| Time |
Nick |
Message |
| 00:07 |
|
paramat joined #minetest-dev |
| 00:56 |
|
Jordach joined #minetest-dev |
| 01:04 |
|
pauloue joined #minetest-dev |
| 01:35 |
|
turtleman joined #minetest-dev |
| 01:50 |
|
Cornelia joined #minetest-dev |
| 01:56 |
|
Nelly joined #minetest-dev |
| 03:16 |
|
Foz joined #minetest-dev |
| 07:01 |
|
pyrollo_ joined #minetest-dev |
| 08:17 |
|
Lone-Star joined #minetest-dev |
| 09:07 |
|
T4im joined #minetest-dev |
| 09:53 |
|
proller joined #minetest-dev |
| 10:14 |
|
proller joined #minetest-dev |
| 10:15 |
|
Beton joined #minetest-dev |
| 10:48 |
|
proller joined #minetest-dev |
| 11:30 |
|
Fixer joined #minetest-dev |
| 11:43 |
|
proller joined #minetest-dev |
| 12:01 |
|
calcul0n joined #minetest-dev |
| 13:50 |
|
proller joined #minetest-dev |
| 14:04 |
|
Wuzzy joined #minetest-dev |
| 14:12 |
|
calcul0n joined #minetest-dev |
| 15:15 |
rubenwardy |
updated #8383 |
| 15:15 |
ShadowBot |
https://github.com/minetest/minetest/issues/8383 -- Add formspec styling using a style[] tag by rubenwardy |
| 15:23 |
|
proller joined #minetest-dev |
| 15:43 |
|
pauloue joined #minetest-dev |
| 15:54 |
rubenwardy |
#8592 |
| 15:54 |
ShadowBot |
https://github.com/minetest/minetest/issues/8592 -- Add core.open_url() to main menu API by rubenwardy |
| 16:22 |
rubenwardy |
I suggest having a requirements discussion to find out what is blocking having mobs in Minetest Game, and what is stopping mobs being higher quality |
| 16:23 |
VanessaE |
a stable MOB API, and good animations |
| 16:23 |
Calinou |
do we have yaw interpolation yet? |
| 16:23 |
Calinou |
(for players too) |
| 16:23 |
Calinou |
that's one of the reasons mob animations don't look great |
| 16:24 |
VanessaE |
tenplus1's mobs redo does a good job in the meantime though |
| 16:24 |
rubenwardy |
what does a stable mob API mean? |
| 16:24 |
rubenwardy |
what is stopping Lua mobs API from doing things well already? |
| 16:24 |
rubenwardy |
Calinou: I'm unsure |
| 16:25 |
VanessaE |
a stable API? well you know, as in something any random modder can hook into to create another creature - where it spawns, under what conditions, how its fighting style works, can it be captured and tamed, etc... |
| 16:26 |
VanessaE |
like I said, mobs redo does a good job |
| 16:26 |
rubenwardy |
what's wrong with mobs redo in that respect? |
| 16:26 |
VanessaE |
nothing. |
| 16:27 |
VanessaE |
I'm trying to say it's a good example one could follow, but OF COURSE you know perfectly well that whatever makes it into mtg will leave out *just enough* that the mod it's based on (e.g. mobs redo) would still have to exist |
| 16:28 |
rubenwardy |
hmm |
| 16:29 |
VanessaE |
am I wrong? |
| 16:29 |
rubenwardy |
well, it would be good to find out what other games do, and how something should be improved |
| 16:29 |
rubenwardy |
also, I'd be for including mobs redo as is |
| 16:29 |
rubenwardy |
having ok mobs is better than having no mobs |
| 16:29 |
VanessaE |
agreed. |
| 16:29 |
VanessaE |
meh, who gives a shit what other games do, though? |
| 16:29 |
rubenwardy |
no no |
| 16:29 |
rubenwardy |
I'm talking about algorithms |
| 16:29 |
VanessaE |
oh |
| 16:30 |
rubenwardy |
like, they might have nice algorithms we can be inspired from |
| 16:30 |
rubenwardy |
also, created a discussion for dev meeting: https://github.com/orgs/minetest/teams/engine/discussions/18 |
| 16:30 |
rubenwardy |
also, it would be good to have this somewhere which is public |
| 16:31 |
VanessaE |
the only things I ever notice wrong with mobs redo is sometimes a rat will jump on top of a kitten as if to avoid being killed, and monsters will often go nuts rocking back and forth if you fly above or below them (instead of realizing you're out of reach and losing interest). |
| 16:32 |
VanessaE |
(kittens will eat rats, it's how you tame one) |
| 16:32 |
VanessaE |
I can't access that linmk |
| 16:33 |
VanessaE |
(I'm not in the appropriate access group on GH) |
| 16:33 |
Nelly |
Hmm... I wonder how hard it would be to implement a hierarchical pathfinding model that can be extended as more chunks are genned |
| 16:33 |
calcul0n |
mineclone and lott also do a great job with mobs, but nothing better than mobs_redo imo |
| 16:34 |
VanessaE |
rubenwardy: there IS one problem with mobs redo: as good as it is, it's a tenplus1 project, and ...well..you guys tend to be hard-ass with him. |
| 16:34 |
rubenwardy |
really? |
| 16:34 |
VanessaE |
yup. |
| 16:34 |
rubenwardy |
have I personally? |
| 16:34 |
VanessaE |
I don't recall. |
| 16:35 |
rubenwardy |
I vaguely recall paramat and 11 falling out |
| 16:35 |
VanessaE |
ohohoho... |
| 16:35 |
rubenwardy |
Santa? |
| 16:35 |
VanessaE |
:P |
| 16:36 |
VanessaE |
to say paramat and 11 have a falling out is something of an understatement |
| 16:36 |
VanessaE |
but since paramat basically holds the keys to mtg, well.. |
| 16:37 |
rubenwardy |
:'( |
| 16:40 |
|
proller joined #minetest-dev |
| 16:52 |
|
LoneWolfHT joined #minetest-dev |
| 17:20 |
rubenwardy |
trivial #8593 |
| 17:20 |
ShadowBot |
https://github.com/minetest/minetest/issues/8593 -- Update Github templates by rubenwardy |
| 17:47 |
|
Krock joined #minetest-dev |
| 17:55 |
Krock |
rubenwardy: where will SECURITY.md appear? |
| 17:58 |
rubenwardy |
In the create an issue page, as a banner asking the user id the issue is a security vuln |
| 17:58 |
rubenwardy |
So they'd click that banner, and it would go to the security page |
| 18:15 |
|
paramat joined #minetest-dev |
| 18:19 |
paramat |
nothing is blocking mobs being in MTG except core dev time. i keep thinking of adding at least an oerkki. having a well designed mob API is the hard part |
| 18:22 |
VanessaE |
G*d no. go with something modern like mobs redo. |
| 18:23 |
paramat |
also, i wouldn't trust tenplus1's coding, it's usually very sloppy, so mobs redo would need studying and alteration, might be a good starting point though |
| 18:24 |
VanessaE |
no. |
| 18:24 |
VanessaE |
no more alterations |
| 18:24 |
VanessaE |
no more half-assed imports. |
| 18:27 |
paramat |
nothing wrong with oerkkis, they're in most mobs mods, probably in mobs redo too |
| 18:27 |
VanessaE |
the CORRECT action here is to take a known, popular implementation and bring it in whole cloth, THEN rewrite it if you want, making sure it never loses compatibility with whatever bits plug into/use it/ |
| 18:27 |
rubenwardy |
it's better to have something good than have nothing waiting on something perfect |
| 18:29 |
paramat |
mobs redo may not be suitable as it is, and is very probably not high enough quality |
| 18:29 |
VanessaE |
but it is arguably the most popular. |
| 18:30 |
rubenwardy |
then how do we get one which is "high enough" quality? |
| 18:30 |
rubenwardy |
why hasn't one been created already? |
| 18:31 |
VanessaE |
paramat: I'm just gonna say it: stop worrying about code quality! |
| 18:31 |
paramat |
popular does not make it suitable or sufficient quality. better i think to create a new API |
| 18:31 |
VanessaE |
give it a fucking rest |
| 18:32 |
paramat |
heh good grief |
| 18:32 |
|
fwhcat joined #minetest-dev |
| 18:32 |
VanessaE |
code that's trash (In whoever's opinion), but works, is better than none at all |
| 18:32 |
paramat |
that's such a ridiculous statement |
| 18:33 |
VanessaE |
*facepalm* |
| 18:33 |
rubenwardy |
Minetest's code is trash |
| 18:33 |
rubenwardy |
it's ridiculous that we have no good mobs 8 years in |
| 18:33 |
rubenwardy |
(those are two different statements) |
| 18:36 |
paramat |
code is bad in places, but that's not an excuse to write bad code, anything new added is high quality |
| 18:36 |
|
DudestOfMen joined #minetest-dev |
| 18:36 |
Krock |
VanessaE: I agree that tenplus1's mobs_redo is what MTG needs, but at the same time I think it might need some refactoring |
| 18:36 |
VanessaE |
Krock: then refactor *after* adding it, or it'll never get in. |
| 18:36 |
Krock |
yes, that can happen in many small steps. Just saying that it might put some off due to that |
| 18:37 |
Krock |
question is also whether ten wants to give the "ownership" to MTG |
| 18:37 |
VanessaE |
that's a fair question |
| 18:37 |
paramat |
mobs redo might be excellent, i don't know yet, but knowing the author's code quality it needs to be studied and understood |
| 18:38 |
Krock |
and if there are new changes which aren't accepted in Minetest (API design reasons, for example), then will there be the risk that there will be a mods_redo2 which is just continued as before? |
| 18:38 |
VanessaE |
paramat: you're missing the point... |
| 18:39 |
paramat |
it can't be reworked after adding because you then break the API. code is added when it is ready |
| 18:39 |
Krock |
paramat: the code works very well, and they're working hard on making it fast for servers. studying it will take a while, especially to find better optimizations |
| 18:39 |
Krock |
1) you don't break the API in first place |
| 18:40 |
Krock |
provide deprecated API functions which still implement the functionality, maybe in a hacky way |
| 18:40 |
Krock |
similar to "userspace shall not be broken" |
| 18:41 |
Krock |
key point is to have compatibility with existing mods, so changing the API to something entirely new isn't an option |
| 18:42 |
paramat |
you can't add code that isn't understood and isn't checked for quality, everything has to pass code review |
| 18:43 |
VanessaE |
what you mean is understood by YOU, paramat. |
| 18:43 |
VanessaE |
I mean, you're not the only person capable of evaluating an addition. |
| 18:44 |
paramat |
but sure it could then be further improved later, but you wouldn't then be able to alter the API, unless we call it unstable, that's a possibility |
| 18:44 |
Krock |
improvements do not directly mean altering the API |
| 18:45 |
paramat |
VE don't be insulting, that's a nasty accusation |
| 18:45 |
VanessaE |
I'm not accusing, I'm stating a direct observation |
| 18:45 |
* Krock |
prepares the fire extinguisher just for the case |
| 18:45 |
paramat |
krock i agree |
| 18:45 |
VanessaE |
you're acting as if not even the AUTHOR could understand their code. |
| 18:46 |
VanessaE |
THAT is what I disagree with |
| 18:46 |
paramat |
no ... |
| 18:46 |
VanessaE |
very well, then "isn't understood" by WHOM then? |
| 18:46 |
VanessaE |
-then |
| 18:46 |
paramat |
VE stop it or i kick you, this is dev channel |
| 18:46 |
Krock |
paramat: I can see that adding non-MT-standard formatted code is a PITA, but look at how much that would need a change before it were compatible in just one PR |
| 18:46 |
VanessaE |
wtf? |
| 18:47 |
fwhcat |
may I add my 2 cents ? hello all. |
| 18:47 |
Krock |
if someone decides to work on it until the next release to adjust it to our standards it's fine to me if we just added it directly. otherwise it will go nowhere as VE stated previously |
| 18:48 |
fwhcat |
I think what's difficult is to take the time to read and understand what the code does exactly, all the implication/events, etc. Some people would have to take serious time before merging mobs_redo into MTG. |
| 18:48 |
Krock |
alternatively it could be done in a separate fork, but I doubt that reviewers of MTG would clone that manually |
| 18:49 |
Krock |
fwhcat: c'mon. api.lua is only 4200 lines long |
| 18:50 |
Krock |
and if any of you looked at the source code already (you probably didn't): there are much more comments than in some places of MTG |
| 18:50 |
paramat |
krock sure, i'm just saying it has to meet the usual minimum quality standards before merge. MTG has always had high standards and should continue to |
| 18:51 |
Krock |
paramat: I agree that this would be the optimal case. Counter-question: do we fix Irrlicht source code style? |
| 18:51 |
paramat |
no, VE is wrong to suggest it 'will never be added' if we do this with care, please don't be misled by that |
| 18:51 |
VanessaE |
bull. |
| 18:52 |
Krock |
how about this: |
| 18:52 |
Krock |
1) ask whether it may become a minetest-mods mod |
| 18:53 |
Krock |
2) ask whether ten is fine when the source code is re-formatted |
| 18:53 |
VanessaE |
if something has to be ideal to be added, then something complex like mobs redo does not stand a chance./ |
| 18:53 |
Krock |
3) re-format it to make it MTG-ready |
| 18:53 |
Krock |
4) merge that stuff |
| 18:54 |
Krock |
minetest-mods has the advantage that multiple people could work on it (at the risk of nobody doing anything) |
| 18:54 |
fwhcat |
Do you guys provide any tools to check for code quality or format ? |
| 18:54 |
paramat |
krock sounds ok to me |
| 18:54 |
Krock |
fwhcat: yes, there's clang-format (lint) |
| 18:55 |
VanessaE |
paramat: all right, then tell me this: when was the last time some complex, third-party project got added to mtg and remained compatible with whatever used it before? |
| 18:55 |
Krock |
buildbot checks for the code style and unused variables when you create a PR or push a commit |
| 18:55 |
Krock |
carts, but that was like 1/10th of the code |
| 18:55 |
Krock |
and the format was already correct |
| 18:56 |
paramat |
isn't mobs redo too complex for MTG? we were intending something very simple. i think it's best mobs redo stays a separate optional mod, however i will probably be outvoted on that =) |
| 18:56 |
Krock |
the goals are still not clear |
| 18:57 |
Krock |
it might not add much content, but some API functions like mobs would be great to give MTG more of a purpose |
| 18:57 |
Krock |
right now it lives from the "default" and "stairs" dependencies |
| 18:58 |
paramat |
VE irrelevant question for me personally, i don't care about a MTG mobs API being mobs redo compatible, i prefer it wasn't |
| 18:58 |
VanessaE |
I didn't say mobs API |
| 18:58 |
Krock |
I don't have such a strong opinion about that, but I've seen various people on the forums asking for content |
| 18:59 |
VanessaE |
I said "complex third-party project" as in ANY feature at all |
| 19:01 |
paramat |
i prefer any large new feature starts with a fresh improved API, so that never happening doesn't bother me |
| 19:01 |
VanessaE |
in other words, paramat, you have nfi. |
| 19:02 |
VanessaE |
(which is fine) |
| 19:04 |
VanessaE |
in fact, I don't think one ever has, unless you count xpanes or doors, and we've had those for years. |
| 19:04 |
paramat |
how intensive is mobs redo? anything in MTG has to be lightweight so that the game runs on most hardware. i'm concerned about the line count, it looks very bloated |
| 19:05 |
VanessaE |
5289 loc, counting all .lua files in the engine/API part. |
| 19:06 |
VanessaE |
(4253 in api.lua, as Krock said) |
| 19:06 |
paramat |
carts was quite big |
| 19:06 |
VanessaE |
and is another that we've had for years, also. |
| 19:07 |
VanessaE |
I run mobs_redo on my Bananaland server, and never have problems with it. |
| 19:07 |
VanessaE |
and that one is a low-lag server. |
| 19:10 |
VanessaE |
so if there's one thing I'm reasonably sure of, its that whatever you think of the code quality, it works just fine and performs well. |
| 19:11 |
paramat |
VE, your question is trying to prove that no big feature was added because our standards stopped it, if this has happened that's a good thing. or, that if a big feature was added it broke compatibilty, but that's never happened |
| 19:11 |
VanessaE |
(if something performs and behaves well enough for a server, then by definition it's likely more than good enough for singleplayer) |
| 19:11 |
paramat |
having carts for years is irrelevant to your question |
| 19:11 |
VanessaE |
*headdesk* |
| 19:12 |
VanessaE |
why do I even bother arguing with you? |
| 19:12 |
paramat |
a feature 'appearing to work' doesn't necessarly mean it is good enough, it may be unnecessarily intensive |
| 19:13 |
VanessaE |
if mtg code standards stopped something from being added, either the "something" is trash, OR the standards are too damned high. |
| 19:13 |
paramat |
why has having carts for years relevant to your question? also, carts was fairly recent |
| 19:13 |
VanessaE |
and clearly you are unwilling to accept the notion that maybe, just maybe the latter is true. |
| 19:14 |
VanessaE |
what about carts? YOU brought that up |
| 19:14 |
paramat |
MTG standards are not too high, it's not that good =) |
| 19:14 |
* VanessaE |
sighs |
| 19:14 |
VanessaE |
point proven. |
| 19:16 |
paramat |
"and is another that we've had for years, also." i'm just wondering what you mean by that and why it's relevant to your question |
| 19:16 |
VanessaE |
it's a measure of feature progression. |
| 19:16 |
VanessaE |
and btw: |
| 19:16 |
VanessaE |
carts was added from, boost_cart back in 2016. |
| 19:16 |
paramat |
your question was about the entire history of MTG after all |
| 19:16 |
VanessaE |
that's a LONG TIME, as gaming goes. |
| 19:17 |
rubenwardy |
Krock: "We only support the latest stable and latest development versions for security issues." |
| 19:17 |
rubenwardy |
hmmm |
| 19:17 |
rubenwardy |
problem with this is that we don't guarantee latest dev is checked for bugs |
| 19:18 |
rubenwardy |
maybe just "We only support the latest stable version for security issues." |
| 19:18 |
paramat |
well, carts was my answer to yourquestio of a big feature added withot being mangled, but you dismissed it by saying we've had it for years |
| 19:18 |
paramat |
*question |
| 19:18 |
VanessaE |
yes, for years. and how long did that mod exist before being added, just offhand. |
| 19:18 |
Krock |
rubenwardy: precisely. |
| 19:18 |
VanessaE |
let's see... |
| 19:19 |
VanessaE |
about 2 years. |
| 19:19 |
rubenwardy |
Krock: which one |
| 19:20 |
Krock |
> maybe just "We only support the latest stable version for security issues." |
| 19:20 |
Krock |
or did I miss your point? |
| 19:20 |
Krock |
oh. the development as well |
| 19:21 |
paramat |
VE ok, so what's your point? |
| 19:21 |
Krock |
ofc. latest development and stable release. both should be mentioned |
| 19:21 |
VanessaE |
paramat: my whole point here is that it takes for-fucking-EVER for some third-party project to get added to mtg |
| 19:21 |
VanessaE |
and you can't tell me Krock's code is garbage. |
| 19:21 |
paramat |
tenplus1 |
| 19:21 |
Krock |
it just happened to be in the MT-style |
| 19:21 |
VanessaE |
huh? no |
| 19:21 |
paramat |
oh sorry |
| 19:21 |
VanessaE |
Krock. |
| 19:22 |
VanessaE |
he wrote it, YOU added it, fwiw. |
| 19:22 |
paramat |
no, krock's code is good because they're a good coder |
| 19:22 |
VanessaE |
yeah, and? |
| 19:22 |
paramat |
carts didn't take long to be added, and no other large project was ever added to MTG, so your point fails |
| 19:23 |
VanessaE |
you literally just made my point right there. |
| 19:23 |
VanessaE |
"no other large project was ever added" |
| 19:23 |
paramat |
because it wasn't good enough or lack of dev time |
| 19:24 |
paramat |
neither of which is anyone's fault |
| 19:24 |
VanessaE |
let's see.. minetest has had modding capability since 2012 I guess, took 2 years to come up with that mod, then another 2 years after it was published before it got added to mtg. |
| 19:25 |
VanessaE |
and THAT was 3 years ago. |
| 19:25 |
VanessaE |
have you no concept of just how long that is, in gaming terms? |
| 19:26 |
|
proller joined #minetest-dev |
| 19:27 |
paramat |
carts was added quickly once it had been agreed it was to be added, it's not as if there was a 2 year fight to get it in |
| 19:28 |
VanessaE |
I'll grant you that bitg. |
| 19:28 |
VanessaE |
bit |
| 19:29 |
VanessaE |
but how long had mtg needed them? |
| 19:29 |
VanessaE |
and how long has mtg needed mobs? |
| 19:30 |
VanessaE |
(hint: for about 7 years, in the latter case) |
| 19:30 |
paramat |
anyway, you're trying to make a negative point but failed, and with that point are trying to argue for dropping standards, which will never happen |
| 19:30 |
VanessaE |
and how many implementations have there been in that time? |
| 19:30 |
VanessaE |
between 5 and 10 by my guess. |
| 19:32 |
paramat |
how long MTG needed something is irelevant, MTG has always needed many things |
| 19:32 |
VanessaE |
wat |
| 19:32 |
VanessaE |
ok I give up. |
| 19:34 |
VanessaE |
oh, and since you decided to pull out the big boots, I'll just save you the hassle. |
| 19:34 |
|
VanessaE left #minetest-dev |
| 19:48 |
rubenwardy |
Krock: updated #8593 |
| 19:48 |
ShadowBot |
https://github.com/minetest/minetest/issues/8593 -- Update Github templates by rubenwardy |
| 19:48 |
paramat |
anyway, my request for mobs redo is: i've reviewed tenplus1's code enough to know it is often very sloppy, so mobs redo will certainly need understanding and checking before adding. but yes it may be possible to improve it without breaking the API |
| 19:51 |
paramat |
i wouldn't work on it though as i don't think a big MTG project is worth my time, i rather think we should be redirecting effort to new games |
| 20:02 |
paramat |
just to reply to this nonsense: i do not 'hold the keys to MTG' obviously due to dev rules, 2 other devs can merge something and 1 disapproval doesn't stop anything |
| 20:08 |
Krock |
it can. disapprovals are there if a core dev thinks there's no way to get it merged in its current state |
| 20:10 |
paramat |
that's not my understanding, single disapprovals are often overruled |
| 20:11 |
paramat |
probably for the best as so much would be blocked |
| 20:11 |
Krock |
well so, I guess that's another point for the upcoming meeting |
| 20:11 |
paramat |
i think it's part of the 'no one can act alone' thing, no one can add or block alone |
| 20:19 |
|
calcul0n joined #minetest-dev |
| 21:05 |
|
paramat joined #minetest-dev |
| 21:13 |
|
proller joined #minetest-dev |
| 22:04 |
|
ssieb joined #minetest-dev |
| 22:44 |
|
turtleman joined #minetest-dev |
| 23:30 |
|
Calinou joined #minetest-dev |
| 23:32 |
|
LoneWolfHT joined #minetest-dev |
| 23:36 |
|
Fixer_ joined #minetest-dev |
| 23:43 |
|
pauloue joined #minetest-dev |
