| Time |
Nick |
Message |
| 00:08 |
Copenhagen_Bram |
hello |
| 00:11 |
Copenhagen_Bram |
so, i'm talking to the dev of a game that's been hacked a lot, and i've been talking to them about how open source software isn't easier to hack and I've been using Minetest as an example of an open source game that you can't hack over multiplayer. It seems to me like minetest servers are well-protected against hacks, since the server is authoritative, there are accounts, and (I think) you can |
| 00:11 |
Copenhagen_Bram |
easily recruit mods without needing to trust them with too much power. |
| 00:13 |
kurtzmusch |
hack is such a broad term... |
| 00:14 |
|
erlehmann_ joined #minetest |
| 00:18 |
rubywarden |
Copenhagen_Bram: security is broken if you rely on the user not changing their software |
| 00:18 |
rubywarden |
the most important thing in security is not to trust clients |
| 00:19 |
Copenhagen_Bram |
Hmm |
| 00:19 |
Copenhagen_Bram |
What would be a good place to introduce this dev to people who know about this stuff and talk to them? |
| 00:19 |
Copenhagen_Bram |
Also, what are some other examples of open source games with secure servers? |
| 00:22 |
rubywarden |
all the good ones |
| 00:22 |
rubywarden |
good security should be the default, not the exception |
| 00:25 |
kurtzmusch |
Copenhagen_Bram: if its done in the client, it can be "hacked". the client should only render things. of course, the developer might choose to offload some stuff to clients, but thats a concious decision and they should know its prone to 'hacking' |
| 00:37 |
|
nri__ joined #minetest |
| 00:45 |
|
est31 joined #minetest |
| 00:53 |
|
nosbig joined #minetest |
| 00:53 |
nephele |
you can let the client do work, as long as you verify it |
| 01:08 |
Copenhagen_Bram |
How do you verify client work? |
| 01:10 |
Copenhagen_Bram |
Also: Why wasn't Minetest written in Godot? If I were to write a game, is there any reason I might want to write the game in Irrlicht instead of Godot? |
| 01:12 |
|
Quiark joined #minetest |
| 01:12 |
Quiark |
Copenhagen_Bram: I think a very good reason for the devs was that godot didn't exist at the time minetest was started |
| 01:12 |
Copenhagen_Bram |
Ahhhhhhhh that's a very good reason lmao |
| 01:13 |
Copenhagen_Bram |
I'm thinking about making a game of my own, in Godot. But I've never made a game before |
| 01:14 |
Copenhagen_Bram |
I wanna start rebuilding the game that the dev I mentioned made, in fact, and I want to make it open source, better written, and more secure this time. |
| 01:18 |
Quiark |
coding games is so hard :( |
| 01:18 |
erlehmann |
Copenhagen_Bram, which game is it? |
| 01:19 |
Copenhagen_Bram |
It's called "Fly Like a Bird 3" |
| 01:19 |
Copenhagen_Bram |
> coding games is so hard :( |
| 01:19 |
Copenhagen_Bram |
How hard? |
| 01:19 |
Copenhagen_Bram |
Tell me about it |
| 01:19 |
Copenhagen_Bram |
What am I getting myself into? |
| 01:19 |
Quiark |
do you still have all your hair? |
| 01:19 |
erlehmann |
Copenhagen_Bram, link? |
| 01:19 |
Copenhagen_Bram |
yes |
| 01:19 |
Quiark |
good for you |
| 01:20 |
Copenhagen_Bram |
are you saying i'll pull it out if I start coding a game? |
| 01:20 |
Copenhagen_Bram |
https://rw-developement.itch.io/flab3-classic |
| 01:20 |
Copenhagen_Bram |
warning, it's not open source |
| 01:21 |
Copenhagen_Bram |
or at the least it's all rights reserved I think? |
| 01:21 |
Copenhagen_Bram |
I've heard a few times that it's open source but all I've gotten is an SDK that doesn't have all the sounds the game has |
| 01:21 |
erlehmann |
open source is a corporate phrase that people invented because they thought “free software” sounds too communist, look it up mate |
| 01:21 |
erlehmann |
i see |
| 01:21 |
erlehmann |
so they are lying |
| 01:22 |
erlehmann |
> Requires Unity 4.7 and some work to get running. |
| 01:22 |
Copenhagen_Bram |
As I've warned, it's nonfree software |
| 01:22 |
Copenhagen_Bram |
I know the difference between open source and free software, I just have a bad habit of saying open source. Sorry. |
| 01:23 |
erlehmann |
btw |
| 01:23 |
erlehmann |
> General, player and modder discussion is on-topic. |
| 01:23 |
erlehmann |
this is not related to minetest, or is it? |
| 01:23 |
erlehmann |
do you want to make it a minetest subgame? |
| 01:23 |
erlehmann |
that would be interested |
| 01:23 |
Copenhagen_Bram |
Oh. where's the off-topic channel? |
| 01:23 |
erlehmann |
pixelbird |
| 01:23 |
erlehmann |
interesting |
| 01:23 |
Copenhagen_Bram |
Hmmmmmmmmmmm |
| 01:23 |
Copenhagen_Bram |
that might even be easier than doing it in godot |
| 01:23 |
Copenhagen_Bram |
I'll have to consider that |
| 01:23 |
Copenhagen_Bram |
But how do you make the player fly in minecraft |
| 01:24 |
Copenhagen_Bram |
And not fly like fly mode, like swimming in air |
| 01:24 |
Copenhagen_Bram |
Bird flight |
| 01:24 |
Copenhagen_Bram |
minetest* |
| 01:27 |
rubywarden |
Copenhagen_Bram: by performing some checks or doing the work again |
| 01:27 |
rubywarden |
whichever is easier |
| 01:27 |
rubywarden |
in response to: |
| 01:27 |
Copenhagen_Bram |
oh okay |
| 01:28 |
rubywarden |
<Copenhagen_Bram> How do you verify [the] client['s] work? |
| 01:28 |
rubywarden |
it's like checking maths homework |
| 01:28 |
rubywarden |
you can either make sure both sides of the equation balance, or you could do the homework yourself and then compared the answers |
| 01:29 |
rubywarden |
sometimes doing the homework yourself is easier than checking, but you can still benefit from the client doing some work because it can reduce perceived lag |
| 01:29 |
erlehmann |
Copenhagen_Bram, if your minecarts have a max speed and a client exceeds the speed limit, ban it |
| 01:29 |
Copenhagen_Bram |
ah |
| 01:29 |
erlehmann |
it's the old P ?= NP thing :D |
| 01:30 |
Copenhagen_Bram |
I'm seriously thinking about this flylikeabirdtest idea |
| 01:30 |
erlehmann |
minetest mods are easy |
| 01:30 |
erlehmann |
look at how boats and gliders work |
| 01:30 |
erlehmann |
you might be able to code something convincing in a few hundred lines at most |
| 01:30 |
Copenhagen_Bram |
Hmmmm |
| 01:30 |
erlehmann |
are there plane mods? |
| 01:31 |
Copenhagen_Bram |
First, can I build some chatrooms so I can get some people together to give me advice and moral support? |
| 01:32 |
Copenhagen_Bram |
You guys are already giving me useful tips and encouragement |
| 01:32 |
Copenhagen_Bram |
Then I can start working on some code |
| 01:32 |
erlehmann |
if you build it in minetest, just stay here? |
| 01:32 |
erlehmann |
also you can create a channel by joining an empty one |
| 01:33 |
Copenhagen_Bram |
Alright |
| 01:33 |
Copenhagen_Bram |
(I know a bit about creating channels on IRC. I've even used chanserv) |
| 01:33 |
erlehmann |
nice |
| 01:33 |
Copenhagen_Bram |
Any tips on using chanserv might be handy though |
| 01:34 |
Copenhagen_Bram |
And I think I'm gonna create an IRC channel, a Matrix channel, and a Discord server, and bridge them together |
| 01:36 |
Copenhagen_Bram |
Hmm... Or maybe I'll try to start coding first and just talk to you guys idk |
| 01:37 |
Copenhagen_Bram |
I'll start a github repo right now I guess |
| 01:40 |
|
acharon joined #minetest |
| 01:40 |
Quiark |
Copenhagen_Bram: thats' what all the game devs do - create website, create forum, set up discord matrix slack, twitter facebook page |
| 01:41 |
Quiark |
and then don't actually code any game because it's kinda hard |
| 01:42 |
erlehmann |
Copenhagen_Bram, discord is scummy |
| 01:42 |
erlehmann |
https://old.reddit.com/r/discordapp/comments/8nzb5d/why_is_discord_so_antiencryption/e001lr1/ |
| 01:42 |
erlehmann |
> It is our legal and moral obligation to do everything reasonable in our power to enforce the law and our Terms of Service. If we were not able to decrypt messages we effectively neuter our ability to do this. |
| 01:43 |
erlehmann |
what they are actually saying is “of course we'll deliver you homos to the thugs if the law requires us to do so” |
| 01:43 |
Copenhagen_Bram |
erlehmann: and that's why the room would be bridged |
| 01:43 |
erlehmann |
Copenhagen_Bram, but by that you reduce the security of the whole room to the security of discord, i.e. none |
| 01:43 |
erlehmann |
Copenhagen_Bram, i would say: set up nothing and code. |
| 01:44 |
erlehmann |
but that's just my advice, disregard it as you like |
| 01:44 |
Copenhagen_Bram |
I'm thinking the same thing too, maybe I should code something first |
| 01:44 |
erlehmann |
you can always setup something later |
| 01:44 |
Copenhagen_Bram |
BUT, what kind of security does a public room need? |
| 01:44 |
erlehmann |
also, if it's nice, you'll get a community |
| 01:44 |
Copenhagen_Bram |
yeah |
| 01:44 |
erlehmann |
whatever, i'm just saying discord is actively malicious and acting against user interests |
| 01:45 |
erlehmann |
(no one reasonably wants their stuff to be *deliberately* not encrypted) |
| 01:45 |
erlehmann |
but tool discussions like that are not for here i guess |
| 01:45 |
Copenhagen_Bram |
ah |
| 01:45 |
erlehmann |
Copenhagen_Bram, want any assistance on starting with mod development? i did it some years ago |
| 01:46 |
Copenhagen_Bram |
Sure! |
| 01:50 |
Copenhagen_Bram |
And yeah Discord sucks. But now I'm using it, because it's popular. But I haven't stopped using IRC or Matrix, and I've actually encouraged some Discord communities into bridging to Matrix |
| 01:51 |
|
scr267 joined #minetest |
| 01:52 |
Copenhagen_Bram |
Thanks to me there are Matrix rooms for some good open source games called Hyperrogue and Hideous Destructor that are bridged to their discord communities. I'm rather proud of that. Now people can take part in those communities without being forced to use proprietary software. Perhaps that is worth having to use proprietary software myself to do it. |
| 01:53 |
Copenhagen_Bram |
s/having//g |
| 01:53 |
erlehmann |
hyperrougue is nice |
| 01:53 |
erlehmann |
Copenhagen_Bram, have you played with the minetest mod glider? |
| 01:53 |
|
riff-IRC joined #minetest |
| 01:53 |
erlehmann |
Copenhagen_Bram, https://forum.minetest.net/viewtopic.php?t=21425 |
| 01:54 |
Copenhagen_Bram |
no, but that's gonna be one of the first things I do |
| 01:56 |
Copenhagen_Bram |
erlehmann: https://riot.im/app/#/room/#hyperrogue:chat.weho.st if you use Matrix |
| 01:56 |
Copenhagen_Bram |
that room was briidged to ##hyperrogue once but looks like it's not anymore, lemme see what I can do about getting it bridged there again |
| 01:57 |
erlehmann |
i try to avoid it, matrix is … let's say, designed following principles that elude me |
| 01:58 |
erlehmann |
also maybe talk more about minetest instead of chat services? |
| 01:58 |
Copenhagen_Bram |
alright |
| 02:12 |
Copenhagen_Bram |
May I PM you? |
| 02:32 |
|
NoctisLabs joined #minetest |
| 02:40 |
|
NoctisLabs joined #minetest |
| 02:46 |
|
galaxie joined #minetest |
| 02:54 |
|
nephele joined #minetest |
| 02:55 |
|
nosbig joined #minetest |
| 02:56 |
erlehmann |
Copenhagen_Bram, yes |
| 02:56 |
Copenhagen_Bram |
Yay |
| 02:59 |
erlehmann |
rubywarden, i suggested Copenhagen_Bram read your book :) https://rubenwardy.com/minetest_modding_book/en/index.html |
| 03:00 |
Copenhagen_Bram |
ooh there's a security section |
| 03:46 |
|
Sokomine joined #minetest |
| 04:45 |
|
Lunatrius` joined #minetest |
| 05:39 |
|
Hawk777 joined #minetest |
| 05:59 |
|
fluxflux joined #minetest |
| 06:16 |
|
Edgy1 joined #minetest |
| 06:21 |
|
galaxie joined #minetest |
| 06:45 |
|
kawaiipunk joined #minetest |
| 06:50 |
|
NoctisLabs joined #minetest |
| 06:50 |
|
erlehmann joined #minetest |
| 07:06 |
|
jluc joined #minetest |
| 07:08 |
|
mmuller joined #minetest |
| 07:17 |
|
supahati joined #minetest |
| 07:28 |
|
pyrollo joined #minetest |
| 07:30 |
|
CWz joined #minetest |
| 07:35 |
|
Norore joined #minetest |
| 08:24 |
|
celeron55_ joined #minetest |
| 08:24 |
|
TGS joined #minetest |
| 08:37 |
|
ShadowNinja joined #minetest |
| 09:05 |
|
erlehmann joined #minetest |
| 09:08 |
|
ektor_ joined #minetest |
| 09:46 |
|
Miniontoby joined #minetest |
| 09:52 |
|
HDMI_STECKDOSE joined #minetest |
| 10:12 |
|
nowhere_man joined #minetest |
| 10:19 |
|
proller joined #minetest |
| 10:31 |
|
Miniontoby joined #minetest |
| 11:04 |
|
zarcade_droid joined #minetest |
| 11:23 |
|
ektor joined #minetest |
| 11:23 |
|
Fixer joined #minetest |
| 11:28 |
|
proller joined #minetest |
| 11:39 |
|
proller joined #minetest |
| 12:02 |
|
proller joined #minetest |
| 12:05 |
|
riff-IRC joined #minetest |
| 12:13 |
|
Miniontoby joined #minetest |
| 12:16 |
|
logo4poo1 joined #minetest |
| 12:43 |
|
SwissalpS joined #minetest |
| 12:46 |
|
Flabb joined #minetest |
| 13:00 |
|
nree joined #minetest |
| 13:19 |
|
Volgaar joined #minetest |
| 13:40 |
|
Miniontoby joined #minetest |
| 13:44 |
|
kurtzmusch joined #minetest |
| 13:51 |
|
CrazyDave joined #minetest |
| 14:09 |
|
nate_1 joined #minetest |
| 14:09 |
|
exio4 joined #minetest |
| 14:19 |
|
nate_1 joined #minetest |
| 14:19 |
|
exio4 joined #minetest |
| 14:30 |
|
erlehmann joined #minetest |
| 15:08 |
|
luk3yx joined #minetest |
| 15:21 |
|
Taoki joined #minetest |
| 15:33 |
|
dabbill joined #minetest |
| 16:29 |
|
jluc joined #minetest |
| 17:05 |
Copenhagen_Bram |
Has anyone noticed weird behaviors with vehicles such as boats on slow computers? |
| 17:05 |
|
Ruslan1 joined #minetest |
| 17:06 |
Copenhagen_Bram |
When I'm on a boat in minetest, sometimes the boat locks up such that I can't steer and the boat still coasts in the direction it was going but keeps teleporting back to where it was earlier |
| 17:12 |
Calinou |
Copenhagen_Bram: that's due to server lag |
| 17:12 |
Calinou |
(or in singleplayer, the local server being unresponsive) |
| 17:12 |
Copenhagen_Bram |
Hmm |
| 17:12 |
Calinou |
vehicles are controlled server-side, there's no client-side prediction like in Minecraft |
| 17:12 |
Copenhagen_Bram |
How do I prevent my local server being unresponsive? |
| 17:13 |
Copenhagen_Bram |
Will there ever be client-side vehicle prediction? |
| 17:14 |
Calinou |
no, I don't think so, not until we have server-provided client-side mods at least |
| 17:14 |
Calinou |
(which likely won't happen anytime soon) |
| 17:15 |
nephele |
I sure hope they get here earlier rather than later ;) |
| 17:16 |
nephele |
There are already quite a lot of edge cases where prediction would be needed (direction of block placement or which block gets placed for some stuff for instance (specially if one wants minecraft like connecting stairs)) |
| 17:19 |
Copenhagen_Bram |
> minecraft like connecting stairs |
| 17:19 |
Copenhagen_Bram |
Doesn't mineclone 2 have that? |
| 17:19 |
nephele |
You can implement them fine, i was talking about prediction |
| 17:19 |
Copenhagen_Bram |
oh |
| 17:19 |
nephele |
i.e preventing the stair to look misplaced for N seconds (untill the server has send the actual placement) |
| 17:20 |
Copenhagen_Bram |
To be fair, I don't think I care much about something like that |
| 17:20 |
Copenhagen_Bram |
I don't mind if a stair looks misplaced for a second. I just mind boats being weird |
| 17:20 |
nephele |
Makes the game look less polished, prediction is important ;) |
| 17:21 |
Copenhagen_Bram |
I'm used to stuff not happening for a second, my internet has a high latency. |
| 17:21 |
nephele |
That certainly includes vehicles |
| 17:21 |
Copenhagen_Bram |
Yes. |
| 17:21 |
|
Rafi59_ joined #minetest |
| 17:21 |
nephele |
Copenhagen_Bram: Then you have played bad games in the past :/ |
| 17:21 |
Copenhagen_Bram |
what do you mean? |
| 17:21 |
nephele |
With good prediction you normally don't notice that you have a bad connection, except in edge cases |
| 17:22 |
|
Rafi59__ joined #minetest |
| 17:22 |
nephele |
I think minetest predicts movements too, so you don't get an input delay of PING :) |
| 17:23 |
Copenhagen_Bram |
Are you saying Minetest is a bad game then? Because when I connect to a server in Minetest, items and nodes sometimes temporarily revert after I've dug them or changed my inventory |
| 17:23 |
Copenhagen_Bram |
Due to my ~700 or so ms latency due to satellite internet |
| 17:23 |
nephele |
But then again any mod commands to change where you look etc. are all serverside, so they will always work horrible (except for maybe singleplayer) since you have per average ~300ms to move around freely till it resets your view |
| 17:23 |
Copenhagen_Bram |
I guess minetest has bad prediction |
| 17:23 |
Copenhagen_Bram |
hmm |
| 17:23 |
|
Rafi59 joined #minetest |
| 17:24 |
nephele |
Copenhagen_Bram: That is an item of prediction yes, and minetest is horrible there, if you go anywhere out of the compfort zone of engine predicted behaviour stuff starts to suck for prediction :) |
| 17:24 |
nephele |
Like, the actuall node placement in the stairs example is predicted, but the direction isn't |
| 17:24 |
Copenhagen_Bram |
hmm |
| 17:25 |
nephele |
also, it is predicted that you are always allowed to break nodes, or place them, while with prediction you could visually enforce protections clientside (even if that isn't securer it certainly looks much better) |
| 17:26 |
nephele |
predict where you can't place seeds without them poping into existance for a second etc. :) |
| 17:26 |
Copenhagen_Bram |
Is it less secure? Why can't the client visually enforcee protections while the server also enforces protections as it does before, at the same time? |
| 17:26 |
nephele |
No, it isn't less secure, it is the same ammount of secure |
| 17:27 |
Copenhagen_Bram |
Oh okay |
| 17:27 |
nephele |
The server would still enforce them, the client can't be trusted with that |
| 17:28 |
nephele |
Just that it now looks much better and easier to understand on the client |
| 17:28 |
Copenhagen_Bram |
indeed |
| 17:29 |
|
Krock joined #minetest |
| 17:29 |
nephele |
Krock: Hello there |
| 17:29 |
Krock |
hi nephele |
| 17:29 |
Copenhagen_Bram |
hey |
| 17:30 |
nephele |
Krock: How are server provided client mods going along, we were just talking about them for the purposes of prediction |
| 17:30 |
Copenhagen_Bram |
yes client side prediction would be neat |
| 17:31 |
Krock |
client-side prediction for node placement and inventory actions already exist |
| 17:31 |
Krock |
FYI |
| 17:31 |
Copenhagen_Bram |
ah, but they seem to time out or something after a while |
| 17:31 |
Copenhagen_Bram |
I have ~700ms latency |
| 17:31 |
nephele |
not really krock |
| 17:32 |
Krock |
there was a PR to send mods to the client and execute the code there - but it's not safe enough |
| 17:32 |
nephele |
assuming that everything suceeds is a bad prediction |
| 17:32 |
Copenhagen_Bram |
if I dig a node, it will often reappear temporarily |
| 17:32 |
nephele |
safe in what way? |
| 17:33 |
Krock |
client-side protection. so that the client does not run whatever the server tells it to do |
| 17:33 |
nephele |
That is what clientside mods are supposed to do though |
| 17:33 |
nephele |
be arbitrary code within the confines of minetest |
| 17:33 |
Krock |
they shall not wipe your filesystem |
| 17:33 |
nephele |
Yeah... obviously, but that holds true for any mod |
| 17:33 |
Krock |
"obviously". you asked for it |
| 17:33 |
nephele |
not just clientside ones |
| 17:34 |
nephele |
So i don't see how this situation would be worse |
| 17:34 |
nephele |
Why isn't minetest namespacing itself away anyhow then? |
| 17:34 |
Krock |
ofc it is. but the "secure" mod feature still has some gaps by using metatables |
| 17:34 |
nephele |
I woulnd't try to build a perfect sandbox in the game |
| 17:35 |
nephele |
especially when the OS gives you sandboxing caps too |
| 17:35 |
Krock |
yeah. just run Minetest inside a VM |
| 17:35 |
nephele |
You can use both certainly, but using just your homecooked ones seems stupid |
| 17:35 |
nephele |
No, not a VM |
| 17:35 |
Krock |
why? it's saf |
| 17:35 |
Krock |
+e |
| 17:35 |
nephele |
No, why? |
| 17:36 |
nephele |
I somehow doubt that they would be any less or more secure than containers conceptually |
| 17:36 |
nephele |
But their performance is muuuuuch worse |
| 17:36 |
nephele |
so really not a good tool for minetest |
| 17:37 |
nephele |
especially since you can't expect users to provide the protections themselves, you should namespace yourself, unconditionally |
| 17:37 |
Krock |
what's a namespace |
| 17:38 |
nephele |
Linues equivalent of solaris zones |
| 17:38 |
nephele |
or linuxes equivalent of FreeBSD jails |
| 17:40 |
Copenhagen_Bram |
i think i tried multiplayer doom once, there was no prediction |
| 17:40 |
Copenhagen_Bram |
i would click to shoot, but the gun would shoot a second later |
| 17:40 |
Calinou |
nephele: Minetest movement is client-side, it's not predicted |
| 17:40 |
Krock |
but you can play dom everywhere |
| 17:40 |
Calinou |
the server accepts your position and runs a few checks if anticheat is enabled |
| 17:40 |
nephele |
Calinou: Server verifying movements? sounds like prediction to me :) |
| 17:41 |
Calinou |
it's more prone to cheating, but also requires less resoruces on the server |
| 17:41 |
Calinou |
nephele: no, prediction is when the client only sends *inputs* to the server, not its position |
| 17:41 |
Krock |
it's not predicted server-side |
| 17:41 |
Krock |
that's what he probably would like to say |
| 17:41 |
nephele |
Calinou: Eh, both can be predicted |
| 17:41 |
Krock |
can but isn't |
| 17:41 |
nephele |
doesn't matter whether you predict that N is a valid position or predict that W is a valid move |
| 17:41 |
Calinou |
Copenhagen_Bram: Zandronum has an option for clientside puffs, should make it feel a bit better at least |
| 17:42 |
Calinou |
though I don't think weapons are unlagged like in modern games |
| 17:42 |
est31 |
nephele: collision checking happens on the client, not on the server |
| 17:42 |
Krock |
o/ est31 |
| 17:42 |
Krock |
yay |
| 17:42 |
est31 |
hi |
| 17:42 |
Krock |
hello :) |
| 17:42 |
est31 |
at least few years back |
| 17:42 |
est31 |
maybe devs have now added it since I went |
| 17:42 |
Krock |
still the same |
| 17:42 |
nephele |
est31: sounds like a bug to me ;) |
| 17:42 |
est31 |
sure is |
| 17:42 |
Krock |
there was a PR but it was so massive that it was never merged |
| 17:43 |
Copenhagen_Bram |
how long has minetest had prediction problems? |
| 17:43 |
nephele |
Krock keeps cheating on my server |
| 17:43 |
Krock |
now it's rotting in the Rebase Needed grave |
| 17:43 |
nephele |
:( |
| 17:43 |
est31 |
Krock: link? I'm curious |
| 17:43 |
est31 |
I had a PR for it myself |
| 17:43 |
Krock |
nephele: "keeps" assumes multiple times. It was once |
| 17:43 |
nephele |
You were on my server once |
| 17:43 |
nephele |
that is 100% qouta ;) |
| 17:44 |
Krock |
#7220 |
| 17:44 |
ShadowBot |
https://github.com/minetest/minetest/issues/7220 -- Move player movement code into Player class by bendeutsch |
| 17:44 |
Krock |
^ est31 |
| 17:44 |
est31 |
kk |
| 17:44 |
Krock |
actually that's only the preparation for it.. |
| 17:45 |
est31 |
yeah I see |
| 17:45 |
Krock |
even that didn't make it through the review in time |
| 17:45 |
est31 |
yeah my attempts were about doing collision checks only |
| 17:45 |
nephele |
Krock: what about my PR btw |
| 17:45 |
est31 |
that would forbid glitching through walls but still allow flying and stuff |
| 17:45 |
est31 |
advantage would be less lag |
| 17:45 |
Krock |
nephele: drowning in ridiculous controversy |
| 17:46 |
nephele |
still? |
| 17:46 |
Krock |
go check it yourself |
| 17:46 |
Copenhagen_Bram |
are there any open source multiplayer games that are still enjoyable with high latency? |
| 17:46 |
nephele |
Last comment was by dezzl i think? |
| 17:46 |
nephele |
Copenhagen_Bram: Yeah, most |
| 17:46 |
Krock |
Copenhagen_Bram: turn-based like chess |
| 17:46 |
Krock |
or tic-tac-toe |
| 17:46 |
nephele |
Garry's mod worked fine for moderately high latencies |
| 17:47 |
nephele |
only was really unberable with pings of over 1S |
| 17:47 |
nephele |
Don't play garrys mod though, it's bad now |
| 17:48 |
Krock |
est31: yes, collision detection is also a good idea. if it's possible to compensate the lag as a "don't calculate: dead zone in time" it would be pretty fail-safe |
| 17:49 |
est31 |
the issue my work had was that it checked the straight connection of latest point sent by the client and the last *verified* point stored on the server |
| 17:49 |
est31 |
the problem with that approach was that it broke when you walked past corners |
| 17:49 |
est31 |
in a laggy situation |
| 17:49 |
est31 |
and walking up steps would be even worse |
| 17:50 |
est31 |
so I tried adding a second check that based on: can you get from A to B at all= |
| 17:50 |
est31 |
that used the path finding algo exant in minetest |
| 17:50 |
est31 |
for that I even made the path finding algo faster |
| 17:50 |
est31 |
that patch actually made it to master I think |
| 17:51 |
Krock |
it still has its bugs, though. |
| 17:51 |
est31 |
but the path finding algo was still far too slow and, more importantly, far to inflexible |
| 17:51 |
est31 |
forgot the details. think it didn't work if you weren't directly on the ground |
| 17:52 |
Calinou |
the corner/step issue could be avoided by adding a tolerance |
| 17:52 |
Krock |
I also thought of using the Raycast functions to use the collision boxes rather than selection boxes. If old mid-player pos to new mid-player pos is clear - no problem |
| 17:52 |
Calinou |
this is even built into some games' physics models, like the CPMA mod in Quake 3 |
| 17:52 |
Calinou |
(it's called "skimming") |
| 17:52 |
Calinou |
in CPMA, if you run against a corner at an high speed while airborne, the game allows you to slide along it, then continue in your initial direction |
| 17:52 |
Calinou |
it can make navigating maps much less frustrating |
| 17:53 |
Calinou |
of course, you can only do it over a small distance |
| 17:53 |
est31 |
Krock: I think my work used raycasts in fact |
| 17:54 |
est31 |
https://github.com/minetest/minetest/commit/9aec701a4cb999b3d1eb097d4b01df0480b4ebd0 |
| 17:54 |
est31 |
this was the speedup commit |
| 17:55 |
nephele |
with tolerance i will clip through any costum meshes that should be just too small for player |
| 17:55 |
Krock |
uh.. I'd need to study that code in order to understand any of your changes |
| 17:57 |
est31 |
#3894 was the noclip pr |
| 17:57 |
ShadowBot |
https://github.com/minetest/minetest/issues/3894 -- Add serverside noclip enforcement by est31 |
| 17:58 |
est31 |
hmmm no it used collisionMoveSimple |
| 17:58 |
est31 |
which IIRC just distributes the path into 1000 small steps and checks each for a collision |
| 17:58 |
Krock |
yeah.. it's before juhdanad's Raycast addition |
| 17:59 |
Krock |
I though it was distance-based steps |
| 17:59 |
Krock |
speed * dtime |
| 17:59 |
est31 |
ok |
| 17:59 |
Krock |
or maybe that's only for the LocalPlayer calculations which have to be very previse |
| 17:59 |
est31 |
something like that |
| 17:59 |
Krock |
*precise |
| 18:00 |
Krock |
seems like you spent quite some time in writing that - at least according to the code graveyard there |
| 18:02 |
Krock |
but with the distance difference and time difference it would be (in theory) possible to limit the pathfinder distance to check |
| 18:08 |
est31 |
one of the issues was that in situations with server lag, distance and time would get large, and the pathfinder would have to check a larger distance |
| 18:08 |
est31 |
which then caused even more server lag |
| 18:09 |
est31 |
which contributed even more to the problem :p |
| 18:12 |
Calinou |
Copenhagen_Bram: Cube 2 games tolerate higher pings well due to their client-side physics and aiming |
| 18:13 |
Calinou |
this also makes them prone to cheating, though their small playerbase these days means there are less cheaters |
| 18:13 |
Calinou |
(Sauerbraten, Red Eclipse) |
| 18:15 |
Copenhagen_Bram |
ooh |
| 18:15 |
Krock |
est31: yay it's a wheel of death .. or something like that |
| 18:15 |
Copenhagen_Bram |
why can't a game have both client side physics and also server authority? |
| 18:16 |
Krock |
Copenhagen_Bram: because compensating the lags is very hard and calculating player movement quite expensive |
| 18:16 |
Krock |
even more if it has to be predicted |
| 18:16 |
Calinou |
most MMOs use client-side physics because of this ^ |
| 18:16 |
Calinou |
computing physics for 400 players isn't like computing physics for a single player |
| 18:16 |
Copenhagen_Bram |
yes |
| 18:17 |
Krock |
closed-source also helps them to have less cheaters |
| 18:17 |
Copenhagen_Bram |
hmm |
| 18:17 |
Calinou |
many people run Minecraft servers at MMO scale, so I guess it can also apply here |
| 18:17 |
Copenhagen_Bram |
does it really help?? |
| 18:18 |
Krock |
well yeah. at least it's surely harder to modify the client. there's CheatEngine, though. |
| 18:18 |
Copenhagen_Bram |
you know, i think open source software, because it's easier to cheat, makes it less interesting to cheat |
| 18:18 |
Krock |
cheating isn't just about flying and no-clipping |
| 18:18 |
Copenhagen_Bram |
I used to use cheat engine, but then I played open source games, and knowing that it would probably be easier to cheat them, didn't cheat them |
| 18:19 |
Krock |
It was very helpful to me to integrate formspec stuff into the client so that I could find and fix security holes |
| 18:19 |
Copenhagen_Bram |
also since cheating in singleplayer is usually well supported in such games |
| 18:19 |
Krock |
cheese steak jimmy's |
| 18:20 |
Copenhagen_Bram |
it's no fun to cheat when you can just type iddqd in the console |
| 18:20 |
Krock |
cheat codes in Minetest would be funny |
| 18:20 |
Copenhagen_Bram |
there are cheat codes in minetest |
| 18:20 |
Copenhagen_Bram |
it's calle grant singleplayer fly, fast, noclip |
| 18:20 |
Krock |
oh? can it spawn a Cobra Car? |
| 18:21 |
Krock |
like easter eggs |
| 18:21 |
Copenhagen_Bram |
oh |
| 18:21 |
Copenhagen_Bram |
nyan cats are easter eggs though aren't they? |
| 18:21 |
Krock |
yeah.. they were |
| 18:21 |
Copenhagen_Bram |
if you want an easter egg, make one of your mods do something spooky when the system date is on halloween |
| 18:22 |
Copenhagen_Bram |
maybe someone here could start adding an easter egg to their mod right now |
| 18:23 |
Calinou |
ILOVESCOTLAND ;) |
| 18:23 |
Copenhagen_Bram |
anyone can add an easter egg to minetest lol |
| 18:23 |
|
Rafi59 joined #minetest |
| 18:23 |
|
erlehmann joined #minetest |
| 18:23 |
Krock |
let's hardcode one |
| 18:24 |
Krock |
I'm pretty sure you can find some funny comments in the code, but regular players won't notice that |
| 18:24 |
nephele |
Eh, easter eggs arent hard to do |
| 18:24 |
nephele |
I removed all commands on my server and replaced them with uhm, other text |
| 18:24 |
rubywarden |
YASQUEEN |
| 18:24 |
nephele |
like /mods -> "You'd like to know that wouldn't you?" |
| 18:24 |
rubywarden |
Unlocks Nyan cats |
| 18:25 |
Krock |
minetest.registered_chatcommands = {} |
| 18:26 |
Krock |
rubywarden: your nickname is broken |
| 18:26 |
nephele |
You are allowed to access that table directly? mkay |
| 18:26 |
Krock |
yes sure. it's documented |
| 18:26 |
Copenhagen_Bram |
yeah isn't it supposed to be rubenwardy? |
| 18:26 |
Krock |
maybe it's cached so I can't tell for sure that this works |
| 18:26 |
Krock |
Copenhagen_Bram: no |
| 18:26 |
Copenhagen_Bram |
oh |
| 18:26 |
Krock |
it's supposed to be rabenwurdy |
| 18:26 |
Copenhagen_Bram |
lol |
| 18:26 |
nephele |
very funny Kroxy |
| 18:26 |
wubenrardy |
helloes |
| 18:27 |
Krock |
hi |
| 18:27 |
* wubenrardy |
creates a moddable game called Tinemest |
| 18:27 |
Krock |
the real estate |
| 18:27 |
nephele |
How come that gta v has a better lua modding model by now than minetest? ;) |
| 18:28 |
rubenwardy |
Ruby warden is what hmm used to call me |
| 18:28 |
rubenwardy |
nephele: info? |
| 18:28 |
|
twoelk joined #minetest |
| 18:28 |
nephele |
It does have server provided clientside mods :) |
| 18:28 |
nephele |
obviously isn't "rockstar official" but whatever, what info do you want? just where to get the mod or? |
| 18:30 |
Krock |
nephele: the reason is called money and full-time jobs |
| 18:31 |
nephele |
Krock: How so? it isn't made by rockstar |
| 18:31 |
|
erlehmann joined #minetest |
| 18:31 |
Elon_Sat0shi |
why hasn't anyone been able to make money with a free software game? |
| 18:31 |
nephele |
The game is, but the modding engine sure isn't and rockstar isnt happy about it either, they just want to sell more cashcards for gta online, so a different multiplayer to that isn't in their interest |
| 18:32 |
Krock |
Elon_Sat0shi: kinda yes.. thanks to Minetest I learned some bits about C++ |
| 18:32 |
Krock |
but not make money in terms of selling it or receiving donations |
| 18:32 |
Elon_Sat0shi |
has anyone tried selling access to their minetest server? (players must pay to play on the server or something) |
| 18:32 |
Elon_Sat0shi |
and donations, merch |
| 18:33 |
Elon_Sat0shi |
sell the latest version of the software on itch |
| 18:33 |
nephele |
Why would anyone pay for that? |
| 18:33 |
Krock |
I've seen some servers with donation links |
| 18:33 |
Elon_Sat0shi |
copyright the assets and sell them |
| 18:33 |
nephele |
Elon_Sat0shi: Then it isnt freee anymore |
| 18:33 |
nephele |
but then again, you can totally sell a game and just include the source code |
| 18:33 |
nephele |
it doesn't have to be foss to have it's sourcecode open |
| 18:34 |
Krock |
LGPL requires you to include a link to the source code no matter whether you copyrighted media or not |
| 18:34 |
nephele |
just put in the license "dont package this", and then distros won't package it, you can also just sell the assets, kind of like how openmw lets you play morrowind wherever, but you still need to have bought the game |
| 18:35 |
Krock |
and just selling an add-on pack might be difficult |
| 18:35 |
Elon_Sat0shi |
maybe even temporarily copyright the assets but make sure they become free after a few years or so |
| 18:36 |
nephele |
That is the "we promise it's foss soon!" mentallity, which people (obviously) avoid |
| 18:36 |
Elon_Sat0shi |
i know a trilogy of books where the first one is free but the next two are copyrighted, but promised to be in public domain by 2030 or something |
| 18:37 |
rubenwardy |
Elon_Sat0shi: that's commonly called the ransom monetisation model |
| 18:37 |
Elon_Sat0shi |
and why is it avoided? |
| 18:37 |
Elon_Sat0shi |
okay, what's wrong with it? |
| 18:37 |
Elon_Sat0shi |
isn't it better than something that's copyrighted forever? |
| 18:38 |
rubenwardy |
let's put it this way - Minecraft had a similar promise |
| 18:38 |
Elon_Sat0shi |
so it's bad when the promise isn't kept? |
| 18:38 |
Elon_Sat0shi |
is there a way to legally guarantee that something will be free eventually? |
| 18:39 |
Krock |
sign a contract with all users |
| 18:39 |
Elon_Sat0shi |
hmm |
| 18:39 |
rubenwardy |
it's not the free software way of doing things |
| 18:40 |
rubenwardy |
there's nothing that stops you from making your own Minetest game and making it proprietary |
| 18:40 |
Elon_Sat0shi |
what's the free software of making a profit? |
| 18:40 |
rubenwardy |
or under that release in Xyrs policy |
| 18:40 |
Elon_Sat0shi |
free software way* |
| 18:40 |
nephele |
Elon_Sat0shi: apache |
| 18:40 |
nephele |
:) |
| 18:40 |
rubenwardy |
Redhat, Canonical |
| 18:40 |
rubenwardy |
they provide services and support |
| 18:40 |
erlehmann |
> <rubenwardy> there's nothing that stops you from making your own Minetest game and making it proprietary |
| 18:40 |
erlehmann |
i guess the licensing would do that? |
| 18:40 |
rubenwardy |
nope |
| 18:40 |
Krock |
erlehmann: it's MIT |
| 18:41 |
Elon_Sat0shi |
hmm |
| 18:41 |
rubenwardy |
it's any license you want |
| 18:41 |
rubenwardy |
the minetest engine is LGPLv2.1 |
| 18:41 |
rubenwardy |
but that doesn't affect the Lua scripts |
| 18:41 |
Elon_Sat0shi |
i've noticed there are sites that sell minetest servers and support |
| 18:41 |
rubenwardy |
yes |
| 18:41 |
rubenwardy |
that's an example of the services monetisation model |
| 18:41 |
erlehmann |
<rubenwardy> the minetest engine is LGPLv2.1 |
| 18:41 |
erlehmann |
that's what i meant |
| 18:41 |
rubenwardy |
you can still make a proprietary game |
| 18:41 |
erlehmann |
i remember being against licensing the engine under a more permissive license |
| 18:42 |
erlehmann |
because then proprietary games would give nothing back |
| 18:42 |
Elon_Sat0shi |
where do you get all these monetisation models? |
| 18:42 |
rubenwardy |
you just need to release the source code for the engine |
| 18:42 |
nephele |
erlehmann: just like how the gplv2 protected linux from beeing used commercially |
| 18:42 |
rubenwardy |
^ |
| 18:42 |
nephele |
no wait... actually that never happened :) |
| 18:42 |
erlehmann |
nephele, commercially is not the point. |
| 18:42 |
Krock |
RHEL? |
| 18:42 |
erlehmann |
nephele, giving back is the point |
| 18:42 |
rubenwardy |
well, free software doesn't disallow commercial use |
| 18:42 |
rubenwardy |
the GPU just requires you to give back |
| 18:42 |
nephele |
erlehmann: AOSP is giving back, shitty vendors aren't |
| 18:43 |
erlehmann |
the “price” for using linux is having to give your improvements to the customers |
| 18:43 |
nephele |
but ganooh/linux doesn't care about what AOSP is giving back at all |
| 18:43 |
erlehmann |
IMO, GPLv2 only was a mistake |
| 18:43 |
nephele |
Nope, you can just not put stuff into kernelspace |
| 18:43 |
nephele |
erlehmann: how so? |
| 18:44 |
rubenwardy |
it's the same with Minetest - if you modify the engine, then you need to give back. But you can create proprietary work on top of the engine/kernel |
| 18:45 |
rubenwardy |
in theory, at least |
| 18:45 |
rubenwardy |
in practice no one has the time to enforce licenses |
| 18:45 |
nephele |
Just look at what a shitshow the busybox gpl enforcement lawsuits were |
| 18:46 |
nephele |
countless vendors sued, now what is the number of code lines busybox gained from released source code? |
| 18:46 |
nephele |
Precisely 0 :D |
| 18:46 |
nephele |
Turns out vendors write shitty code nobody wants |
| 18:51 |
* Krock |
looks at the BIOS code writers |
| 18:53 |
erlehmann |
nephele, GPLv3+ prevents two ways of saying “you can HAVE the code, but you can't RUN it” – tivoization and patent lawsuits. first is creating code-signing mechanisms that allow only the vendor code to run. second is giving other the code, but threatening legal action if they use it. both prevents the goal of the license, giving others the ability modify the code and run their own version. |
| 18:54 |
rubenwardy |
!title https://www.youtube.com/watch?v=PaKIZ7gJlRU |
| 18:54 |
MinetestBot |
rubenwardy: Linus Torvalds says GPL v3 violates everything that GPLv2 stood for - YouTube |
| 18:54 |
erlehmann |
lel |
| 18:55 |
erlehmann |
fun thing that corporate lawyers (that AFAIK were consulted by FSF) have a different opinion |
| 18:56 |
erlehmann |
rubenwardy, IIRC the whole “home routers for cheap” stuff (openWRT etc.) came because someone sued linksys |
| 18:57 |
erlehmann |
rubenwardy, http://ebb.org/bkuhn/talks/OpenWRT-2016/gpl.html |
| 18:57 |
erlehmann |
> OpenWrt may be the most successful historical example of GPL enforcement. |
| 18:57 |
erlehmann |
> The project began as the outcome of GPL enforcement action. |
| 18:57 |
erlehmann |
such cases |
| 18:57 |
erlehmann |
> SVN check-in r1 of the OpenWrt project was the actual source code we reviewed in our GPL compliance efforts. |
| 18:58 |
erlehmann |
rubenwardy, kinda goes against the “in practice no one has the time to enforce licenses” |
| 18:58 |
rubenwardy |
I was referring to Minetest |
| 18:58 |
erlehmann |
there is even the which has a lot of this stuff https://en.wikipedia.org/wiki/Software_Freedom_Law_Center#BusyBox_litigation |
| 18:58 |
erlehmann |
oh okay |
| 18:59 |
erlehmann |
rubenwardy, is there still code from me in minetest and a possible defendant? i may be willing to sue them for breach of license. |
| 18:59 |
rubenwardy |
well, not even that far - but even DMCA s |
| 18:59 |
erlehmann |
DMCA? |
| 18:59 |
rubenwardy |
also, good luck sueing someone in china |
| 18:59 |
rubenwardy |
Digital Millenium Copyright Act is a framework for getting content providers to take down content |
| 18:59 |
erlehmann |
oh, but i don't want it to get taken down |
| 19:00 |
rubenwardy |
ie: you file a notice with Google Play to take down the apps of a dev |
| 19:00 |
erlehmann |
the best outcome is the one cisco had, making them giving up the source code |
| 19:00 |
erlehmann |
much preferably to being taken down on all accounts |
| 19:00 |
erlehmann |
devs get more source code, users keep the app |
| 19:00 |
rubenwardy |
you want people to give the source code without being sued |
| 19:00 |
rubenwardy |
DMCAs are the better way than sueing |
| 19:00 |
rubenwardy |
https://en.wikipedia.org/wiki/Digital_Millennium_Copyright_Act |
| 19:01 |
rubenwardy |
you could file the notice with the app dev first |
| 19:01 |
erlehmann |
reminds me of that minecraft (minetest?) streamer who always talked about “demonetizing” mobs |
| 19:01 |
erlehmann |
instead of killing them |
| 19:01 |
Krock |
but he killed them? |
| 19:02 |
erlehmann |
yes |
| 19:04 |
nephele |
erlehmann: tivoization prevention doesn't help anyone, it just means they run mit code instead |
| 19:05 |
nephele |
and dual licensing doesn't help anyone since people will just say "yeah but i used the gpl2 variant" |
| 19:05 |
nephele |
and gpl2 is incompatible with gpl3 and incompatible with dual gpl2+3 so, sucks pretty much |
| 19:06 |
nephele |
>also, good luck sueing someone in china |
| 19:06 |
erlehmann |
nephele, if there is MIT code available that does the same thing, that is the case. but there often isn't. |
| 19:06 |
nephele |
Seeing as china never signed the internation thingy for copyright the chances of that suceeding are pretty much 0 anyway :) |
| 19:06 |
rubenwardy |
yeah |
| 19:06 |
nephele |
erlehmann: not really, the gpl is deiing |
| 19:07 |
erlehmann |
nephele, ok, so where is the MIT minetest engine? |
| 19:07 |
nephele |
there are only very nieshe things where you can't get an equivalent MIT thing |
| 19:07 |
nephele |
erlehmann: There are other similar games, and games with a better lua api |
| 19:07 |
nephele |
that /dont/ have the GPL |
| 19:08 |
nephele |
Like gmod, even though their eula is now pretty hostile i still present them as an example, could argue the same for the unity engine, the unreal engine, the source engine etc. etc. |
| 19:08 |
erlehmann |
the readline library is GPL and AFAIK that lead to some software being released because they wanted readline. if you frame it in having to release code as a payment, i think it makes sense to understand that “well, if you want a price, i'll go somewhere else” does not always apply |
| 19:08 |
rubenwardy |
apache: https://github.com/MovingBlocks/Terasology |
| 19:09 |
erlehmann |
nephele, for games that is pretty interesting actually. quake mods are free software nowadays because quake, quake2, quake3 were released as free software. half life mods, not so much. |
| 19:09 |
erlehmann |
so while “natural selection” is a nicer game than “tremulous” IMO, i can't easily play natural selection now. |
| 19:10 |
erlehmann |
(first is a half life 1 mod, second a quake 2 (?) mod) |
| 19:10 |
erlehmann |
similar earth 2150 and warzone 2100. first was never released as free software, second was. first is the nicer game. now, after 20 years, the second still has players. the first one … well, if you make a LAN party. |
| 19:10 |
nephele |
mods usually adopt similar licenses as their donor games |
| 19:11 |
erlehmann |
in the long run, proprietary engines sadly vanish. |
| 19:11 |
erlehmann |
or get reimplemented as free software (openttd being one example) |
| 19:11 |
nephele |
the source engine is open source, but that is rather irrelevant now |
| 19:11 |
nephele |
Openmw is also a good example, of how the engine may die |
| 19:12 |
nephele |
also, quake{2,3} wasn't free software no? it was commercial software with source code afaik |
| 19:12 |
erlehmann |
the engine was AFAIK? |
| 19:13 |
nephele |
not at the start |
| 19:13 |
nephele |
And there are many reimplementation of it not based on the original |
| 19:14 |
nephele |
also, goldsrc /was/ based on the quake engine afaik |
| 19:16 |
erlehmann |
nephele, can you give me a pointer to the open source source engine? |
| 19:16 |
nephele |
https://github.com/ValveSoftware/source-sdk-2013 |
| 19:16 |
erlehmann |
also, does it mean being able to play HL2 with fully free software? |
| 19:16 |
erlehmann |
nephele, which license, MIT? |
| 19:17 |
nephele |
> SOURCE 1 SDK LICENSE |
| 19:17 |
erlehmann |
nephele, that is not an open source license |
| 19:17 |
nephele |
The source code is there, people have used it |
| 19:17 |
nephele |
Doesn't matter whether you like the license or not |
| 19:18 |
erlehmann |
nephele, yeah, but “open source” has a very specific meaning and since it is an invented term, it had that meaning from the start. https://opensource.org/osd |
| 19:18 |
nephele |
Not correct |
| 19:18 |
nephele |
Free software is the invented term |
| 19:18 |
erlehmann |
o.0? |
| 19:18 |
rubenwardy |
open source is also the invented term |
| 19:18 |
nephele |
all software was open source |
| 19:19 |
nephele |
propritare software came later |
| 19:19 |
erlehmann |
to put it more clearly: if i only eat animals that eat plants, i may not be eating vegan. |
| 19:19 |
erlehmann |
whatever i think the terms mean. |
| 19:19 |
rubenwardy |
open source is almost identical to free software, except for the motivations: . https://opensource.org/osd |
| 19:19 |
nephele |
erlehmann: did you read the license? |
| 19:19 |
erlehmann |
nephele, yes. what makes you think i did not? |
| 19:19 |
nephele |
It's basically you can make games from it, but you can't sell it |
| 19:19 |
rubenwardy |
source-available is the name for licenses which aren't open or free software, but you are allowed access to the source code |
| 19:19 |
erlehmann |
which is not compatible with point 6 of the Open Source Definition |
| 19:20 |
nephele |
source-available is also an invented term |
| 19:20 |
erlehmann |
> The license must not restrict anyone from making use of the program in a specific field of endeavor. For example, it may not restrict the program from being used in a business, or from being used for genetic research. |
| 19:20 |
rubenwardy |
all terms are invented |
| 19:20 |
nephele |
But just like MiB GiB etc, just because a new term exists doesn't mean people are going to start using it |
| 19:20 |
erlehmann |
nephele, https://en.wikipedia.org/wiki/Open_source#Origins |
| 19:20 |
nephele |
erlehmann: you really won't like the BSD-no-nuclear license |
| 19:20 |
nephele |
I know better than to trust wikipedia on anything IT related ;) |
| 19:20 |
erlehmann |
> Peterson suggested "open source" at a meeting[4] held at Palo Alto, California, in reaction to Netscape's announcement in January 1998 of a source code release for Navigator. Linus Torvalds gave his support the following day, and Phil Hughes backed the term in Linux Journal. |
| 19:21 |
erlehmann |
i gather the term literally did not exist before? |
| 19:21 |
erlehmann |
nephele, BSD-no-nuclear is not a free software or open source license probably? |
| 19:21 |
erlehmann |
what if nuclear war is my field of endeavour |
| 19:21 |
erlehmann |
(half-joking) |
| 19:21 |
nephele |
It's not a term, it's an adjective and a term |
| 19:21 |
nephele |
erlehmann: well, the license would forbid you from using it |
| 19:22 |
nephele |
But regardless, i don't have a problem with the license of the valve sdk, regardless of whether OSI like it or not |
| 19:22 |
nephele |
I have problems with licenses that OSI aproves too ;) |
| 19:22 |
erlehmann |
nephele, yeah, just don't claim it is an open source license |
| 19:22 |
nephele |
one doesn't always have to agree fully |
| 19:22 |
erlehmann |
i have a problem with it because it would make it impossible, e.g. to include it on a debian DVD |
| 19:22 |
nephele |
erlehmann: No, i claimed the engine is open source, not the license |
| 19:22 |
nephele |
No, it would not |
| 19:23 |
nephele |
There is nothing preventing debian from doing that, other than their own policies |
| 19:23 |
erlehmann |
nephele, did you mean “the engine is open source” in any other way than “the source code of the engine is licensed under an open source license” ? |
| 19:23 |
nephele |
erlehmann: nowhere did i say "OSI conform" |
| 19:23 |
nephele |
I don't have to agree with their definition, so i don't |
| 19:24 |
erlehmann |
nephele, so if i claimed a photo i bought from a store was “photoshopped” you would have no problem with my use of language? |
| 19:24 |
nephele |
Yes, photoshopped has become an adjective to reffer to forgery, not specific to photoshop anymore |
| 19:25 |
nephele |
just like how googling no longer reffers to google.com |
| 19:25 |
nephele |
language changes all the time :) |
| 19:26 |
erlehmann |
nephele, words have meaning. if the people around you assume a different meaning because you are using a term differently, it makes the discussion harder due to misunderstandings. |
| 19:26 |
nephele |
erlehmann: linquistics is my hobby, i know this, and so does everyone else |
| 19:27 |
nephele |
but words have meanings only in a specific context |
| 19:27 |
erlehmann |
our context was software licenses |
| 19:27 |
nephele |
Yes |
| 19:27 |
nephele |
and "OSI compliant" is a valid term |
| 19:27 |
nephele |
and i deliberately did not use it |
| 19:27 |
nephele |
actually the context is a bit more than "software licenses" though |
| 19:27 |
rubenwardy |
words and phrases have meanings |
| 19:27 |
erlehmann |
i hereby humbly suggest instead of abbreviating stuff with terms that are commonly understood you spell it out in future, as in “the source code for the engine is available online” (which would AFAIK also be true for some unreal engine). |
| 19:28 |
erlehmann |
that concludes my part of the discussion about the term “open source” for now. |
| 19:28 |
erlehmann |
rubenwardy, we live in a society!!!111 ;) |
| 19:28 |
erlehmann |
words have meanings and yet you use them, you hypocrites! :P |
| 19:28 |
nephele |
erlehmann: i did not abreviate, i wrote what i ment, and ment what i said |
| 19:31 |
|
ensonic joined #minetest |
| 19:32 |
|
karamel joined #minetest |
| 19:34 |
|
Volgaar joined #minetest |
| 19:57 |
|
HDMI_STECKDOSE joined #minetest |
| 20:01 |
|
NoctisLabs joined #minetest |
| 20:15 |
|
sec^nd joined #minetest |
| 20:39 |
MinetestBot |
[git] paramat -> minetest/minetest_game: README: Use correct link, make link clickable 5b21d1d https://git.io/JeDV4 (2019-12-04T20:38:01Z) |
| 20:48 |
|
FreeFull joined #minetest |
| 20:48 |
|
fruitsnack joined #minetest |
| 20:49 |
fruitsnack |
sfan5: hello, what ndk version was used to compile minetest for android? |
| 20:49 |
sfan5 |
no idea |
| 20:49 |
sfan5 |
probably the latest |
| 20:50 |
fruitsnack |
Who's compiling it? |
| 20:50 |
fruitsnack |
For the "releases" section on github |
| 20:59 |
|
sec^nd joined #minetest |
| 21:09 |
Elon_Sat0shi |
ugh |
| 21:09 |
Elon_Sat0shi |
i thought i would start working on that mod today |
| 21:10 |
Elon_Sat0shi |
but right now, i'm too depressed/stressed/burnt out to even take a bath |
| 21:18 |
|
Pie-jacker875 joined #minetest |
| 21:23 |
|
FreeFull joined #minetest |
| 21:26 |
|
twoelk left #minetest |
| 21:32 |
Elon_Sat0shi |
ugh |
| 22:20 |
|
Hirato joined #minetest |
| 22:49 |
|
proller joined #minetest |
| 22:50 |
|
nowhere_man joined #minetest |
| 23:03 |
|
Ruslan1 joined #minetest |
| 23:18 |
|
Unit193 joined #minetest |
| 23:19 |
|
galaxie joined #minetest |
| 23:19 |
|
Fusl joined #minetest |
| 23:19 |
|
Pie-jacker875 joined #minetest |
| 23:21 |
|
scr267_ joined #minetest |
| 23:22 |
scr267_ |
Hi everyone, my server ran into the same error that p_gimeno experienced back on June 6th (saw that on the IRC logs): minetest/src/serverenvironment.cpp:e5: std::string LBMManager::createIntroductionTimesString(): A fatal error occurred: attempted to query on non fully set up LBMManager |
| 23:22 |
scr267_ |
Just wondering if anyone has any idea what might be the cause? |
| 23:29 |
scr267_ |
FYI, both minetest and minetest_game were pulled from branch stable-5 |
| 23:29 |
scr267_ |
I'm going to try tag 5.0.1 just to see if it makes any difference. |
| 23:56 |
rubenwardy |
Weird |
| 23:58 |
|
milkt joined #minetest |
