Time Nick Message 03:19 MTDiscord Would we want a vulkan render backend in minetest-rust? 03:19 MTDiscord I just want some sort of approval, even "eh yeah I guess" 15:06 MTDiscord you might be interested in: https://arewegameyet.rs/ecosystem/3drendering/ 15:08 celeron55 you need to select something for graphics and input and windowing and whatnot that's portable, robust and appropriate well into the future 15:08 celeron55 and also works right now, and also works on older hardware 15:10 celeron55 (if you have to drop one of those aspects, then drop older hardware) 15:11 celeron55 (you aren't making anything that the regular joe would even know exists until after many years have passed. you need to be very, very forward looking) 16:00 MinetestBot 02[git] 04rubenwardy -> 03minetest/minetest_game: Update README.md for 5.8 13b9c0894 https://github.com/minetest/minetest_game/commit/b9c089474cc343b37f90cb8cefc4b5553af2e9ea (152023-12-17T15:57:57Z) 19:46 MinetestBot 02[git] 04JosiahWI -> 03minetest/minetest: Extract Game::drawScene from Game::updateFrame 137162b53 https://github.com/minetest/minetest/commit/7162b536eb43c70f41cac1403323e128e66cef19 (152023-12-17T19:44:45Z) 19:49 MinetestBot 02[git] 04grorp -> 03minetest/minetest: Add `touch_controls` boolean to `get_player_window_information()` (#1… 1391ba024 https://github.com/minetest/minetest/commit/91ba02449bd6649f15730520c12d8d2d8031b6b0 (152023-12-17T19:47:07Z) 19:49 MinetestBot 02[git] 04sfan5 -> 03minetest/minetest: MinGW toolchain refresh 135d3e830 https://github.com/minetest/minetest/commit/5d3e83017679317c27fe02b7087effd9d67f79cc (152023-12-17T19:47:26Z) 23:08 MTDiscord c55 I'm gonna do something really, very dumb 23:09 MTDiscord I'm gonna remake the current progress for minetest-rust in go so you can see side by side in the two languages to see which flavor you like better 23:10 MTDiscord Go seems like a decent idea tbh 23:10 MTDiscord Go's garbage collector is actually advanced enough to the point where it can guarantee that pauses aren't too bad 23:11 rubenwardy I was about to say that you'd be more productive if you focused on something rather than rewriting things into every language under the sun... but I'm not one to talk 23:11 MTDiscord that is still a valid point 23:11 MTDiscord however, under the premise that we're gonna port stuff, i think a go port would at least initially be much less work than a rust port :P 23:11 MTDiscord I'd literally rewrite the entire project into pure luajit/bun/deno but that would simply make everyone equally angry and it would be DOA 23:12 MTDiscord the thing is: Minetest's code base is very massive 23:12 MTDiscord 200 kloc of finest C++ iirc 23:12 MTDiscord if you consider Irrlicht even more 23:12 MTDiscord It's massive because we're jumping through hoops to have elements duct taped on 23:12 MTDiscord now fair, let's say half the code is c++ bullshit 23:13 MTDiscord and let's say your able to port 1kloc a day 23:13 MTDiscord you'd still have to work 100 days full time 23:13 MTDiscord just for feature parity 23:13 MTDiscord I do not want to do a direct port lmao, if I do a direct port we end up with an over complicated code base with memory leaks in another language 23:14 MTDiscord so you want to write another engine, let's call it crafter? 23:14 MTDiscord I want feature parity, even if that initial goal is 0.3.1 23:14 MTDiscord protocol-level? lua api level? 23:15 MTDiscord Surface level, 0.3.1 didn't even have a lua implementation, but that can be the goal from the base, 0.3.1 but it uses a modified implementation of minetest's lua api 23:15 MTDiscord so should your client be able to talk to a Minetest server and vice-versa? 23:15 MTDiscord Absolutely not, this will be the hard reboot 23:15 MTDiscord so.. what does this have to do with Minetest then? 23:18 MTDiscord Just like windows 2000 transferred to NT architecture and still felt like windows 98, minetest-next will have the same iteration 23:22 MTDiscord But now, I have lost the drive to do the side by side and I'll just keep writing the rust base 23:25 MTDiscord don't get me wrong, i'm not saying this is a bad idea, but i think it will have to stand on its own, just like Qub3d (which unfortunately failed) 23:25 MTDiscord not every libre voxel game platform / engine has to be tied to minetest 23:35 MTDiscord I think the idea is much like making (a multiple year supply of) free cookies for everyone for free. - If you make a much better Minetest (which is probably quite doable, considering its state is literally that of Keith, the c++ mascot), everyone will LOVE IT! - On the flip side, you will have invested years of your life, without much "real" return. People will know you built it and respect you for it, that's basically it. - Thus, 23:35 MTDiscord either you REALLY LOVE THE IDEA that much that you are willing to invest massively, for the sake of it, OR you should ABANDON IT, because if you don't really love it you will do so anyway sooner or later, for good. 23:37 MTDiscord I think you guys are thinking about the premise of minetest-rust a bit too hard 23:40 MTDiscord One question remains: @Sr. Christmas Researcher (MD), do you take your Minetest project for your lawful project, to have and to hold, from this day forward, for better, for worse, for richer, for poorer, in sickness and in health, until death or file loss do you part? 23:41 MTDiscord Hmm, well it seems I still have 220 other projects in my git repos, so I think we can rule out file loss 23:47 MTDiscord Maybe they could lock me into the lang by making me the core dev of the minetest/minetest-rust repo 23:48 MTDiscord https://tenor.com/view/dog-eyebrow-funny-eyebrow-raise-gif-13185653 23:48 MTDiscord The point is that such idealistic mega projects are just like marriages, they are pursued for themselves, they stretch for years, go through good and bad times, usually there are near breakups due to attractive side projects... So you really have to ask yourself the question: Do you? Really? 23:49 MTDiscord Because just like a failed marriage is no good, a failed megaproject is no good either. 23:50 MTDiscord If you think about it like that, we probably wouldn't even have linux, or windows, or like anything. "Too much work" 23:50 MTDiscord You take it step by step 23:51 MTDiscord Not to say you don't think forward, but you do not think about it as a whole, you must take the unix approach and treate each of your modules as a program 23:51 MTDiscord Just like a successful marriage. 23:51 MTDiscord ~e 23:52 MTDiscord ? 23:52 MTDiscord I called the destructor on e 23:52 MTDiscord And why? 23:52 MTDiscord Because when you edit your message it doesn't get reflected into the irc 23:53 MTDiscord np. 23:54 MTDiscord I need moar caffeine