Time |
Nick |
Message |
01:05 |
|
Noisytoot joined #minetest-dev |
01:20 |
|
Noisytoot joined #minetest-dev |
03:17 |
|
Noisytoot joined #minetest-dev |
03:39 |
|
fluxionary joined #minetest-dev |
04:00 |
|
MTDiscord joined #minetest-dev |
04:06 |
|
YuGiOhJCJ joined #minetest-dev |
04:40 |
|
v-rob joined #minetest-dev |
06:15 |
|
YuGiOhJCJ joined #minetest-dev |
06:40 |
|
diceLibrarian joined #minetest-dev |
07:45 |
|
Noisytoot joined #minetest-dev |
07:52 |
|
Noisytoot joined #minetest-dev |
08:01 |
[MTMatrix] |
<grorp> josiah_wi: that would of course be better. |
08:22 |
|
Warr1024 joined #minetest-dev |
08:47 |
|
Warr1024 joined #minetest-dev |
08:57 |
MTDiscord |
<herowl> MC is using F3 as a modifier for some hard-coded bindings |
09:31 |
MTDiscord |
<herowl> BTW regarding scancodes vs keysyms: I'm in favor of allowing any custom bindings, which I think requires scancodes. Stuff like: Scroll Lock, Pause|Break, additional mouse buttons, F-keys as modifiers |
09:37 |
[MTMatrix] |
<grorp> what do other core devs think about using a formspec library for a mainmenu replacement? using luk3yx's Flow would make layout and quick iteration much easier |
09:38 |
[MTMatrix] |
<grorp> also stuff like having the layout/sizes be screen-size-dependent |
09:40 |
[MTMatrix] |
<grorp> we'd have to ask the author for a different license though, it's GPL 3 |
09:42 |
[MTMatrix] |
<grorp> one reason we don't have a new mainmenu yet is probably that raw formspecs are a pain to work with, but Flow makes them quite usable |
09:54 |
luk3yx |
I am happy to relicense it, though I'd need permission from Lazerbeak12345 (they have contributed code) |
10:05 |
[MTMatrix] |
<grorp> thank you, that's great. |
10:43 |
rubenwardy |
If the API could smoothly switch from formspecs to a new gui system in the future that would be ideal |
11:40 |
|
Desour joined #minetest-dev |
12:08 |
|
Noisytoot joined #minetest-dev |
12:33 |
[MTMatrix] |
<Zughy> Why can't we switch to GPL3 instead? |
12:36 |
celeron55 |
i like to be in the "somewhere in between GPL and MIT" which i think LGPL2 almost unqiuely provides. and moving to GPL3 would be permanent - there's no way back no matter how much it is regretted later |
12:40 |
celeron55 |
(or more specifically LGPL2.1+). it is valuable to have the path open to GPL3 there (i like having options), but that does not mean that it is wise to ever go there |
14:18 |
Noisytoot |
I would prefer AGPLv3 |
14:26 |
Noisytoot |
Hardcoding shift+7 to mean / may break other keyboard layouts. For example, on a UK QWERTY keyboard shift+7 is & |
14:30 |
[MTMatrix] |
<grorp> if the user has bound something to & |
14:31 |
[MTMatrix] |
<grorp> which is at least theoretically possible: https://github.com/minetest/minetest/blob/90fccc15ebf86b58656c28ed9853780d92feb69e/src/client/keycode.cpp#L219 |
14:31 |
[MTMatrix] |
<grorp> well, but that keybinding would already be broken now |
14:31 |
Desour |
we only want to fix the common cases, iirc. so as long as they can still write & with shift+7 in text inputs, it should be fine |
14:32 |
[MTMatrix] |
<grorp> they can since that's a different codepath (SDL textinput) |
14:33 |
[MTMatrix] |
<grorp> but I'm still curious to see what josiah_wi comes up with |
16:04 |
|
fluxionary joined #minetest-dev |
16:06 |
|
imi joined #minetest-dev |
19:13 |
MTDiscord |
<greenxenith> FWIW, if we switched to GPL3, a lot of people (including myself) would immediately fork the project and disown upstream |
19:16 |
MTDiscord |
<greenxenith> GPL is a virus |
19:24 |
|
Desour joined #minetest-dev |
19:28 |
celeron55 |
i do believe the choice of LGPL2.1 has the ability to unite certain groups of people who never otherwise would |
19:29 |
Mantar |
I didn't know anybody took that MS "viral GPL" meme seriously. It's only "viral" in the same way that all copyrighted material is: you can't build it into your project without being very careful, unless you have a license to do so |
19:35 |
MTDiscord |
<greenxenith> Mantar: GPL is so much worse than most other licenses |
19:36 |
Desour |
greenxenith: can you elaborate why it would be so much worse for the minetest engine? IANAL |
19:37 |
luk3yx |
Flow is licensed under LGPL 3 so relicensing it to LGPL 2.1 wouldn't be a very big change |
19:37 |
Mantar |
greenxenith: it seems pretty straightforward to me. Do you plan to honor the terms of the GPL? Great, you can use the code. If not, you don't have a license to use this copyrighted code. All its teeth come from copyright and apply to all copyrighted code unless you have a license that waives some of it |
19:38 |
Mantar |
other licenses are more permissive, which is great if you're MS and want to make money off other people's hard work without paying them or giving back in any way, but that doesn't make those licenses better, just different |
19:40 |
celeron55 |
(L)GPL is not about using code, it's about distributing the program |
19:41 |
Mantar |
true |
19:42 |
Mantar |
by "use the code" I mean "in your public project that you're distributing" |
19:42 |
celeron55 |
all widely used licenses allow you to use the thing in any form personally, as long as you have it in some form. the license terms come into play when you want to give an unmodified or modified version to someone else |
19:46 |
[MTMatrix] |
<Zughy> GPL is about not taking advantage of the free work of people to make money out of it (exploitation). At least not without giving anything back |
19:46 |
[MTMatrix] |
<Zughy> If companies are scared of strong licenses, aak yourself why |
19:46 |
[MTMatrix] |
<Zughy> *ask |
19:49 |
celeron55 |
i don't believe GPL is about that. but if you insist it's about that, so is LGPL |
19:51 |
[MTMatrix] |
<fgaz> There is also a third way: incorporate flow, but continue writing other code under LGPL. The combined work will be GPL, with the option of downgrading it back to LGPL by removing flow. |
19:52 |
celeron55 |
the fact that companies find it a bit more compelling to attempt to use LGPL-licensed stuff in their products compared to GPL-licensed ones doesn't mean they're fundamentally different. It's just that LGPL draws the line slightly differently. in reality, they're still not allowed to rip it off. what happens in practice of course is that they don't know what they're doing and end up not following |
19:52 |
MTDiscord |
<wsor4035> fgaz: flow is lgpl3, not gpl |
19:52 |
celeron55 |
the license terms, but that's outside of the scope of comparing license terms |
19:54 |
[MTMatrix] |
<fgaz> wsor4035: lgpl3 does not exist |
19:54 |
celeron55 |
all LGPL adds to GPL is that it permits bundling the open source thing within a closed source thing, if certain technical details are followed. if it's modified, then you have to provide the source code and a way to swap in further modified source code for the open source component |
19:54 |
[MTMatrix] |
<fgaz> oh wait, it does |
19:54 |
MTDiscord |
<wsor4035> fgaz: wrong again |
19:54 |
celeron55 |
eh, actually |
19:55 |
celeron55 |
there needs to be a way to swap in a modified version of the open source component in any case |
19:56 |
MTDiscord |
<wsor4035> this whole mess seems to be kicked off because grorp forgot a l in there message |
19:56 |
celeron55 |
GPL already allows using the open source component from a closed source program as long as it's a separate program. LGPL allows them being linked together into a single program. it's not a large difference |
19:57 |
MTDiscord |
<wsor4035> the difference is lgpl means the gpl doesnt infect whatever your using it with |
19:57 |
[MTMatrix] |
<fgaz> anyway, my point still stands. flow is lgpl3, the rest of minetest can be developed under lgpl2.1, and the combined work is gpl3, but flow can be removed at any time |
19:58 |
MTDiscord |
<wsor4035> 🤦♀️ |
19:58 |
MTDiscord |
<wsor4035> . s/gpl3/lgpl3 and you would be correct |
19:59 |
celeron55 |
fgaz: practically if you combine licenses like that in an upstream project like MT, the things become inseparable |
19:59 |
MTDiscord |
<wsor4035> also what c55 said |
19:59 |
[MTMatrix] |
<fgaz> wsor4035: no, this time I'm sure it's gpl3 https://www.gnu.org/licenses/gpl-faq.html#gpl-compat-matrix |
19:59 |
MTDiscord |
<wsor4035> this does get back to the whole old debate of a. should minetest move to lgplv3/ should it do it to be able to use code from multicraft which is lgplv3 |
20:01 |
MTDiscord |
<wsor4035> fgaz: your again wrong, because minetest is lgpl2.1 or later |
20:01 |
[MTMatrix] |
<fgaz> ...nevermind again. I thought mt was lgpl2.1-only |
20:02 |
[MTMatrix] |
<fgaz> wsor4035: yeah, I'll shut up now, looks like I'm too tired to think about licenses 😅 |
20:03 |
MTDiscord |
<wsor4035> another advantage to license like mit family, bsd family, etc is they arent a dumpsterfire to read and understand, plain and simple |
20:03 |
MTDiscord |
<wsor4035> also combinable into other projects easily |
20:04 |
celeron55 |
LGPL3 mainly clarifies some things that don't really come into play in MT in particular |
20:05 |
celeron55 |
if someone can point out a real world thing regarding to MT where LGPL3 would help, I'm open to it. but changing it for the sake of it doesn't make sense |
20:06 |
celeron55 |
anyway yes, MT is LGPL2.1+. I forget to write it in full most of the time, sorry about that. but the choice of exactly that is intentional |
20:07 |
MTDiscord |
<wsor4035> the only real world to minetest in lgplv2.1 vs lgplv3 would take code from lgplv3 projects now, instead of nicely asking them to let us use it under lgplv2.1. the most notable application of that would probably be multicraft |
20:07 |
celeron55 |
i don't think it's sufficient as a reason |
20:08 |
MTDiscord |
<wsor4035> im not fighting for it, just the only thing i could think of thats plausible |
20:11 |
celeron55 |
i'm actually surprised multicraft is using LGPL3 |
20:11 |
MTDiscord |
<wsor4035> they are because its a part of there claim that they can list minetest on the apple app store then iirc |
20:12 |
celeron55 |
LGPL2.1 would allow them to possibly get away with not actually giving people the means to build the source into the full app |
20:13 |
celeron55 |
i didn't know LGPL3 is compatible with the apple app store and LGPL2.1+ isn't. is that true? |
20:14 |
celeron55 |
i'd think LGPL3 would be _less_ compatible |
20:14 |
MTDiscord |
<wsor4035> the whole apple app store and gpl software is an unlitigated nightmare |
20:14 |
MTDiscord |
<wsor4035> everyone has there own version of what they think, and nothing has ever been to court over it, so who knows |
20:14 |
celeron55 |
ok so it's all BS |
20:16 |
MTDiscord |
<wsor4035> gpl in apple app store is most certain a no, lgpl is a probably fine???, not a lawyer however |
20:16 |
MTDiscord |
<wsor4035> gpl/apgl |
20:17 |
celeron55 |
the thing about LGPL, especially LGPL3 which states it explicitly, is that users have to be able to modify the open source portion of the app and build it back into a functional app |
20:17 |
MTDiscord |
<wsor4035> becasically its comes down to how you think apps are put together, if the lgpl part is linked, and then views on if the lgpl part can be swapped out or not (and your interpretatio of that) |
20:17 |
celeron55 |
doing that on an apple device is a nightmare for a normal user |
20:18 |
celeron55 |
you have to pay and sign a developer account even if you don't want to distribute the app, afaik |
20:19 |
celeron55 |
of course, apple is so completely incompatible with open source anyway that even talking about this seems silly |
20:21 |
celeron55 |
or, well, libre software |
20:24 |
Mantar |
If nobody's ever litigated it, that probably means that that everybody who brought the idea to lawyers had the lawyers tell them to settle. GPL court cases are rare, because it's very clearly written and binding |
20:26 |
MTDiscord |
<wsor4035> there is the whole vlc lgpl dumpsterfire, which was the most notable. i think however even they are still on ios under lgpl |
20:35 |
celeron55 |
given that apple is not going to change, and users are going to buy their products anyway, and it's impossible to install apps on their products from anywhere other than their app store, it's better to be there than to leave it all to the proprietary programs |
20:36 |
celeron55 |
e.g. multicraft is still better than actual minecraft |
20:37 |
celeron55 |
from a software freedom standpoint |
20:41 |
Mantar |
well said |
20:48 |
Desour |
because of the EU digital markets act (DMA), apple as gatekeeper has to provide means to install apps from other places than their contentdb. of course they're doing it in a malicious way, e.g. charging fees. and they're whining that users can also install malware mimimimi. see here: https://developer.apple.com/support/dma-and-apps-in-the-eu/ |
20:48 |
Desour |
disclaimer: idk anything about development on apple platforms |
20:51 |
celeron55 |
there's no freedom if a user is not allowed to install malware |
20:51 |
celeron55 |
that's just how it is |
20:53 |
Desour |
and there's no free will if a human is not allowed to do malicious decisions |
20:58 |
Mantar |
Yeah, what if I want Bonzi Buddy on my windows 98 install? |
21:01 |
celeron55 |
looks like even if you want to offer an iOS app to be downloaded directly from your website, you still have to send it to apple first for them to look at it, to fill in metadata and to sign it |
21:02 |
celeron55 |
the difference from before is.... well, before, there was absolutely no way of doing that |
21:31 |
|
v-rob joined #minetest-dev |
21:48 |
|
v-rob joined #minetest-dev |
22:33 |
|
panwolfram joined #minetest-dev |
23:01 |
MTDiscord |
<mistere_123> I understand that relicensing to lgpl3 allowing presence on the ios store is an open question, but if it is true, why wouldn't we create a mirror of minetest that has a script to relicense it to lgpl3 and add the apple build scripts automatically? That way we keep minetest on v2 while maintaining a presence on ios |
23:03 |
MTDiscord |
<wsor4035> someone has to pay the $100/year, else file non profit status for minetest, then file with apple to get a free dev account. after all that someone has to contribute the code for apple support |
23:04 |
MTDiscord |
<mistere_123> sound like an investment of time someone could make. Aneissue with help wanted tag could be made |
23:05 |
|
Eragon joined #minetest-dev |