| Time |
Nick |
Message |
| 04:00 |
|
MTDiscord joined #minetest-dev |
| 05:55 |
|
calcul0n joined #minetest-dev |
| 06:17 |
|
independent56 joined #minetest-dev |
| 06:19 |
|
independent56 joined #minetest-dev |
| 07:08 |
|
independent56 joined #minetest-dev |
| 07:13 |
|
independent56 joined #minetest-dev |
| 07:40 |
nrz |
if there is donations i think this it should go to the expenses in terms of OPEX, ie. forum & site hosting, maybe some coredev tooling (SaaS solutions ? software licenses to dev ?) and after let's see what can we do. |
| 08:54 |
|
independent56 joined #minetest-dev |
| 08:57 |
|
independent56 joined #minetest-dev |
| 09:09 |
|
independent56 joined #minetest-dev |
| 09:11 |
|
Fixer joined #minetest-dev |
| 10:36 |
|
HuguesRoss7 joined #minetest-dev |
| 11:14 |
|
independent56 joined #minetest-dev |
| 11:18 |
|
behalebabo joined #minetest-dev |
| 11:27 |
|
independent56 joined #minetest-dev |
| 11:51 |
|
Warr1024 joined #minetest-dev |
| 12:15 |
schwarzwald[m] |
Looking for some guidance on how to verify the correctness of the IrrlichtMt mesh buffer handling. I can also skip it for now and work on the next report item though. |
| 12:33 |
|
independent56 joined #minetest-dev |
| 12:39 |
|
independent56 joined #minetest-dev |
| 13:00 |
rubenwardy |
Nrz: you're late and missing the point, as usual |
| 13:15 |
MTDiscord |
<luatic> rubenwardy: and you're impolite, as unusual |
| 13:15 |
rubenwardy |
True |
| 13:17 |
rubenwardy |
nrz: we're in a position where we already have enough to cover expenses. If there's particular other things that are needed, then they can be added. We're already in "after that" |
| 13:18 |
rubenwardy |
Oh, one thing I've thought of: cloud Android testing. There's online servers that allow you to test your app on thousands of different devices, I'm not sure how it works but that would be useful |
| 13:18 |
rubenwardy |
*services |
| 13:35 |
Zughy[m] |
well, also money for bug bounties and hiring some devs wouldn't be bad I guess |
| 13:35 |
Zughy[m] |
more money = faster development pace |
| 13:41 |
celeron55 |
if someone has the time and is ready to eg. use a month of such android testing efficiently, just tell me. as long as i'm funneling donations towards a service and not a person, i can do it |
| 13:42 |
celeron55 |
the problem in what nrz said is the "after let's see what can we do". how do we see that? |
| 13:43 |
celeron55 |
we're already there, and "what we can do" is balatantly unclear |
| 13:43 |
celeron55 |
unless the answer is "nothing" |
| 13:43 |
|
appguru joined #minetest-dev |
| 13:43 |
celeron55 |
then it's very clear |
| 13:56 |
MTDiscord |
<Warr1024> It sounds to me like the theory is (1) make it easier for people to donate above and beyond expenses, (2) have extra money, then (3) think of things you can spend it on that would benefit the project. I'm skeptical [a] that having more money WOULD cause more things to spend it on to surface, as often volunteers spend above and beyond already, and [b] that it would actually be a good, or even sane, idea to go LOOKING for expenses. |
| 13:57 |
MTDiscord |
<Warr1024> On the flip side, faster hosting for stuff like the forums would be nice, but I dunno how far out on a limb that could put people financially. |
| 14:10 |
celeron55 |
yes i (and i think others too) want the things to spend on first, not the money first |
| 14:11 |
celeron55 |
it's borderline illegal to collect funds when you don't have a non-profit organization and don't have anything to directly spend them on |
| 14:12 |
celeron55 |
unless it's implied that it's payment for work |
| 14:12 |
celeron55 |
of course depends on the country |
| 14:12 |
celeron55 |
anyway, i (and i think others too) want the things to spend on first, not the money first |
| 14:13 |
celeron55 |
i could certainly move the forum on a faster host, and i don't even need donations right now for that, i have plenty of buffer for that. but doing that makes it really difficult to scale back in the future |
| 14:13 |
celeron55 |
so it needs to be 100% certain i don't have to downscale before i do it |
| 14:15 |
celeron55 |
for me it's way more comfortable to have the buffer and have a slightly slow forum, than to have an as-fast-as-possible forum and slim buffer, and having to count the pennies and beg for money |
| 14:15 |
MTDiscord |
<Warr1024> Yeah, if you could have a subscription-based system instead of one time donations then that would make it easier to predict future cashflow, but I'm not sure I like the framing around tools like patreon where you're expected to give exclusive perks to patrons and it becomes a tit-for-tat kind of arrangement. |
| 14:17 |
|
independent56 joined #minetest-dev |
| 14:17 |
celeron55 |
personally i'm just really happy i'm so far from having to say "please donate or i'll shut down the forum because i'm not here to pay it myself" |
| 14:18 |
celeron55 |
i don't have further monetary goals regarding to MT |
| 14:18 |
celeron55 |
it's exactly where i want to be |
| 14:20 |
celeron55 |
and i'm very happy the relatively buried donation button gives me this amount of funds. it's an absolute luxury not having to beg |
| 14:21 |
celeron55 |
as i said, if someone wants to use a paid service for some development task, i can help. even if you're doing it just to learn the service it's fine as long as it benefits MT |
| 14:22 |
celeron55 |
other than these uses, i can't really justify any spending |
| 14:37 |
nrz |
rubenwardy, as usual yes, is this a criticism ? i am busy and i talk asynchronously, does than mean my idea is not right ? :p |
| 14:38 |
nrz |
note #12462 always needs love :)p |
| 14:38 |
ShadowBot |
https://github.com/minetest/minetest/issues/12462 -- fix: various unitinialized or badly initialized variables by nerzhul |
| 15:07 |
|
independent56 joined #minetest-dev |
| 15:19 |
|
independent56 joined #minetest-dev |
| 15:33 |
|
independent56 joined #minetest-dev |
| 16:22 |
|
Soni joined #minetest-dev |
| 16:27 |
Soni |
can we have server access API? (ability to bind server to socket, including unix socket, from mods) for "Open to LAN"/"Open to Friend Code" functionality? (for mods with system privileges ofc) |
| 16:29 |
MTDiscord |
<LandarVargan> You can bind to a unix socket with mods I think |
| 16:33 |
Soni |
not the same as binding the (internal) server to a socket |
| 16:36 |
rubenwardy |
no, that doesn't have much chance for happening |
| 16:36 |
rubenwardy |
binding to sockets is a server owner thing |
| 16:37 |
rubenwardy |
See `bind_address` in minetest.conf |
| 16:37 |
rubenwardy |
https://github.com/minetest/minetest/blob/5.5.1/minetest.conf.example#L1227 |
| 16:39 |
Soni |
then why does singleplayer allow binding to sockets - but only at startup? |
| 16:40 |
Krock |
huh? singleplayer creates an internal loopback |
| 16:42 |
Soni |
can... can one use that internal loopback with a proxy? |
| 16:42 |
Krock |
why would you do that? singleplayer is called singleplayer for a reason |
| 16:43 |
Krock |
what you need is to host a local server by either running minetestserver or ticking the "Host server" checkbox in the main menu |
| 16:43 |
Krock |
and in that case the bind_address setting becomes relevant. |
| 16:44 |
Soni |
for "open to LAN" functionality |
| 16:44 |
Soni |
or "open to proxy" |
| 16:44 |
MTDiscord |
<ROllerozxa> I assume Soni wants something like Minecraft's "Open to LAN" option where you can make an already existing singleplayer server open for others to join |
| 16:44 |
Soni |
or "open to tailscale" |
| 16:45 |
Krock |
I don't understand what that's supposed to be. Minetest knows the concept of regular servers. whether or not socket files are accepted is up to the porting implementation that I'm not familiar with |
| 16:45 |
MTDiscord |
<ROllerozxa> since, well, it's still a server that's running on a random port, people could theoretically join but they will be immediately kicked with a message that the server is in singleplayer mode |
| 16:46 |
Krock |
just set a default server password |
| 16:46 |
Krock |
people will have to log in with that to register a new account |
| 16:47 |
Krock |
Minetest does not have a "hybrid singleplayer" mode. either it's broadly accessible or not. |
| 16:48 |
Soni |
sure, so, once again: <Soni> can we have server access API? (ability to bind server to socket, including unix socket, from mods) for "Open to LAN"/"Open to Friend Code" functionality? (for mods with system privileges ofc) |
| 16:48 |
Soni |
"it doesn't have a hybrid mode" isn't an answer to "can we have a hybrid mode" |
| 16:49 |
Krock |
mods cannot create a server instance. the server loads mods instead. |
| 16:49 |
Krock |
you'd need to mod the main menu to overwrite the default behaviour |
| 16:49 |
Krock |
"can we have a hybrid mode" is not a question but a feature request |
| 16:50 |
rubenwardy |
the problem with opening up a singleplayer instance is that the host is called "singleplayer" which will break things |
| 16:50 |
rubenwardy |
the solution may be to get rid of this behaviour, and require players to enter a player name |
| 16:50 |
Soni |
so we're better off emulating the minetest protocol with fake entities? |
| 16:51 |
rubenwardy |
but there's not a huge amount of benefit over just clicking Host server |
| 16:51 |
Krock |
you could create a new issue on Github to note down your feature request |
| 16:51 |
rubenwardy |
I'm not a fan of the singleplayer user personally, it's weird to see in world like "singleplayer's chest" etc |
| 16:53 |
Soni |
(also what's that about modding the menu?) |
| 16:55 |
rubenwardy |
the main menu is in Lua. You can override it by using the setting to set the path to the scripts |
| 17:20 |
Soni |
hmm alright that's 3 issues opened, anything we forgot? |
| 17:42 |
Zughy[m] |
4 new issues in not even an hour: that's evil |
| 17:43 |
Soni |
one of those is purely bonus points ^^ |
| 17:46 |
Zughy[m] |
tip: you can remove the "A clear and concise etc." text when you open a new issue ? |
| 17:47 |
Zughy[m] |
I would edit them, but I can't :)))) |
| 17:57 |
|
independent56 joined #minetest-dev |
| 18:01 |
Soni |
we wonder if the proxy thing would be easy |
| 18:01 |
Soni |
it's not purely network, you need a DB for proxy passwords |
| 18:02 |
Soni |
but it should be more or less easy |
| 18:03 |
Soni |
(also server admins might argue you need a way to add/remove proxies at runtime. also this would potentially allow minetest to work in a browser, using wasm.) |
| 18:10 |
schwarzwald[m] |
Soni: It looks like the WebAuthn key is based on the hardware of the host machine; what do I need to do to transfer it to another set of hardware? |
| 18:11 |
Soni |
schwarzwald[m]: it is not. unless you're using the windows or android webauthn stuff, in which case it is. but if you're using an actual webauthn key, like a yubikey or a solokeys, it's independent of the machine. |
| 18:13 |
schwarzwald[m] |
Ah, so it's an authentication key you carry around with you as another device you plug in? Do you usually log into applications that way? |
| 18:14 |
Soni |
yes |
| 18:14 |
Soni |
https://github.com/openssh/openssh-portable/blob/master/PROTOCOL.u2f |
| 18:15 |
Soni |
(this might be more helpful) |
| 18:18 |
Soni |
(definitely non-trivial tho) |
| 18:21 |
|
independent56 joined #minetest-dev |
| 18:28 |
|
independent56 joined #minetest-dev |
| 18:50 |
Soni |
btw has anyone done anything about the lack of loading screens e.g. in the menu? |
| 18:51 |
Soni |
hmm server list does have a loading screen, it's just browse content that locks up |
| 19:54 |
|
Baytuch joined #minetest-dev |
| 22:32 |
|
panwolfram joined #minetest-dev |
| 22:39 |
|
independent56 joined #minetest-dev |
| 22:42 |
|
independent56 joined #minetest-dev |
| 23:25 |
MTDiscord |
<MisterE> Soni: minetest.dustlabs.io |
| 23:29 |
Soni |
ah *nod* |
| 23:30 |
Soni |
yeah, doesn't work with IP bans |
| 23:37 |
MTDiscord |
<MisterE> Thats right :| |
