5 |
regarding platform support see for example this: https://github.com/LuaJIT/LuaJIT/issues/628 |
| 21:05 |
sfan5 |
telling people that "oh we have to wait for <project with opinionated maintainer and a bus factor of 1> until you get to play MT on <shiny new thing>" is far from a good look |
| 21:06 |
MTDiscord |
<warr1024> So requiring LuaJIT will give us more risk but less RISC... |
| 21:17 |
MTDiscord |
<warr1024> Looking at that homebrew link in ruben's list, it seems kind of funny that it didn't seem to occur to them to stop and ask before going from 5.7.0 engine and 5.7.0 game to 5.8.0 engine and 5.7.0 game... |
| 21:25 |
MTDiscord |
<paradust> Does the engine maintain backwards compatibility? |
| 21:28 |
MTDiscord |
<jordan4ibanez> sometimes |
| 21:29 |
|
grorp joined #minetest-dev |
| 21:32 |
grorp |
there are sometimes breakages, and some stuff is broken after deprecation periods, but in general it does |
| 21:40 |
MTDiscord |
<warr1024> We don't require that the 5.8 version of MT match up with the 5.8 version of MTG ... but just the fact that they would have before but suddenly don't now should still give somebody pause and make them think "what's happening here? maybe I should investigate a bit further and find out why". |
| 21:42 |
MTDiscord |
<jordan4ibanez> sfan5 I'm pretty sure if it were up to mike pall everyone would be using windows 95 |
| 22:32 |
[MTMatrix] |
<fgaz> rubenwardy: NixOS: removed https://github.com/NixOS/nixpkgs/pull/273557 |
| 23:07 |
|
Desour joined #minetest-dev |
| 23:32 |
|
panwolfram joined #minetest-dev |
