| Time |
Nick |
Message |
| 00:44 |
|
FileX joined #luanti |
| 00:54 |
|
kimapr joined #luanti |
| 01:20 |
|
FileX joined #luanti |
| 03:03 |
|
lmisu joined #luanti |
| 03:14 |
|
FileX joined #luanti |
| 03:23 |
|
FileX joined #luanti |
| 03:24 |
|
FileX joined #luanti |
| 03:37 |
|
amfl2 joined #luanti |
| 03:39 |
|
alias joined #luanti |
| 03:53 |
|
FileX joined #luanti |
| 04:00 |
|
MTDiscord joined #luanti |
| 04:14 |
|
Fulgen joined #luanti |
| 05:48 |
|
Talkless joined #luanti |
| 06:26 |
|
FileX joined #luanti |
| 06:48 |
|
YuGiOhJCJ joined #luanti |
| 07:09 |
|
jaca122 joined #luanti |
| 07:12 |
|
gregon joined #luanti |
| 07:12 |
|
gregon left #luanti |
| 07:18 |
|
dabbill joined #luanti |
| 07:54 |
|
FileX joined #luanti |
| 08:18 |
|
dabbill joined #luanti |
| 09:09 |
|
alias joined #luanti |
| 09:31 |
|
Kimapr_ joined #luanti |
| 09:43 |
|
Kimapr_ joined #luanti |
| 09:49 |
|
Kimapr_ joined #luanti |
| 09:56 |
[MatrxMT] |
<Blockhead256> we have a report on the forums that the server list is being abused with bad usernames and probably spoofed client numbers https://forum.luanti.org/viewtopic.php?p=446383#p446383 |
| 09:59 |
[MatrxMT] |
<birdlover32767> a username blacklist should solve this, but only temporarily |
| 10:00 |
[MatrxMT] |
<Blockhead256> we have had serverlist abuse before. Thankfully we are small enough that it doesn't happen often |
| 10:00 |
[MatrxMT] |
<Blockhead256> if Luanti were big enough the serverlist would have to change a lot or go extinct |
| 10:01 |
[MatrxMT] |
<leca> please don't |
| 10:01 |
sfan5 |
i don't see how the list can effectively prevent bad usernames appearing on genuine servers |
| 10:01 |
sfan5 |
(preventing bad or fake servers is much easier) |
| 10:01 |
[MatrxMT] |
<birdlover32767> i think what blockhead said was more about servers with inappropiate names |
| 10:02 |
[MatrxMT] |
<leca> I'd like Luanti to have a much bigger community (as it is community-centered game, it is up to community to create games/mods/etc content and servers) but I don't want it to be flooded with hackers and trolls |
| 10:02 |
[MatrxMT] |
<birdlover32767> +1 |
| 10:03 |
[MatrxMT] |
<birdlover32767> also luanti being bigger would also come with othe baggage |
| 10:03 |
[MatrxMT] |
<birdlover32767> such as pay to win servers |
| 10:03 |
[MatrxMT] |
<birdlover32767> other* |
| 10:03 |
[MatrxMT] |
<leca> I really love it, it gives me those warm feelings that minecraft did, but with much bigger taste of freedom which I really enjoy |
| 10:03 |
[MatrxMT] |
<Blockhead256> scunthorpe problem at the scale of the serverlist... no thanks |
| 10:04 |
[MatrxMT] |
<leca> Just looked up "scunthorpe problem" and it is hillarious |
| 10:04 |
[MatrxMT] |
<leca> Can't like a well-trained small AI model help with it? |
| 10:04 |
[MatrxMT] |
<leca> instead of dumb soulless regexp filters |
| 10:06 |
[MatrxMT] |
<birdlover32767> what about the false positives |
| 10:06 |
[MatrxMT] |
<Blockhead256> I take issue with your use of the word soulless there |
| 10:06 |
[MatrxMT] |
<Blockhead256> "just use AI" is also about as accurate as saying "use computer science" |
| 10:07 |
[MatrxMT] |
<leca> Obviously, perfect solution (that is 100% accurate) doesn't exist. But I think we should work toward a *better* solution |
| 10:07 |
[MatrxMT] |
<leca> Sorry, it was a kind of honestly bad word choose |
| 10:09 |
[MatrxMT] |
<leca> Not "just use AI", we could train our own model, not the big llamas or chatgpt which have billions of params, but a relatively small that could be calculated pretty fast even on CPU, to help filter out "scunthorpe" cases |
| 10:09 |
[MatrxMT] |
<Blockhead256> this is a do-ocracy, so get back to us when you have something deployable and lightweight enough if you really want to make it your goal |
| 10:11 |
[MatrxMT] |
<leca> I'd like to first discuss the idea and then get to work if its really worth it. Sadly, but I don't really have time and trying to do my best and help this community grow. I have some ideas I'd like to implement on my own but it's really complicated.. |
| 10:14 |
[MatrxMT] |
<Blockhead256> true, well, the do-ing isn't just hard dev work, you could look at what's on the issue tracker https://github.com/luanti-org/serverlist/issues |
| 10:14 |
[MatrxMT] |
<Blockhead256> and consider making a proposal |
| 10:14 |
sfan5 |
IMO there isn't urgent need to do anything right now |
| 10:15 |
[MatrxMT] |
<leca> I thought about making a CLI mod manager to manage mods and games and about improving the server console, but I don't know if it will be really helpful for the people and if I, myself, can do it. I'm really insecure about my code and I'm afraid that people will say 'this is crap don't do that', but I fell like I already wrote it here.. |
| 10:16 |
[MatrxMT] |
<leca> right now yes, but I think we should just think about it in case if it will be needed. It's better to have a plan than nothing π |
| 10:18 |
[MatrxMT] |
<Blockhead256> my first step when I consider what to make is to consider what's already out there, "prior art" |
| 10:18 |
[MatrxMT] |
<Blockhead256> I've been around long enough to remember stuff, and offhand I remember there's already two CLI mod managers posted to the forums |
| 10:19 |
[MatrxMT] |
<Blockhead256> that doesn't mean you should make your own, just try to make it different to and learn from what's come before |
| 10:19 |
[MatrxMT] |
<leca> I searched if it was already made, of course, but I found the repos to be abandoned like some years ago |
| 10:21 |
[MatrxMT] |
<Blockhead256> Luanti moves fairly slowly, so that doesn't really mean that what you found is obsolete |
| 10:22 |
[MatrxMT] |
<leca> I thought about writing it in C++ so that I could try merging it into the game, so you could write in server console 'mod install author/modname`, reboot the server and it would be installed. It would keep track of the installed mods and notify/update the mods at command and such helpful things... |
| 10:22 |
[MatrxMT] |
<leca> Didn't like the API for mods changed a bit for past few years? or is it just me? |
| 10:23 |
sfan5 |
having a cli mod manager inside luanti would be far more useful |
| 10:23 |
[MatrxMT] |
<leca> You can't imaging how got it is to hear it from you π |
| 10:23 |
|
ireallyhateirc joined #luanti |
| 10:24 |
[MatrxMT] |
<leca> You can't imagine how got it is to hear it from you π |
| 10:24 |
luatic |
i think a "CLI mod manager" is not *that* important to have. either you're a player and don't want to touch the command line - then you just manage content via the GUI. |
| 10:24 |
[MatrxMT] |
<leca> You can't imagine how good it is to hear it from you π |
| 10:24 |
[MatrxMT] |
<Blockhead256> 5.13 has some minor problems accumulated if your mod was made for earlier entries, but it should load mods that were mod for 5.0, just with minor issues |
| 10:24 |
luatic |
or you're a "power user", say a server admin, then you probably manage your game as a collection of submodules or something similar and don't need a CLI mod manager either. |
| 10:24 |
luatic |
(git submodules) |
| 10:24 |
[MatrxMT] |
<Blockhead256> git for fetching mods hasn't changed, and I can't really say, but I don't think fetching ContentDB stuff across HTTP has changed much either |
| 10:25 |
[MatrxMT] |
<leca> I thought about the content.luanti.org |
| 10:25 |
luatic |
because contentdb has a decent HTTP API, it's pretty easy to write a script which uses that |
| 10:25 |
[MatrxMT] |
<leca> agreed |
| 10:26 |
sfan5 |
i think there's a space between player and power use where people are comfortable with a CLI to install mods but don't want to mess with git |
| 10:27 |
[MatrxMT] |
<Blockhead256> I would urge people in that stage to just learn git, but there are still use cases like content that is not in git or fetching specific ContentDB releases |
| 10:28 |
[MatrxMT] |
<leca> I also thought about something like pip freeze, so you could save the exact state of versions of your mods |
| 10:29 |
rubenwardy |
I use a git repo with submodules for pip freeze like workflow |
| 10:29 |
rubenwardy |
for servers |
| 10:29 |
[MatrxMT] |
<Blockhead256> ^ what rubenwardy says is typical, https://github.com/pandorabox-io/pandorabox-mods is what I usually point to |
| 10:30 |
[MatrxMT] |
<Blockhead256> but honestly, git submodules are a pain in the neck |
| 10:33 |
[MatrxMT] |
<leca> So it is a generally poor/useless idea and I shouldn't really waste time on it. |
| 10:33 |
[MatrxMT] |
<leca> But what about server console? I think everyone here at least tried to create a minecraft server, maybe long ago, but tried. |
| 10:33 |
[MatrxMT] |
<leca> Personally, I really love minecraft console with its tab hints, command history, logs directory (its automatic compression on some software), working arrows and so on... |
| 10:33 |
[MatrxMT] |
<leca> When I tried luanti, it was really sad to find that it doesn't have the same functionality (nor the normal mode, or the curses). And I thought about rewriting this stuff, but is it that needed? I understand that while I enjoy using server console, it seems that I am pretty lonely it this so will it be accepted if I do it? |
| 10:36 |
[MatrxMT] |
<birdlover32767> you can make a feature request issue on the luanti repo to get feedback |
| 10:37 |
[MatrxMT] |
<leca> Sounds like a good idea, thank you! |
| 10:38 |
[MatrxMT] |
<Blockhead256> a lot of people manage their server with SSH into a linux system that holds the server, but your idea would really help average windows users |
| 10:39 |
[MatrxMT] |
<Blockhead256> a competing idea is web-based interfaces for server management |
| 10:39 |
[MatrxMT] |
<Blockhead256> that's more common in paid server hosting |
| 10:39 |
[MatrxMT] |
<leca> I'm doing it too. I start the server using tmux and its pretty enough for me rn to use, but I'm talking about all that QoL additions |
| 10:40 |
[MatrxMT] |
<Blockhead256> that's good to be Β thinking about, smoothing the workflow is good work to be doing |
| 10:40 |
repetitivestrain |
i tend to administer my server by running /lua (which is provided by luatic's excellent dbg mod) |
| 10:40 |
repetitivestrain |
and yes, i do not connect to my server as an administrator without connecting through a secure VPN |
| 10:41 |
repetitivestrain |
on which account i disabled the condition which disables /lua by default in multiplayer sessions |
| 10:41 |
repetitivestrain |
which i assume is far more versatile than minecraft's swing server console |
| 10:43 |
[MatrxMT] |
<birdlover32767> what about having a console with all privs in the terminal the server is launched in, but that might cause problems with the player arg to commands |
| 10:43 |
[MatrxMT] |
<Blockhead256> `/lua` is quite versatile, but unless you've built a decent library of lua functions to call, then it's not suited to repetitive tasks |
| 10:43 |
[MatrxMT] |
<Blockhead256> Also, do chatcommands have tab completion? I can't find it but I seem to recall something... |
| 10:44 |
[MatrxMT] |
<birdlover32767> they do not |
| 10:44 |
[MatrxMT] |
<leca> they don't afaik |
| 10:44 |
[MatrxMT] |
<leca> and it's the saddest thing for me |
| 10:44 |
[MatrxMT] |
<Blockhead256> whereas you could write bash (or other system) autocompletion rules |
| 10:45 |
[MatrxMT] |
<birdlover32767> autocompleting the command is easy but autocompleting the arguments would need a new api |
| 10:46 |
[MatrxMT] |
<birdlover32767> you can't know if an argument is a player or an itemstring or a message in the current definitions |
| 10:46 |
[MatrxMT] |
<leca> I thought about creating a kind of "completion server", it would receive the command do some priv checks and then send back possible variants for the completion. But i think it would require command's description. |
| 10:46 |
[MatrxMT] |
<leca> Like, describe a command like this: |
| 10:46 |
[MatrxMT] |
<leca> /ban <player> <time(s)> |
| 10:46 |
[MatrxMT] |
<leca> and the aforementioned server would parse this desc and complete <player> with cycling player names if there is no name or complete the name of an online person from first letters |
| 10:47 |
[MatrxMT] |
<leca> Like create something like "types" for arguments: players, time, coordinates, items, nodes and so on |
| 10:47 |
[MatrxMT] |
<Blockhead256> watch out that you don't send a message too long or with too many newlines, the IRC bridge has been banned before.. :\ |
| 10:48 |
[MatrxMT] |
<leca> Oh, sorry |
| 10:48 |
[MatrxMT] |
<birdlover32767> then define a new api for defining new argument types |
| 10:48 |
[MatrxMT] |
<leca> yes |
| 10:49 |
[MatrxMT] |
<leca> I think I need to think it through very much to create a really good and flexible api |
| 10:49 |
[MatrxMT] |
<Blockhead256> I would add a callback mechanism like on_player_tabcomplete(strCommandSoFar) -> returns a new string |
| 10:49 |
luatic |
^ yep, that's the kind of thing we'll want to do using SSCSM |
| 10:49 |
[MatrxMT] |
<Blockhead256> that would be in C++ mind |
| 10:49 |
[MatrxMT] |
<birdlover32767> that would be better done as a SSCSM |
| 10:49 |
sfan5 |
guys there's an issue for this on github |
| 10:49 |
sfan5 |
no need to recount everything |
| 10:50 |
[MatrxMT] |
<leca> What is (SS)CSM? |
| 10:50 |
[MatrxMT] |
<birdlover32767> server sent CSM |
| 10:50 |
luatic |
the answer to life, the universe, and all the rest |
| 10:50 |
[MatrxMT] |
<birdlover32767> csm is client side mod |
| 10:50 |
[MatrxMT] |
<leca> π |
| 10:51 |
[MatrxMT] |
<Blockhead256> https://github.com/luanti-org/luanti/issues/9663 |
| 10:51 |
[MatrxMT] |
<Blockhead256> https://github.com/luanti-org/luanti/issues/6450 |
| 10:51 |
[MatrxMT] |
<leca> And what's the problem with the sscsm? |
| 10:51 |
[MatrxMT] |
<birdlover32767> it doesn't exist yet |
| 10:52 |
[MatrxMT] |
<Blockhead256> https://github.com/luanti-org/luanti/issues?q=is%3Aissue%20state%3Aopen%20sscsm%20label%3ASSCSM |
| 10:52 |
[MatrxMT] |
<leca> oh heh |
| 10:53 |
[MatrxMT] |
<leca> As I see, it is a very powerful and needed thing. It is not being implemented for security reasons, an I understand? |
| 10:53 |
[MatrxMT] |
<birdlover32767> no one is developing it yet afaik |
| 10:54 |
[MatrxMT] |
<birdlover32767> unless someone is and its stuck in draft |
| 10:54 |
[MatrxMT] |
<Blockhead256> that's one main reason, the other is the complexity of adding sending Lua to the client and having the client execute it |
| 10:54 |
[MatrxMT] |
<Blockhead256> the main problem is making an acceptably-secure implementation, not a hackjob |
| 10:54 |
[MatrxMT] |
<leca> how does a normal mod works then? isn't it being downloaded from server |
| 10:54 |
[MatrxMT] |
<birdlover32767> it runs on the server |
| 10:55 |
[MatrxMT] |
<Blockhead256> no, Luanti's mods all run on the server, and only some definitions of nodes, entities etc. and media are transmitted to the client |
| 10:55 |
[MatrxMT] |
<birdlover32767> it is not being downloaded from the server, only the media (sounds, textures, models) are |
| 10:55 |
[MatrxMT] |
<leca> oh wow |
| 10:56 |
[MatrxMT] |
<Blockhead256> It's a two-edged sword: You can connect to any server without downloading anything beforehand, but the server can't do much at all with client-side prediction and offloading of tasks like drawing the inventory |
| 10:58 |
|
FileX joined #luanti |
| 10:59 |
luatic |
it's a difficult tradeoff between security, performance, and ease of implementation |
| 10:59 |
[MatrxMT] |
<leca> I see now |
| 11:10 |
|
gregon joined #luanti |
| 11:11 |
|
gregon left #luanti |
| 11:18 |
|
jaca122 joined #luanti |
| 11:32 |
|
Kimapr_ joined #luanti |
| 11:40 |
|
kimapr joined #luanti |
| 11:45 |
|
erstazi_ joined #luanti |
| 12:06 |
|
mrkubax10 joined #luanti |
| 12:30 |
cheapie |
I'm poking around at an old Core 2 Duo laptop here from about 2008, in a world with a bunch of mapgen mods installed, and... I must say, sitting around waiting on /emergeblocks is reminding me a lot of waiting on emerge back when I tried Gentoo on a VIA C3 :P |
| 12:30 |
cheapie |
(maybe that's why it's called that, since they both use a lot of CPU time and take a while...) |
| 12:51 |
|
SFENCE joined #luanti |
| 13:28 |
|
qqe joined #luanti |
| 13:53 |
|
kimapr joined #luanti |
| 14:00 |
|
SFENCE joined #luanti |
| 14:10 |
|
SFENCE joined #luanti |
| 14:20 |
|
SFENCE joined #luanti |
| 14:34 |
|
SFENCE joined #luanti |
| 14:50 |
|
silverwolf73828 joined #luanti |
| 15:05 |
|
SFENCE joined #luanti |
| 15:14 |
|
SFENCE joined #luanti |
| 15:25 |
|
SFENCE joined #luanti |
| 15:34 |
|
SFENCE joined #luanti |
| 15:58 |
|
kimapr joined #luanti |
| 16:15 |
|
Kimapr_ joined #luanti |
| 16:22 |
|
Kimapr_ joined #luanti |
| 16:29 |
|
kimapr joined #luanti |
| 16:48 |
|
Kimapr_ joined #luanti |
| 17:07 |
|
kimapr joined #luanti |
| 17:20 |
|
kimapr joined #luanti |
| 17:39 |
|
Kimapr_ joined #luanti |
| 17:42 |
|
SFENCE joined #luanti |
| 17:45 |
|
kimapr joined #luanti |
| 18:00 |
|
Kimapr_ joined #luanti |
| 18:21 |
|
kimapr joined #luanti |
| 18:26 |
|
kimapr joined #luanti |
| 18:27 |
|
mrkubax10 joined #luanti |
| 19:06 |
|
___nick___ joined #luanti |
| 19:10 |
|
___nick___ joined #luanti |
| 19:11 |
|
krisbug joined #luanti |
| 19:52 |
|
___nick___ joined #luanti |
| 19:56 |
|
SFENCE joined #luanti |
| 20:07 |
|
dabbill joined #luanti |
| 20:07 |
|
Warr1024 joined #luanti |
| 20:07 |
|
HumanG331 joined #luanti |
| 20:07 |
|
z812 joined #luanti |
| 20:07 |
|
basxto joined #luanti |
| 20:07 |
|
Cork joined #luanti |
| 20:07 |
|
comrad joined #luanti |
| 20:07 |
|
frostsnow joined #luanti |
| 20:07 |
|
mmuller joined #luanti |
| 20:07 |
|
nickodd joined #luanti |
| 20:07 |
|
rodrigo-morales joined #luanti |
| 20:09 |
|
swee joined #luanti |
| 20:09 |
|
chilledfrogs joined #luanti |
| 20:09 |
|
zleap joined #luanti |
| 20:09 |
|
runxiyu joined #luanti |
| 20:09 |
|
jonadab joined #luanti |
| 20:09 |
|
kamdard joined #luanti |
| 20:09 |
|
HamAdams joined #luanti |
| 20:09 |
|
calculon_ joined #luanti |
| 20:09 |
|
ShadowBot joined #luanti |
| 20:09 |
|
izzyb joined #luanti |
| 20:09 |
|
runxiyu_ joined #luanti |
| 20:09 |
|
Sokomine joined #luanti |
| 20:09 |
|
dv^_^ joined #luanti |
| 20:09 |
|
wsor4035 joined #luanti |
| 20:09 |
|
the_sea_peoples8 joined #luanti |
| 20:09 |
|
Thermoriax joined #luanti |
| 20:09 |
|
Unit193 joined #luanti |
| 20:09 |
|
dv^_^ joined #luanti |
| 20:09 |
|
the_sea_peoples joined #luanti |
| 20:10 |
|
thelounge656495 joined #luanti |
| 20:11 |
|
HumanG331 joined #luanti |
| 20:17 |
|
turtleman joined #luanti |
| 20:35 |
|
kamdard_ joined #luanti |
| 20:39 |
|
kamdard_ joined #luanti |
| 20:47 |
|
Sharpman joined #luanti |
| 21:31 |
|
SFENCE joined #luanti |
| 22:03 |
|
Kimapr_ joined #luanti |
| 22:06 |
|
CRISPR joined #luanti |
| 22:32 |
|
panwolfram joined #luanti |
| 23:04 |
|
SFENCE joined #luanti |
| 23:05 |
|
Eragon joined #luanti |
| 23:23 |
|
SFENCE joined #luanti |
| 23:32 |
|
dabbill joined #luanti |
| 23:46 |
|
SFENCE joined #luanti |
| 23:57 |
|
whosit joined #luanti |
| 23:57 |
MinetestBot |
whosit: May-07 06:48 UTC <sfan5> range_all has a different culling behavior than just setting a high view range |
