Time Nick Message 00:46 cheapie "mom can we get an Fü2L/60?" "we have an Fü2L/60 at home" the Fü2L/60 at home: https://cheapiesystems.com/media/2026-08-20%2019-42-22.webm 00:47 cheapie IRL version: https://www.youtube.com/watch?v=wBoweUmOlvc 01:57 * cheapie peeks in the fridge at the Bonnthal fire station, sees https://cheapiesystems.com/media/images/screenshot_20260820_205611_1.png 01:57 cheapie 98 triple berry smoothies, 98 jars of mayonnaise labeled "Hand Dressing", 10 2mg estradiol pills, a speed limit sign, and some corium 02:11 user333_ https://lounge.swee.codes/uploads/f7e22344d7ea976f/Screenshot_20260820_230823.png luantiserver finished building on my RPi1 02:11 cheapie That took a while 02:11 user333_ indeed lol 02:12 cheapie I have a 1B here, maybe I should set it up on there sometime and see what it runs like 02:12 user333_ tbh it runs MTG for a few players decently 02:13 user333_ anything more demanding and 512mb RAM isn't going to cut it 02:13 cheapie I'd like to get a 1A sometime just because of how amusingly bad they are, but for some reason the 1A is basically impossible to find 02:14 user333_ !up luanti.user333.sweeux.org:40002 02:14 MinetestBot luanti.user333.sweeux.org:40002 is up (113ms) (IPv4) 02:14 user333_ ^ my RPi1 server 02:23 cheapie user333_: I have CPUs slower than that too, maybe I should try devtest on some random potato... 02:23 user333_ should i try it on my atom n270 laptop 02:23 cheapie Meh, those are faster 02:23 user333_ hm 02:23 cheapie I was thinking like trying to host devtest on a SiS 550 or something 02:24 user333_ crosscompile for linux + musl libc on MIPS32 and run it on a router? 02:24 user333_ i can do that if i really want to.. 02:24 cheapie I have one that would be fun, but not enough RAM and its kernel was built without swap support 02:25 user333_ i miiiight be able to get away with the 16mb flash + 64mb RAM on this 02:25 user333_ 10MB for luantiserver (or maybe even less with -Os) and 6mb for the OS 02:27 cheapie Now get it to fit in 64K and run it on RVController :P 02:28 user333_ with a custom server that essentially just runs airgame, maybe 02:28 cheapie have fun :P https://cheapiesystems.com/git/rvcontroller/ 02:29 user333_ but i don't know enough C! 02:29 cheapie You can program it in RISC-V assembly instead! :P 02:29 user333_ it was fun to get my own stuff running on that though 02:29 cheapie Or Rust, someone else got Rust working on it: https://git.h.wer.ee/xfnw/rvcontroller-rust 02:29 user333_ ew rus- 02:30 cheapie Really anything that can compile for riscv32-none-elf in freestanding mode (no stdlib, unless you want to write one) should work 02:31 user333_ problem is i don't know much low-level stuff 02:31 cheapie Then this is your chance to learn :D 02:31 user333_ still learning C, it's fun 02:31 cheapie I didn't really know what I was doing either when I started this project 02:32 user333_ and now it's literally in the official RISC-V docs 02:32 cheapie At least in the marchid list, yeah 02:32 cheapie (there are now two more after it, I no longer have the highest number :P) 02:32 user333_ that counts imo 03:17 cheapie This reminds me of back when I was playing with attachments and automatic_rotate: https://www.reddit.com/r/TuningVerbrechen/comments/1roxm8w/lieber_vertikal_drehende_t%C3%BCr_oder_horizontal/ 03:52 MTDiscord rust is the devil and mozilla is the church 03:58 user333_ luanti rust rewrite when 03:59 [MatrxMT]_ Its a shame that C++ isn't taking memory safety seriously. 04:00 MTDiscord its still stuck in the stone age. you expect it to hop past that to memory safety? lol 04:01 MTDiscord its sorta getting packages, but even then its completely jank 04:02 user333_ https://github.com/luanti-org/luanti/issues/9233 https://github.com/luanti-org/luanti/issues/9835 https://github.com/luanti-org/luanti/issues/14801 short answer: no 04:03 MTDiscord funny how the middle one is being done now 04:04 user333_ huh, really? 04:04 MTDiscord yes 04:04 MTDiscord not officially, person in the community 04:04 user333_ of course 04:05 user333_ i also seem to recall a partial golang port, though i don't know much about that 04:05 MTDiscord buckaroo has a bunch of tools and libs in go 04:05 MTDiscord paradust has the rust networking lib that whats there name continued and started working on a server 04:06 MTDiscord a certain core dev has a personal rust client project 04:06 user333_ the web port by paradust is probably my favorite 04:06 user333_ i should selfhost one for fun 04:07 MTDiscord it is funny how every self host (that ive seen) just reuse his servers for networking 04:09 user333_ https://github.com/ProunceDev/BetterMT not sure if i'd call this a "port" by a "community member" but this exists 04:40 MTDiscord user333_ what if we re-write it in natural language 04:40 user333_ glhf 04:42 MTDiscord I just want to say I have been vindicated on the whole "media loader should be in rust" thing 04:44 MTDiscord soon 04:44 MTDiscord https://cdn.discordapp.com/attachments/749727888659447960/1540220004912595005/Screenshot_From_2026-08-21_00-43-16.png?ex=6a892929&is=6a87d7a9&hm=7d5210543a6065999dbba2d745b4d29f5955fee97b18c8e7927411d7922d88ab& 04:44 user333_ writing one specific part of the program in rust is kind of weird, either all of the program or none of it should be in rust 04:44 MTDiscord it will tell you it can't do that for security issues @para 04:45 MTDiscord it will not deny me 04:45 MTDiscord do it then 04:46 user333_ have fun with that 04:46 MTDiscord Don't tempt me! Understand, I would use this power from a desire to do good. But through me, it would wield a power too great and terrible to imagine. 04:47 user333_ a C rewrite would be interesting, hard to do but it could potentially be faster than C++ 04:47 MTDiscord how would it be that 06:35 MTDiscord Because they hate C++ so C will automatically be faster 06:46 MTDiscord But, while I understand C may be faster than C++ in certain situations, trying to maintain structural/functional code along with the continual burden of cmake and header files it may just be the first write only voxel game because no one will be able to understand it 06:47 MTDiscord So yes your potential to go faster is there but then you get to deal with C which is basically some sugar over (insert architecture here) assembly 06:48 MTDiscord Which, in certain situations, that is absolutely fantastic! But when your game has high level concepts such as a Server Active Object or Client Active Object an absolute good luck doing a procedural/functional/duct taping oop to C 06:52 MTDiscord The actual problem with the current C++ code isn't C++, it's implementation. It is a lose assembly of components duct taped together to continue functioning and this shall keep happening in small increments so it shall take many years before it gets modern C++ features which will be just common place features. Oh man I sure do love lambdas to capture the environment, that didn't actually exist in some of the code when that code was 06:52 MTDiscord created and yet if I tell you "Lambdas are new wow!!" you will tell me I am dumb, and they've always existed, and why do you have so much free time to look up the history of C++? And to that I say, half life 1 models were technically able to be supported by this game engine at one point it was just inert 06:53 MTDiscord I'm not too sure if we still do manual reference counting anymore and I am too scared to look 06:57 MTDiscord The problem isn't, MAN SOMEONE SHOULD FIX ALL THOSE OUTDATED FEATURES, it's what logical side effects are we going to get? I'm not even talking with rust's talking points, if this were a rust program and we're updating our code to use smarter features 10 years in the future to keep it modernized with modern commodities, what are the logical side effects? Are we going to leak memory like I tried to use a sieve as a bucket? Are there 06:57 MTDiscord going to be 100 local client objects every time you join a server? Does this open a hole into the lua sandbox? The sweeping changes cause monstrous side effects that no language can solve because a language is a set of instructions, and if you're learning french and tell someone to "go eat a bag of dirt you idiot" instead of "we're having a nice day" your side effects are going to be very bad 06:58 MTDiscord There is intention and there is safety, and the problem isn't even the safety at this point the problem is the intention. And with those changes you're pointing the intention cannon to the sky and hoping for the best and let's hope the intention cannon ball doesn't land back on you 07:04 MTDiscord So I suppose now that I think about this even further the only thing I can think of to fix it is, the code base needs to be a bit more modular. There is quite a lot of tight coupling in a few places. And that should probably be a target for 5.19 or 6.0 or whatever. Modulate the code base so components can be ripped out at will or rewritten 14:28 MinetestBot 02[git] 04ubuntuyou -> 03luanti-org/luanti: Improve volumetric lighting performance (#17360) 13e2ca283 https://github.com/luanti-org/luanti/commit/e2ca283f26b05d4f14293a1742ebcf617ecc09b8 (152026-08-21T14:28:09Z) 14:28 MinetestBot 02[git] 04jordan4ibanez -> 03luanti-org/luanti: Add textcolor to label docs 13a6292f6 https://github.com/luanti-org/luanti/commit/a6292f6fe129876563738bf411be304d025c7b0f (152026-08-21T14:28:21Z) 14:30 MinetestBot 02[git] 04sfan5 -> 03luanti-org/luanti: Optimize image transform modifier (#17384) 137d5626b https://github.com/luanti-org/luanti/commit/7d5626ba100a9edaaf7df82f8a70024437a414f7 (152026-08-21T14:28:49Z) 14:30 MinetestBot 02[git] 04cx384 -> 03luanti-org/luanti: Don't use getNodeNoEx for mapblock meshgen (#17411) 13e918474 https://github.com/luanti-org/luanti/commit/e9184745e2a39c0a6de3047fc4dbdeaf12e34a48 (152026-08-21T14:29:39Z) 14:30 MinetestBot 02[git] 04SmallJoker -> 03luanti-org/luanti: Improve utilities 13148af5e https://github.com/luanti-org/luanti/commit/148af5efc4aba7049a4711d552a72078bbca10b5 (152026-08-21T14:29:53Z) 14:30 MinetestBot 02[git] 04SmallJoker -> 03luanti-org/luanti: Chat: Prevent leaking RETURN key to newly opened formspec 13362ddef https://github.com/luanti-org/luanti/commit/362ddefe2dc0b448c122d672c586e27ec370cc34 (152026-08-21T14:29:53Z) 14:30 MinetestBot 02[git] (9 newer commits not shown) 15:24 cheapie Quite a long time ago I made this, a 6-floor movestone elevator with a relay logic controller: https://cheapiesystems.com/media/images/screenshot_20260821_102223.png 15:24 cheapie I finally got around to drawing a ladder diagram of how that thing is wired: https://cheapiesystems.com/media/6-floor-relay-logic-elevator.pdf 15:43 sfan5 usually the elevator logic is not bigger than your entire building 15:56 cheapie sfan5: For all you know, that building in the bottom left could extend several km. It doesn't, but it could :P 18:00 [MatrxMT] i got around to reading openslop 18:00 [MatrxMT] and apparently "inclusive language" is a thing 18:01 [MatrxMT] normal chess is outlawed immediately in favor of inclusive chess where the beginning side is chosen by rock paper scissors 18:24 [MatrxMT] Starting white or black is just mirroring the game. 18:29 [MatrxMT] yeah but it's now ✨️inclusive✨️ 18:32 sfan5 can we be more inclusive towards ai 18:43 bgstack15 birdlover32767: which openslopware page are you reading about "inclusive language"? 18:43 [MatrxMT] here's a link: https://codeberg.org/ethical-foss/open-slopware/src/branch/main/CODE_OF_CONDUCT.md#inclusive-language 18:44 MTDiscord so? 18:44 technomancy wow, so edgy, complaining about a code of conduct. is it somehow 2016 again? 18:45 bgstack15 Oh, their CoC about their own project. For some reason I thought you meant they were basing their criteria for "slopware" based on those other projects' "inclusive language." 18:45 [MatrxMT] it's 2026 so probably 18:45 MTDiscord "Some example nouns you can use if you don't know the gender of someone: maintainer" 18:45 bgstack15 Well, I'm a huge believer of freedom of association. 18:45 MTDiscord non-binary people don't own you maintaining your software project some would argue 18:45 bgstack15 If somebody dislikes that somebody else has standards, I suppose it would be best if they parted ways rather than go all Poettering 18:48 [MatrxMT] and so i will 18:48 MTDiscord this does seem a little bit like choosing to fighting a battle no-one asked you to fight, and then losing it, can't really see how getting upset at a code of conduct makes you look like the responsible person in the room with views on LLM usage that are taught out 18:48 Menchers lol, claiming "master" and "blacklist" are non-inclusive 18:48 MTDiscord thought* 18:48 bgstack15 Good thing we moved past IDE cables a long time ago, lol 18:49 MTDiscord it's still part of the ATA standard might be marked obsolete but the bit is still there and even in 2001 it was device 0 and device 1 unless you were non-technical 18:50 cheapie I've called them "device 0" and "device 1" for ages now 18:52 cheapie Similar story with most of these other terms, I usually use the newer ones just to avoid the drama but I generally don't care a whole lot what anyone else says 18:53 technomancy cheapie: ah, but have you tried complaining about it on IRC? 18:53 * cheapie complains to technomancy on IRC about... I dunno, advtrains or something 18:53 cheapie (*cough* https://gist.githubusercontent.com/cheapie/104377ee27f90c0ce1f814d5ad173375/raw/0992cb4c20319f0289f31bf420bc081d9eaa7113/advbugs.txt *cough*) 18:56 MTDiscord cheapie: that's just Postel's Law applied to human communication ;) 18:59 MTDiscord cheapie: correct technical term in the first place, "master" and "slave" made no sense and I have to assume were made up out of an assumption non-technical people are afraid of numbers, there's nothing in the protocol that makes one drive primary or secondary apart from some slight details in the state transitions 19:00 MTDiscord or I guess I should say state changes least the anti PC language police gets upset about that 19:00 cheapie I'm wondering if it might be one of those things lingering around from the old pre-standardization versions of it, like the "IDE" name itself 19:02 MTDiscord maybe yes? I would need to research it, and I'm not sure I personally care enough to do that, it seems like a lot of work, one of those things where you really can't trust the later sources much since someone is going to just lie and come up with something or misremember 19:03 cheapie Yeah, I'd probably have to buy yet more standards PDFs to do that and I don't really want to spend actual money to settle some argument on the internet /again/ 19:04 MTDiscord you can get the ata ones for free for 7, I think 6 is out there too I presume the earlier ones are as well, master-slave is actually used in one diagram label for ata 7, one of those physical interface specification with respect to the jumper switches. 19:04 MTDiscord https://www.hddguru.com/documentation/2006.01.27-ATA-ATAPI-7/ 19:05 cheapie Those are just the drafts that tend to float around 19:05 MTDiscord I doubt they are different from the standard 19:05 cheapie probably not much 19:07 MTDiscord it's a compliance thing, you put out the draft for free, you do a revision that removes the word draft from the Word document, and then you make that the standard, I don't think any liked the standards being paid but couldn't get around it so just dump the draft on the mailing list/members area 19:08 MTDiscord from the ATA-3 (draft) 19:08 MTDiscord "2.1.12 master: In ATA-1, Device 0 has also been referred to as the master. Throughout this document the term Device 0 is used" 19:10 MTDiscord they then proceeded to again miss that they used it for the physical interface spec for the 68-pin connector 19:11 MTDiscord > This standard provides the capability of operating on the AT Bus in a daisy > chained configuration with a second drive that operates in accordance with > these standards. One drive (selected as Drive 0) has been referred to as the > master in industry terms and the second (selected as Drive 1) has been > referred to as the slave (see figure 3 From the ATA-1 spec, it seems like yes this was a pre-standard term cheapie 19:11 MTDiscord well from the draft but you know what I mean 19:16 MTDiscord it looks like it might have been related to the soon to be removed spindle sync, probably yes came from a pre-standard world where there were different architectures 21:50 Sheriff_U3 Some thing funny I noticed about openslopware... they are using git... which happens to also be on their list of "bad" software... https://codeberg.org/ethical-foss/open-slopware#version-control-clients-and-tooling 21:50 crazylad haha lmao 21:51 crazylad "we're gonna complain about it but we're still gonna use it" - that's how you know these people don't actually believe in anything 21:51 Sheriff_U3 I would open an issue to tell them, but they banned me... 21:51 [MatrxMT] ah well apparently they're using `got` instead 21:51 MTDiscord <.zenonseth> they should have the linux kernel there too, if they don't already 21:51 [MatrxMT] ^ they do 21:51 Sheriff_U3 they have it there 21:52 MTDiscord <.zenonseth> great, and i guarantee most parts of codeberg run on that! 21:52 MTDiscord <.zenonseth> good for them 21:52 Sheriff_U3 lol 21:53 Sheriff_U3 These are the comments they banned me over... https://codeberg.org/ethical-foss/open-slopware/pulls/221 21:53 MTDiscord <.zenonseth> I'm sure many of you have read it but i liked Linus's view on this: https://lore.kernel.org/linux-media/CAHk-=wi4zC+Ze8e+p3tMv8TtG_80KzsZ1syL9anBtmEh5Z40vg@mail.gmail.com/?x=0 21:59 [MatrxMT] i don't see why that would be bannable 22:03 Sheriff_U3 I countered their arguments. I guess that's all it takes. 22:05 MTDiscord <.zenonseth> Also, not sure this is the right channel for this, but while I don't really care if they maintain a list of "slopware," I think calling the organization "ethical-foss" is misleading at best. Their focus seems to be just against LLMs (or "AI"), but that's only one small, and most recent, part of software ethics. Things like user privacy, surveillance, security, accessibility, licensing, deceptive UX, labor practices, sustainability, etc 22:05 MTDiscord etc. are also real ethical issues, but no they're not really (at least at the moment) for "ethical" dev, but more just "anti-LLM" and they try to make it about ethics without caring about the other parts of ethics /rant 22:06 MTDiscord <.zenonseth> also which comment were you banned for exactly? I didn't see anything in there 22:06 technomancy yeah sure would suck to solve one part of a problem without solving the entire thing 22:07 Sheriff_U3 I don't know for sure... I just logged on the next day and noticed I was banned, no warning, no email, no notice of why. 22:07 MTDiscord <.zenonseth> not the point, if my agenda is to get rid of LLMs but i phrase it as "I'm just being ethical" then it's misleading isn't it? And I'm really going to argue that the ethics of LLMs is debatable and not the highest priority imho 22:07 technomancy that makes no sense 22:08 MTDiscord <.zenonseth> which part? 22:10 MTDiscord <.zenonseth> also fyi, I am not a fan of what LLMs are using, but I don't like this constant anti-LLM, "if you use llms you're doing unethical things" bit - i wrote a rant here https://github.com/ZenonSeth/about/blob/main/README.md 22:11 MTDiscord have they "desloppified" curl yet of all the fixes to vulnerabilities found using LLMs? :juanchi_face: 22:23 MTDiscord <.zenonseth> also Sheriff_U3 - those comments are absolutely no reason to ban anyone, read the entire thread, ruben even replied after you with similar sentiment, I'm sure they didn't ban him 22:23 rubenwardy nope no ban 22:24 Sheriff_U3 Idk why they banned me then... That was the first and last time I've had contact with them. 22:27 rubenwardy could it be something you've said elsewhere? 22:28 Sheriff_U3 Not that I can think of, I didn't comment on anything else in the repo and haven't contributed either. 22:28 MTDiscord crazylad ""we're gonna complain about it but we're still gonna use it"": or it could be a joke, though the org name isn't telling of that 22:28 MTDiscord <.zenonseth> if they read this chat, would they ban me, even though I've never made any comments on their entire org? 22:29 Sheriff_U3 Possibly based off of their Code of Conduct actually 22:29 Sheriff_U3 If that 22:29 MTDiscord <.zenonseth> feel free to send them my comments if anyone wants to experiment, here's my codeberg profile: https://codeberg.org/ZenonSeth 22:30 Sheriff_U3 If thats the case then who knows what I could have been banned over... 22:30 MTDiscord <.zenonseth> lol, banning people who have never engaged directly with you would be the epitomy of.. something, idk what exactly to call it, but it's not a good sign 22:30 technomancy been a while since I've seen https://knowyourmeme.com/photos/1259257-we-should-improve-society-somewhat acted out in real time 22:31 Sheriff_U3 Cause things like not using neutral language are against their CoC. 22:33 MTDiscord <.zenonseth> technomancy it's literally like you're strawmanning this, and you think I'm talking about this as some kind of 'gotcha' and not trying to address it like the complex issue that shouldn't have extremes, nor that if you call some org "ethical" maybe you should focus on all tings ethical 22:34 MTDiscord <.zenonseth> by which i mean their organization is literally called "ethical-foss" - i didn't put that there, i'm not saying that protesting llm's as un-ethical is wrong because they're not also protesting - but i think that's a misleading name for what their organzation is actually doing 22:35 MTDiscord <.zenonseth> anyway, this is probably not going to change your or anyone's mind, feel free to dislike me or whatever it is in your head, i'm out, peace 22:37 pgimeno I finally got around to drawing a ladder diagram of how that thing is wired: https://cheapiesystems.com/media/6-floor-relay-logic-elevator.pdf 22:37 pgimeno Holy ladder, Batman! Also amusing to see "Modbus" here. What are those relays? From the top view there don't seem to be any mesecons AND gates, which would probably be a logical choice for relay logic off the top of my head 22:38 cheapie pgimeno: The mention of modbus is because I used a PLC program editor to draw the diagram 22:38 pgimeno oh haha ok 22:38 cheapie The relays are made with pistons moving power plants or conductors 22:38 pgimeno ahhh 22:40 cheapie Rungs 107 and 108 (which in the real thing are combined into one with an output at both ends) are the only ones where any real complex logic happens 22:41 cheapie And all those really do is compare the numbers stored by the xPM and xCM relays and energize down if PM>CM or up if PM a thought: LLMs can cause problems yes, but I don't think there are problems exclusive to LLMs