01:45 < bridge> If you mean workshop mods they just get dropped into a folder and it's entirely up to the official application to manage the functionality of the mods. 01:46 < bridge> Some games just support re-textures or changing assets. 01:49 < bridge> Most games that have highly advanced mods on the steam workshop either have an extensive modding api, or they use C# and let the mods do whatever runtime patching they want. 01:57 < bridge> time to ditch rust and c++ in favor of c# 02:01 < bridge> OR, the mods can monkeypatch ddnet by compiling their code at runtime and injecting into memory 02:01 < bridge> with a launcher, one could support multiple executables, given, they live in some specific folder 02:01 < bridge> sounds reasonable 02:07 < bridge> whats that 02:15 < bridge> the sub tick aiming stuff? 02:17 < bridge> it functionally applies antiping to your tee based on the inputs you're currently holding 02:18 < bridge> idk if that makes any sense 02:28 < bridge> <01000111g> what do u mean exactly, can u elaborate? I thought it just adds some visuals, does it change how pred works completely? 02:28 < bridge> oh, is it for gores gameplay 02:38 < bridge> it was just explained 10 minutes before u replied (with the ping on :steamsuffering: ) 02:39 < bridge> it's not really related to the gamemode at all but it's probably much more useful in gores? idk 02:39 < bridge> I couldn't tell you because I suck at ddrace 02:39 < bridge> works on ddnet too but the use of pretty much anything but static map tiles and player movement is buggy 02:39 < bridge> I remember smth abt antipi g applying whether youre frozen or not too 02:39 < bridge> that's a different feature but that one is almost entirely for gores 02:40 < bridge> related, itd be nice for mods to send their custom prediction or antiping 02:40 < bridge> yea there’s not really a model that works for that atm 02:40 < bridge> what does that mean 02:41 < bridge> server could theoretically send some kind of data to help the client predict its stuff 02:41 < bridge> ah 02:41 < bridge> ddpg works that way 02:41 < bridge> the server sends the entire game physics so everything is predicted 02:41 < bridge> it would be nearly impossible to do that in ddnet though 02:42 < bridge> It would just require major refactoring and would be incompatible with old clients and servers 02:42 < bridge> so yeah. nearly impossible 😃 02:42 < bridge> i will be switching to ddpg on release 🙋 02:42 < bridge> I have bad news for u 02:43 < bridge> the issue is you need ddpg servers and every mod needs to be written based on ddpg 02:43 < bridge> there's no cross compatibility 02:43 < bridge> but maybe some mod authors will make cool mods that use super different physics 02:44 < bridge> its okay ill be good at rust by the time it releases 02:44 < bridge> gerdöh already implemented insta :kek: 02:44 < bridge> you have ~1 1/2 months 02:45 < bridge> i'll be implementing zombies boys 02:45 < bridge> ill do bomb tag :poggers2: 03:10 < bridge> I think that's reserved to furo: and bencie I'm afraid :feelsbadman: 03:12 < bridge> ok then im taking zombies so you can find something else 03:14 < bridge> Ok I code battlefield 06:05 < bridge> https://cdn.discordapp.com/attachments/293493549758939136/1290887452554297418/e85b8a6fda51d647aeb4f02d310a05a2.jpg?ex=66fe1820&is=66fcc6a0&hm=543afb19bdd3cd9098e306f6e6511fc2236f1f0d30ecb5b1312de0c41765c892& 07:53 < bridge> I am in a public space and was just enjoying some reading in dev chat :( 07:54 < bridge> :nouis: 08:08 < bridge> I didn't even need a separate rule for this, I just slapped this and it works xd: 08:08 < bridge> ``` 08:08 < bridge> ((identifier) @variable.builtin 08:08 < bridge> (#eq? @variable.builtin "this") 08:08 < bridge> (#has-ancestor? @variable.builtin method_definition)) 08:08 < bridge> ``` 08:08 < bridge> I didn't even need a separate rule for this, I just slapped this and it works xd: 08:08 < bridge> ```scm 08:08 < bridge> ((identifier) @variable.builtin 08:08 < bridge> (#eq? @variable.builtin "this") 08:08 < bridge> (#has-ancestor? @variable.builtin method_definition)) 08:08 < bridge> ``` 09:24 < bridge> I just spam you with my brainstorming about it. But it kinda removes the time until next tick 09:24 < bridge> ## Fast prediction code (lower input lag) brainstorming 09:24 < bridge> 09:24 < bridge> Biggest problem is a direction change (looks jittery and is FPS dependent). 09:24 < bridge> One thing is clear. this gives _up to_ 20ms less input lag (visually). 09:24 < bridge> ``` 09:24 < bridge> Tick 0 --------------- 1 Tick 09:24 < bridge> ^ input to move right at 0.25 intra tick 09:24 < bridge> Tee-pos (0,0) --------------- (1,0) Predicted tee-pos 09:24 < bridge> <----------> in this range the tee has to do 09:24 < bridge> the full movement from non-redicted 09:24 < bridge> to predicted pos 09:24 < bridge> ``` 09:24 < bridge> 09:24 < bridge> holding move: 09:24 < bridge> 09:24 < bridge> ``` 09:24 < bridge> Tick 1 --------------- 2 Tick 09:24 < bridge> ^ input to STOP moving at 0.75 intra tick 09:25 < bridge> Tee-pos (1,0) --------------- (1,0) Predicted tee-pos 09:25 < bridge> ^ (1.75, 0) at this moement the tee is here visually 09:25 < bridge> <-> in this range the tee has to do 09:25 < bridge> the RESET the movement from non-redicted 09:25 < bridge> to predicted pos 09:25 < bridge> So literally reverting 0.75 moved units 09:25 < bridge> Additionally it would need to keep a second prediction 09:25 < bridge> world, to even have something to interpolate against. 09:25 < bridge> ``` 09:25 < bridge> 09:25 < bridge> - Since this is kind of anti-input lag (similar to anti-ping) this should maybe be configurable 09:25 < bridge> - How can we make the impact of the reset less? 09:25 < bridge> - How to keep track of the second prediction world (or even more?) 09:40 < bridge> I've had a lot of people on 60hz say they prefer it 09:41 < bridge> idk if fps has any impact 09:41 < bridge> the real issue is that it makes jump timing way harder 09:42 < bridge> any action that sets your velocity instead of giving you acceleration feels worse 09:43 < bridge> > How to keep track of the second prediction world (or even more?) 09:43 < bridge> the client already saves the positions of every tee on every tick of prediction so you don't need multiple worlds for most things unless you care about non-tee objects 09:47 < bridge> The number of ppl losing their e-mail account is not very accurate as they can login with their username password on the website. But after most people registered and have their ingame login, they dont reach out to us for months until they leaked it & then the fun begins. 09:47 < bridge> 09:47 < bridge> As we cant give out information & mods dont see e-mails at all, they have to escalate it to admins 09:47 < bridge> ah, i think that part was about our idea to make it more smooth 09:48 < bridge> it kinda would need multiple prediction worlds i think 09:48 < bridge> to interpolate between the predicted worlds xd 09:48 < bridge> ah 09:48 < bridge> But that part is defs the hardest 09:49 < bridge> esp. for dd-pg bcs all mods would need to implement it 09:49 < bridge> I wonder if it would look better anyway xd 09:50 < bridge> >> I've had a lot of people on 60hz say they prefer it 09:50 < bridge> idk if fps has any impact 09:50 < bridge> 09:50 < bridge> Ah ok, well I'm not very used to 60hz. But for me it felt extra terrible 😄 09:50 < bridge> does regeneration of the ingame login invalidate the old token? 09:50 < bridge> > I've had a lot of people on 60hz say they prefer it 09:50 < bridge> > idk if fps has any impact 09:50 < bridge> 09:50 < bridge> Ah ok, well I'm not very used to 60hz. But for me it felt extra terrible 😄 09:50 < bridge> Yes 09:51 < bridge> This scheme doesn’t sound very fps dependent. Why is it? 09:51 < bridge> At least my brain thinks the impact is negitble on 240hz 09:51 < bridge> I think it's a brain thing 09:51 < bridge> the frames come out faster so the brain has more info 09:52 < bridge> On 60hz it felt like it jumped quite hard xd 09:52 < bridge> > they dont reach out to us for months until they leaked it & then the fun begins. 09:52 < bridge> @avolicious what's the fun there? going back and cleaning up the malicious logins? 09:52 < bridge> I was going to implement variable amount of fast input between 0-20ms so you can set it to 10ms and experience less jitter but the issue with this is that it requires the client to lerp between 3 world states during a single intratick period. 09:52 < bridge> 09:52 < bridge> Basically like this 09:52 < bridge> 09:52 < bridge> ``` 09:52 < bridge> 0 -------------- 1 -------------- 2 09:52 < bridge> ^ Intratick = 0 here 09:52 < bridge> ^ Intratick = 1 here 09:52 < bridge> ``` 09:52 < bridge> but the client only saves 2 values at a time which makes it so you have to change a bunch of stuff 09:53 < bridge> Yeah that is defs also a cool idea 😄 09:53 < bridge> At least to see if that still feels smoother 09:54 < bridge> yeah but it probably doesn't feel as good 09:54 < bridge> idk if there is any perfect solution, I think there probably isn't because the game will always run on 50tps steps 09:55 < bridge> yep 09:55 < bridge> They forget their e-mail they used, mostly creating new discord accounts, so its hard to verify if they are the real owner of this account 09:55 < bridge> This is the "fun" part 09:55 < bridge> ah 09:55 < bridge> so it does happen 😦 09:55 < bridge> There can be no perfect solution to prediction, it’s a prediction after all 09:55 < bridge> let's hope there's less of this with steam-bound accounts 09:56 < bridge> Unless you invent time travel I guess 09:56 < bridge> well this is like a meta sub-tick prediction where you try to fake a higher tickrate 09:56 < bridge> you're predicting where the tee will be in 0.5 ticks from now 09:57 < bridge> I was also thinking of linking steam accounts too 😄 09:57 < bridge> But it still increases the odds of a mispredict by predicting at all 09:57 < bridge> well if you don't predict there cannot be a mispredict 09:57 < bridge> so infinity% increase in odds 09:57 < bridge> Exaclty, and thus less jitter 09:57 < bridge> Tbf "normal" prediction almost always works for the local tee. 09:57 < bridge> 09:57 < bridge> This one _always_ breaks it even for the local tee 09:58 < bridge> fast input is like a bet 09:58 < bridge> that u don't change the direction xD 10:00 < bridge> also you pay 20ms ping penalty if you have antiping on for the privilege to make this bet 10:00 < bridge> If it always breaks you might be doing something wrong. Afaik, the client technically actually has all the information needed to pretend there is 0 ping and even do this subtick fakery perfectly. As long as it’s the local tee and there are no other tees in the way 10:00 < bridge> this isnt anti ping 10:00 < bridge> this is predicting the future 10:00 < bridge> i also call it future prediction 😬 10:01 < bridge> it doesnt have enough info inside the sub tick 10:01 < bridge> All predictions are by definition about the future, it’s just that you are using a different reference time 10:01 < bridge> so it kinda has to hope the input doesnt change during the same sub tick 10:01 < bridge> Yeah but, normally u talk about ticks 10:01 < bridge> 10:01 < bridge> here u talk about subticks 10:02 < bridge> in some sense, if u say 1 tick is 1 time unit. then this doesn't make sense 10:02 < bridge> regular prediction is just the client pretending to be the server, this is like trying to know what the next tick will be before the server or the client knows 10:04 < bridge> Ah, and we have people trying to iterate over login codes 😄 10:04 < bridge> Generating them randomly & trying to login 10:04 < bridge> Why don’t you stay behind one subtick at all times? Then you will never get jitter from this and you will only be 1 subtick behind of “truth” 10:04 < bridge> lmao 10:04 < bridge> I thought the login codes are huge, why would anyone even attempt that 10:05 < bridge> ¯\_(ツ)_/¯ 10:05 < bridge> Kids ™️ 10:05 < bridge> It gets detected anyways 10:05 < bridge> then they open a ticket & ask why they cant login 10:05 < bridge> Seen that a few times now 10:05 < bridge> 🤣 10:05 < bridge> u'd need to be 1 tick behind. in other words, u get what ddnet is doing rn anyway 😄 10:05 < bridge> well up to 1 tick 10:06 < bridge> to be precise 10:06 < bridge> to be fair I think all games do it how ddnet does it atm, it's sort of the objectively correct way to do it 10:06 < bridge> I don’t seem to have the right words to convey this. Worry not, I’ll show you how when I make ddnet-asm 10:07 < bridge> idk what minecraft does tho 10:07 < bridge> guys, who knows how to add such a thing to their server, or create their own??? 10:07 < bridge> https://cdn.discordapp.com/attachments/293493549758939136/1290948227805347851/image.png?ex=66fe50ba&is=66fcff3a&hm=2f0ee50884b11f0dc11cc4216f236cfa36591c79eacf77ebeb7c62a9f4788a9f& 10:07 < bridge> somehow it runs at 20tps and I don't think the client has any input lag or server desyncs 10:07 < bridge> yeah, normally games simply do more ticks if u need better gameplay xd 10:07 < bridge> A fair bit of games just let the client run their own simulation entirely and they trust the client for it mostly 10:08 < bridge> true 10:08 < bridge> And in CS case the server and client can negotiate the tick-rate 😏 10:08 < bridge> i think minecraft also trusts the client? 10:08 < bridge> at least in past u could just fly xD 10:08 < bridge> This would be my guess with all the rampant cheating 10:08 < bridge> it's not so simple 10:08 < bridge> the minecraft client and server have a complex relationship 10:08 < bridge> a highly janky one 10:08 < bridge> time for minetee, the better minecraft 10:09 < bridge> https://wiki.vg/Main_Page 10:09 < bridge> @avolicious is it a secret how your login tokens are generated? 10:09 < bridge> like what algo 10:09 < bridge> Algo = os.random 10:09 < bridge> i see so just random bytes, 32? 10:09 < bridge> Yep 10:10 < bridge> No need for complex computing 10:10 < bridge> who knows??? 10:10 < bridge> yeah, that is defs unbreakable for any normal player 😄 10:10 < bridge> ok wtf 10:10 < bridge> minecraft has 50ms input delay?? 10:10 < bridge> > The physics engine runs at 20 ticks per second, meaning the game's physics are updated every 50ms. 10:10 < bridge> > 10:10 < bridge> > This includes the player’s position and speed, the environment, and entities. 10:10 < bridge> > 10:10 < bridge> > In-game actions are performed at the end of each tick, regardless of the time or order in which they were called. 10:10 < bridge> > 10:10 < bridge> > Due to the game working this way, inputs may take up to 50ms of delay between the button press and their activation. 10:10 < bridge> i mean in 3d everything feels different anyway 10:11 < bridge> idk how I never noticed it 10:11 < bridge> I guess you get used to it 10:11 < bridge> guys help please 10:11 < bridge> also minecraft is kinda slow 10:11 < bridge> jumping is a fast action 10:11 < bridge> if it would add osu gameplay u'd probs notice it easier xdd 10:11 < bridge> We detect bruteforce attacks & block the ip 😄 So yeah, its pretty unbreakable 10:12 < bridge> i guess even without it would be unbreakable, but yeah that makes it impossible 10:12 < bridge> https://github.com/PaperMC/Folia might be interesting tho 10:12 < bridge> I think some crypto wallets might use less bits than the kog login token 10:13 < bridge> also kog servers need to actively report if the token was correct. that alone is a rate limit 10:13 < bridge> yes 10:13 < bridge> It works async, so you could theoretically spam login tokens 10:14 < bridge> As its just a chat message/command 10:14 < bridge> every chat message/command is an event 10:14 < bridge> and treated that way 10:14 < bridge> a ed25519 is also just 32bytes long 10:14 < bridge> and u often have the public key to it 10:14 < bridge> 1) you need to come up with a name for the community 10:14 < bridge> 2) draw a logo sized 128x64 10:14 < bridge> 3) launch about 20 servers with a Linear map 10:14 < bridge> 4) wait until there are 30-40 people online 10:14 < bridge> 5) create a request to add a community via GitHub https://github.com/ddnet/ddnet/issues or Discord bot #✉-create-a-ticket 10:15 < bridge> idk this is a lot of work to get a less stable version which doesn't support plugins and still only has 2x-3x total players 10:15 < bridge> It’s probably unbreakable period. I don’t imagine any modern linux kernel shipping with subpar `/dev/urandom` 10:15 < bridge> Paper is fairly stable, Folia is just an addition 10:15 < bridge> thank you, I'll try 10:15 < bridge> did you read the github? 10:16 < bridge> > Current broken API 10:16 < bridge> > Most API that interacts with portals / respawning players / some player login API is broken. 10:16 < bridge> > ALL scoreboard API is considered broken (this is global state that I've not figured out how to properly implement yet) 10:16 < bridge> > World loading/unloading 10:16 < bridge> > Entity#teleport. This will NEVER UNDER ANY CIRCUMSTANCE come back, use teleportAsync 10:16 < bridge> > Could be more 10:16 < bridge> Conclusive proof that tickrate isn’t that important. It’s all in the brains of impressionable young men 10:16 < bridge> 20 servers? and how do people who have only 3 servers create? 10:16 < bridge> 20 for example 10:17 < bridge> I think Folia is a fork of paper not an addon, it's rewrites the entire engine 10:17 < bridge> okay 10:17 < bridge> it is cool tho 10:17 < bridge> Folia and Paper is from the same devs 😄 10:17 < bridge> But yeah, they are adding multithreads to it 10:18 < bridge> or atleast they try 10:18 < bridge> but all the minecraft server implementation differ a bit 10:18 < bridge> from the original implementation 10:19 < bridge> Did they multithread minecraft? 10:19 < bridge> yes 10:19 < bridge> it's region based 10:20 < bridge> That’s smart, I wonder how they handled the boundaries of the threads 10:20 < bridge> Hardware 10:20 < bridge> > What hardware will Folia run best on? 10:20 < bridge> > Ideally, at least 16 cores (not threads). 10:20 < bridge> I think it would be cool if there was a multi instance version and associated client mod so you could scale a single world to infinite players at the cost of boundaries 10:21 < bridge> Lets assume they have 16 cores and HT enabled 😄 10:21 < bridge> Thats an expensive server tho 10:21 < bridge> I meant as in how they handled something crossing the boundary from one threads authoritative region to the other 10:21 < bridge> I think it's dynamic 10:22 < bridge> > For a server with many spread out players, Folia will create many spread out regions and tick them all in parallel on a configurable sized threadpool. Thus, Folia should scale well for servers like this. 10:22 < bridge> contention incoming 10:23 < bridge> 10:23 < bridge> i wonder how fast it would be in a system programming language 10:23 < bridge> java is always like twice as slow xd 10:23 < bridge> it appears very complicated 10:23 < bridge> it appears to be very complicated 10:24 < bridge> Hm, this sounds very hard to do, I’ll have to read it 10:24 < bridge> I agree and it looks like they tried very hard 10:24 < bridge> > Inter-region operations 10:24 < bridge> > Inter-region operations refer to operations that work with other regions that are not the current ticking region that are in a completely unknown state. These regions may be transient, may be ticking, or may not even exist. 10:25 < bridge> Stuff like redstone crossing the boundary sounds impossible to handle properly 10:25 < bridge> they probably do not care 10:25 < bridge> paper does not even have vanilla redstone mechanics normally 10:33 < bridge> hi, what should I do if I was banned for a bot client, but I don't use it and no CFGs, please help 10:34 < bridge> were you using any custom client 10:34 < bridge> no 10:34 < bridge> client or stiam 10:34 < bridge> steam* 10:34 < bridge> for how long did you get banned 10:35 < bridge> 1 day , next time for a year , I don 't want to , it 's a Varna - type system 10:35 < bridge> \ 10:35 < bridge> https://cdn.discordapp.com/attachments/293493549758939136/1290955341970018308/image.png?ex=66fe575a&is=66fd05da&hm=35b2baba7653cd5192e2016fbb1fe6f8b1a2acd940514c65c7fa32a8298df004& 10:35 < bridge> it should be a week next time if i remember correctly 10:36 < bridge> if you are 100% certain u used steam version. #✉-create-a-ticket -> ban appeal 10:36 < bridge> 10:36 < bridge> But be prepared to be rejected 10:36 < bridge> okey 10:37 < bridge> thanks 10:39 < bridge> https://www.reddit.com/r/yuzu/s/5XNp0ISC2i Nintendo going all out 10:40 < bridge> Is this the past? 10:40 < bridge> Or what is different this time 10:40 < bridge> It's today. They took down another emulator, Ryujinx 10:41 < bridge> Ah I was confused by the yuzu subreddit xD 10:41 < bridge> Oh, yeah the subreddit is still there 11:02 < bridge> morning fellas 11:06 < bridge> ohayou mr ryozuki :owo: 11:14 < bridge> fuck Nintendo lmao 12:40 < bridge> who knows how to fix the fact that emotions cannot be spammed on my server? 12:42 < bridge> https://ddnet.org/settingscommands/ 12:42 < bridge> :f3: 12:42 < bridge> thx 12:42 < bridge> sv_emoticon_ms_delay 12:42 < bridge> is probably what u want 12:59 < bridge> @learath2 what do you think about variable shadowing? 13:00 < bridge> Idk, no strong opinions on it 13:01 < bridge> depends on the language, I'd guess 13:03 < bridge> I like it in rust so Ill yoink it to my language xd 13:41 < bridge> https://github.blog/changelog/2024-10-01-evolving-github-issues-public-preview/ 16:22 < bridge> @learath2 ? :( 16:49 < bridge> LOL, I guess BagleR got bored of waiting for me to add people 16:49 < bridge> Ok ok, I'll do it now 16:50 < bridge> It's your weird name you know. I started adding people, then I couldn't find your player profile link then I stopped 17:03 < bridge> xd 17:03 < bridge> @heinrich5991 sorry for ping, just for clarification, did you test 0.5 spec .json and its outcome from generate.py in libtw2? 17:04 < bridge> seems it misses `NETMSG_SERVERINFO` 17:05 < bridge> ah seems it's hardcoded, there's nothing in spec xd 17:07 < bridge> OMFGGGG CHILLER 17:07 < bridge> YOU DID NOT XDDDDDD 17:09 < bridge> > “Systems program building is an entropy-decreasing process, hence inherently metastable. Program maintenance is an entropy-increasing process, and even its most skillful execution only delays the subsidence of the system into unfixable obsolescence.” 17:09 < bridge> > ― Frederick P. Brooks Jr., The Mythical Man-Month: Essays on Software 17:09 < bridge> what u think 17:10 < bridge> Better dialog that anything in frieren 17:10 < bridge> i wanna read this book https://www.amazon.es/Mythical-Man-Month-Software-Engineering-Anniversary/dp/0201835959 17:10 < bridge> Better dialog than anything in frieren 17:10 < bridge> its from there 17:10 < bridge> What holds you back? 17:11 < bridge> money 17:11 < bridge> and that 17:11 < bridge> i barely read books i buy 17:11 < bridge> XD 17:11 < bridge> ill ask my company to buy it 17:11 < bridge> they bought me the x86 asm book 17:11 < bridge> I never read a book from start to finish 17:11 < bridge> So I can relate 17:12 < bridge> That is also why I suck so hard at expressing myself 17:12 < bridge> 😏 17:14 < bridge> xd 17:14 < bridge> wrong ❗ 17:14 < bridge> @jupeyy_keks i like this quote more 17:14 < bridge> > The programmer, like the poet, works only slightly removed from pure thought-stuff. He builds his castles in the air, from air, creating by exertion of the imagination. Few media of creation are so flexible, so easy to polish and rework, so readily capable of realizing grand conceptual structures.... Yet the program construct, unlike the poet's words, is real in the sense that it moves and works, producing visible outputs separate from the co 17:14 < bridge> > 17:14 < bridge> > —Fred Brooks, The Mythical Man-Month 17:15 < bridge> programmers are artists that build things that can make stuff that also build things 17:15 < bridge> a artist just makes a painting and thats it 17:15 < bridge> well programming is like playing god 17:15 < bridge> yes 17:15 < bridge> your own physics 17:15 < bridge> we gods 17:15 < bridge> and if u use rust 17:15 < bridge> ur godlier 17:15 < bridge> your own what do i know 17:15 < bridge> kek 17:15 < bridge> ur a safer god 17:15 < bridge> :gigachad: :poggers2: 17:15 < bridge> rust is the language of gods 17:16 < bridge> > 17:16 < bridge> > “The general tendency is to over-design the second system, using all the ideas and frills that were cautiously sidetracked on the first one.” 17:16 < bridge> so true 17:16 < bridge> is dd-pg over designed? 17:17 < bridge> In some aspects probably it is already 17:17 < bridge> > 17:17 < bridge> > “By documenting a design, the designer exposes himself to the criticisms of everyone, and he must be able to defend everything he writes. If the organizational structure is threatening in any way, nothing is going to be documented until it is completely defensible.” 17:17 < bridge> this guy got some nice quotes 17:18 < bridge> One thing I like about many rust things is, that you understand the intend without the docs 17:18 < bridge> I think this is generally also better 17:18 < bridge> gimme generics 17:18 < bridge> But everytime that is not the case, it sucks even more 17:18 < bridge> ill make any code unreadable 17:19 < bridge> But the way rust does generics is abstracted in a sense, that you most likely need traits too 17:19 < bridge> But I understand what you mean 17:19 < bridge> rust generics are not that abstracted tbh 17:19 < bridge> generics and traits are a way to express logic that u cant do wrong 17:19 < bridge> its kinda painless to work with cpp templates 17:20 < bridge> what doesthat mean 17:20 < bridge> rust is more strongly typed than cpp 17:20 < bridge> Don't make me crime you 17:20 < bridge> also it has lot of nice patterns 17:20 < bridge> like new types 17:20 < bridge> please give me some notes on how to make good abstracted proxy trait 17:20 < bridge> or type state pattern 17:20 < bridge> https://cliffle.com/blog/rust-typestate/ 17:20 < bridge> https://doc.rust-lang.org/rust-by-example/generics/new_types.html 17:21 < bridge> Clearly false, the bible is in Latin 17:21 < bridge>