Minetest Game now in maintenance/bugfix-only mode, no feature PRs accepted

User avatar
paramat
Developer
Posts: 3700
Joined: Sun Oct 28, 2012 00:05
GitHub: paramat
IRC: paramat
Location: UK

Minetest Game now in maintenance/bugfix-only mode, no feature PRs accepted

by paramat » Post

The Minetest Game core devs have decided to move the game to a development state where only code maintenance and bugfixes occur. New features will not be added.
It will of course continue to be carefully maintained long term to support the many servers, games and mods that rely on it, the commitment to this has not changed.

New PRs for new features will not be accepted.
If you have an idea for a contribution but are unsure whether it qualifies to be accepted, open an issue first to discuss it with us. Code maintenance includes translation work. New textures will not be accepted unless they count as maintenance or bugfix.
Currently existing feature PRs will continue to progress if they have support from core devs.

User avatar
DrFrankenstone
Member
Posts: 231
Joined: Tue May 24, 2016 05:36
GitHub: treer
Location: Australia
Contact:

Re: Minetest Game now in maintenance/bugfix-only mode, no feature PRs accepted

by DrFrankenstone » Post

I'll add a link to the discussion behind this, as this post came as a bit of a surprise and left me wondering about the intention to sunset MTG.
Last edited by DrFrankenstone on Thu Jul 30, 2020 00:29, edited 1 time in total.

User avatar
rubenwardy
Moderator
Posts: 6972
Joined: Tue Jun 12, 2012 18:11
GitHub: rubenwardy
IRC: rubenwardy
In-game: rubenwardy
Location: Bristol, United Kingdom
Contact:

Re: Minetest Game now in maintenance/bugfix-only mode, no feature PRs accepted

by rubenwardy » Post

Issue is here: https://github.com/minetest/minetest_game/issues/2710

Please please make your own games! Either from scratch or by forking Minetest Game (MTG). MTG has historically suffered too much from bureaucracy and design-by-committee. I'd like to see both individual and group game projects, too

Also, note that this is not the first time Minetest Game has been frozen. The last freeze ended in 2014, when a fork was merged back into MTG. So this isn't necessarily a nail in the coffin, and good quality forks are very valuable

My more detailed thoughts on this:
MTG certainly does have management problems. I've been burned out with Minetest Game for a while, because it takes so much effort to do so little that it's not rewarding.

The skills for writing games and mods for Minetest are different. The bar is also much lower in terms of programming, and art/design is much more important. This means that it doesn't make sense to have such an overlap between engine devs and game devs.

The way I see it, the problem we're trying to solve here is the stale development of MTG. Freezing development and encouraging forks is one way to do that, but I'm not sure if it's the best way.

I think one cause of the stale development is that no one agrees on what MTG should be. The way to solve this would be to split the project into two games - one a stable base for servers and modding, the other a decent game. The new "decent game" project should have lower requirements in terms of compatibility and code quality, such that it's easier to iterate and improve on the gameplay. It should also have someone in charge of it, to not end up with design-by-committee. I don't know whether freezing MTG and encouraging forks is the best way to do that, but it's one way I guess.

Ultimately, we need someone to step up on Minetest Game-related development, whether that's to make a fork or lead a (semi-)official version. Bagsy not me
Renewed Tab (my browser add-on) | Donate | Mods | Minetest Modding Book

Hello profile reader

User avatar
Wuzzy
Member
Posts: 4786
Joined: Mon Sep 24, 2012 15:01
GitHub: Wuzzy2
IRC: Wuzzy
In-game: Wuzzy
Contact:

Re: Minetest Game now in maintenance/bugfix-only mode, no feature PRs accepted

by Wuzzy » Post

These are great news!
The focus in the last years has been way too much on Minetest Game. To many players, MTG = MT, which is limiting potential. Minetest really should not revolve all around MTG.

Note that if you're a player, there's nothing to worry about: MTG is NOT dead with this decision. The servers based on MTG still exist, you can still play on them perfectly fine. If you're a player, nothing actually changes for you. And bugfixes are still being promised as well.

Also, another important context: The last update (5.3.0) only came with less than 10 new features. So development was already stalling very much. Even if this announcement would not have been made, development was de facto already at a near standstill.

This is not an accident. More and more MTG developers became disillusioned with the course of MTG. It makes perfect sense to move on.

And yes, I agree with rubenwardy that more games are needed. Obviously.

User avatar
Zughy
Member
Posts: 301
Joined: Thu Mar 26, 2020 18:23
GitHub: belongs_to_microsoft
In-game: Zughy
Location: Italy
Contact:

Re: Minetest Game now in maintenance/bugfix-only mode, no feature PRs accepted

by Zughy » Post

Well, I can't really say I'm sorry. I'm actually very happy about it, and after reading the whole GitHub issue I'm even more convinced now. One problem less, that's great news!

User avatar
Saturn
Member
Posts: 57
Joined: Sat Jul 11, 2020 12:03
Location: Lombardy, Italy

Re: Minetest Game now in maintenance/bugfix-only mode, no feature PRs accepted

by Saturn » Post

Wuzzy wrote:
Thu Jul 30, 2020 18:03
The focus in the last years has been way too much on Minetest Game. To many players, MTG = MT, which is limiting potential. Minetest really should not revolve all around MTG.
To many players MT & MTG == Cheap Minecraft clone
In my not-so-humble opinion, skilled developers like you should stop developing stuff like MineClone, and mods that only exist to make MT & MTG look like Minecraft; people having coding experience, git experience, translating experience like you should focus on creating something new, instead of reinventing what already exists.
Also Known As: dad installed Linux, how can I make it look like Windows?

Nothing wrong with it, just please don't make it your main goal: no "original" content added to the Minetest-verse and attracts only that kind of people - hardly the kind of people that will create something new (or something at all) for Minetest. Because they are used to buy software, or to only use it - not creating it.
Wuzzy wrote:
Thu Jul 30, 2020 18:03
And yes, I agree with rubenwardy that more games are needed. Obviously.
Games are created also to create a community around it, to share thoughts, have fun and socialize.
If the only thoughts that are constantly brought in discussion are "Minetest lacks this/that compared to Minecraft", after a while one loses the will of contributing, fun is spoiled and the community doesn't grow.

So, do not take this personally: there are others that focused on cloning Minecraft's features, making this their main goal; my eyes bleed every time I think that if I were as skilled as you are I would not surely spend my time reinventing the nether, the witches, the whole Minecraft-verse.

Minetest has a HUGE potential, it could be used as MMORPG, or as virtual environment like Second Life/OpenSim.

Please, stop cloning - at least as main job - and start creating.
This perhaps will add value to the Minetest-verse and attract other people interested in spending their time creating original contents - making the community grow.

/me steps down from the soap-box and shuts his mouth.

Mineminer
Member
Posts: 325
Joined: Mon Mar 05, 2018 04:05

Re: Minetest Game now in maintenance/bugfix-only mode, no feature PRs accepted

by Mineminer » Post

Saturn wrote:
Thu Jul 30, 2020 19:38
Wuzzy wrote:
Thu Jul 30, 2020 18:03
The focus in the last years has been way too much on Minetest Game. To many players, MTG = MT, which is limiting potential. Minetest really should not revolve all around MTG.
To many players MT & MTG == Cheap Minecraft clone
In my not-so-humble opinion, skilled developers like you should stop developing stuff like MineClone, and mods that only exist to make MT & MTG look like Minecraft; people having coding experience, git experience, translating experience like you should focus on creating something new, instead of reinventing what already exists.
Also Known As: dad installed Linux, how can I make it look like Windows?

Nothing wrong with it, just please don't make it your main goal: no "original" content added to the Minetest-verse and attracts only that kind of people - hardly the kind of people that will create something new (or something at all) for Minetest. Because they are used to buy software, or to only use it - not creating it.
Wuzzy wrote:
Thu Jul 30, 2020 18:03
And yes, I agree with rubenwardy that more games are needed. Obviously.
Games are created also to create a community around it, to share thoughts, have fun and socialize.
If the only thoughts that are constantly brought in discussion are "Minetest lacks this/that compared to Minecraft", after a while one loses the will of contributing, fun is spoiled and the community doesn't grow.

So, do not take this personally: there are others that focused on cloning Minecraft's features, making this their main goal; my eyes bleed every time I think that if I were as skilled as you are I would not surely spend my time reinventing the nether, the witches, the whole Minecraft-verse.

Minetest has a HUGE potential, it could be used as MMORPG, or as virtual environment like Second Life/OpenSim.

Please, stop cloning - at least as main job - and start creating.
This perhaps will add value to the Minetest-verse and attract other people interested in spending their time creating original contents - making the community grow.

/me steps down from the soap-box and shuts his mouth.

Time is not wasted if you enjoys it. Honestly I seen people waste away their lives on things they don't enjoy whether that's long term jobs/careers they don't enjoy at all because money is at stake, people nigging others to do things they rather not do and so on.

If people genuinely enjoys cloning Minecraft how is that doing harm to anyone I am right? :)

User avatar
v-rob
Developer
Posts: 970
Joined: Thu Mar 24, 2016 03:19
GitHub: v-rob
IRC: v-rob
Location: Right behind you.

Re: Minetest Game now in maintenance/bugfix-only mode, no feature PRs accepted

by v-rob » Post

Mineminer wrote:
Thu Jul 30, 2020 21:11
If people genuinely enjoys cloning Minecraft how is that doing harm to anyone I am right? :)
Well, if it gets very popular, it might make Microsoft lose business, so it could affect someone... :)

Anyway, I'm glad for this news. Good for the developers, but I have another secret reason. I've never liked building stuff with MTG because I was always afraid that it might update and change the map in new mapchunks and add different gameplay elements that I would have missed out on earlier. Now I can build in peace, knowing that my worlds will always be up to date :) (Until I find a game that I like better, of course)
Core Developer | My Best Mods: Bridger - Slats - Stained Glass

Sokomine
Member
Posts: 4276
Joined: Sun Sep 09, 2012 17:31
GitHub: Sokomine
IRC: Sokomine
In-game: Sokomine

Re: Minetest Game now in maintenance/bugfix-only mode, no feature PRs accepted

by Sokomine » Post

I'm not so glad about this decision but I can understand that it became necessary. It's true that developing the engine and developing a game requires diffrent skils and that time is also often far too short to do all.

One of the defining concepts of the type of games the engine supports is that things can be crafted from other things. It is definitely part of the lua_api.txt - so we may consider it important. The trouble is: In order to craft something new (which you can happily define in your mod, no problem there) is that the things you want to craft your new thing out of have to exist. If not, you can still try to spawn them at mapgen time (too bad if you don't want to define new biomes) or let mobs drop them when they die (which requires letting the mobs spawn). Or you can use /giveme or an equivalent command - which breaks immersion.

In theory, groups are a way to get around that. The drawback is that using groups for your recipes limits the number of diffrent recipes considerably because there are not that many groups in the first place - and requirements for easy-to-remember-shapes place further restrictions upon possible recipes. Futhermore most groups are not that well defined and subject to interpretation. Their original purpose is a diffrent one.

If game developers are disencouraged from basing their game on MTG, they'll invent new names for the same things and leave modders with a total, needless mess. This has already happened to some degree. Some games such as MineClone2, RealTest and Ethereal (mostly a mapgen) already did that in the past. I tried to support them, but that costs me valuable time and may be in vain if things are changed in the future in these games.

There are basic "concepts" most games have. It's ok if a specific game doesn't have one concept and is diffrent there, but most will share most of these concepts. One such concept may be "dirt" - stuff that is sometimes covered with a thin layer of vegetation (that is part of the block and can spread) or snow and is then called diffrently, but in general you can hoe it and/or plant plants on it. I want this thing to be either called "default:dirt" or - if a game developer decided he can't abide with that name - with a *translation/compatibility* function that allows other mods to rely on the existence of said "default:dirt" even if the game creator decided to call it "my_very_special_private_mod:the_strange_material_you_get_if_you_dig_down_with_a_shovel" or "g:d" or "wuzzy:1234" or whatever. The game creator won't have to implement that translation function - that has to be available to all games - the game creator would just have to define that his "default:dirt" is called so-and-so. It may still behave diffrently internally, but other mods won't have to worry about how it's called in any particular game - they could just go for "default:dirt" and be done with it if they want the concept of "dirt".

Another thing are trees. The concept of a tree - place a sapling, wait, and the sapling gets replaced with a structure containing wood and leaves and something that can be turned in a new sapling - that's pretty universal as well. My treeslib tried to provide that general functionality, but then the code for leaf decay was improved and I never ported that over (not that anyone showed serious intrest in a library anyway). Other such concepts are beds, doors, dyes, stair-like nodes and so on. Not to forget about mobs either.

Most intresting mods that implement a library lead a pitiful life as nobody ever uses them apart from perhaps their creator. It's ok to implement something anew if you want to try a diffrent approach - even with standarization there ought to be room for that - but you don't have to. And if you want to, it'd be best to create your new mod as well as possible as a drop-in replacement for the "default" one so that other mods can still interact with your mod.

Thus, what I want is standarization of a set of nodes/craftitems that have to be available in each game under the same name. The game may call them diffrently interanally, but other mods have to be able to access these things by their standard name. The game may even implement just a compatibility node if it doesn't use that node/item itself and can prove that it doesn't have anything that at least halfway fits the concept. The availability of these nodes/items can be checked automaticly and mods not conforming to the standard ought to be flagged with a very big red warning in the contentdb and when starting the game.

The other thing are libraries. They have to be available. Not hidden somewhere on the forum in threads. And not maintained by individual modders either. Please get a library collection running somewhere! With libraries that expose as much as possible to their public namespace, and with libraries that are designed to be replaced with individual libs with compatible calls if a game developer such decides.

A game creator wanting a tree would then just have to give the thing a name and textures - and likewise with mobs. If the game developer wants what is implemented in the library to behave diffrently, then he can still replace the library for his game while maintaining compatibility as well as possible. Please - standards and libs are important! It doesn't matter that much if a standard is defined to be 99, 100 or 101 - it just has to be a standard so that others can interact with it.
A list of my mods can be found here.

User avatar
firefox
Member
Posts: 1709
Joined: Wed Jan 14, 2015 07:34
In-game: Red_Fox
Location: Xanadu

Re: Minetest Game now in maintenance/bugfix-only mode, no feature PRs accepted

by firefox » Post

Sokomine wrote:
Fri Jul 31, 2020 23:15
If game developers are disencouraged from basing their game on MTG, they'll invent new names for the same things and leave modders with a total, needless mess.
but aren't most games "just" MTG with mods?
if they are based on MTG, they would still retain the original group/node names, so universal mods based on (or made for) MTG would still work on those games.
MTG not getting new features could actually bring stability, as games and mods wouldn't have to be updated everytime something in MTG changes.

original games would require original mods, but the same goes for texture packs.
most texture packs only support the default MTG, some do some popular mods, but usually any mod added to a world with a texture pack looks out of place. this is not really a new problem.
Sokomine wrote:
Fri Jul 31, 2020 23:15
Thus, what I want is standarization of a set of nodes/craftitems that have to be available in each game under the same name.
sounds good, but it also means that all games would have to use the same standards regarding the value of materials and the process of crafting.
what if a game uses a theme where wood is a rare treasure? or crafting is done in multiple steps, cutting and bending materials into specific shapes and then putting the pieces together?
a simple universal recipe wouldn't work for those.
i believe universal mods that add things to all games indiscriminately are quite game breaking. recipes aside, just adding a new metal and tools can totally mess up the game balance if the distribution of ores is different from the game that the mod creator used.

as i see it, most mods were created to add things to MTG which didn't exist in MTG.
when (or if) new games are created, they most likely already include those or similar things, so they don't need to be compatible with those mods.
if someone makes a mod for another game, because the game creator didn't include something that the person wanted, then this mod would be for that game, specifically adjusted to that game's properties.
i don't think said mod would do any good in another game that is fundamentally different.
✨🏳️‍🌈♣️✨

Sokomine
Member
Posts: 4276
Joined: Sun Sep 09, 2012 17:31
GitHub: Sokomine
IRC: Sokomine
In-game: Sokomine

Re: Minetest Game now in maintenance/bugfix-only mode, no feature PRs accepted

by Sokomine » Post

firefox wrote: but aren't most games "just" MTG with mods?
In the end, all MT games are just collections of mods.
firefox wrote: f they are based on MTG, they would still retain the original group/node names, so universal mods based on (or made for) MTG would still work on those games
Those games claim not to be based on MTG - yet they share a lot. Ground for the player to stand on is rather popular in most games. Many also have dirt, stone with ores, trees...Of course these elements may not be present in one particular game, or they may behave slightly diffrent, but most are present in one form or another.

Game developers creating new games (RealTest and MineClone2 for example) tend to name the same thing diffrently. No matter if you call it default:wood, planks:wooden_planks, my_own_namespace:something_like_wood, w:1234 or whatever - if it mostly behaves as wood does, it ought to be named that way. Individual chairs, tables, houses and other things vary - yet we have common names for them to fascilitate communication and life. I want standard names for those things to fascilitate mod interoperability.
firefox wrote: MTG not getting new features could actually bring stability, as games and mods wouldn't have to be updated everytime something in MTG changes.
That's right. Name changes in MTG where no fun either. But at least it was just one place where the names where changed, and most of the time it was just new things beeing added.
firefox wrote: most texture packs only support the default MTG, some do some popular mods, but usually any mod added to a world with a texture pack looks out of place. this is not really a new problem.
Hm. That's right. Sounds like an additional problem. Some sort of interoperability might help there as well?
firefox wrote: sounds good, but it also means that all games would have to use the same standards regarding the value of materials and the process of crafting.
That's not even so for all those MTG-based servers out there. They have vastly diffrent economics due to the mods and tweaks they use.
firefox wrote: what if a game uses a theme where wood is a rare treasure?
Still, it's wood - the things the new mod adds can be created. They may be terribly expensive, but that may even be desired in that particular game. It's also possible that the node does not really exist in the game normally (doesn't spawn and cannot be crafted but is defined in lua) and that you have to get it from an expensive admin shop ("Wood from earth - imported to space! Not that crappy plastic-wood you see so much around!"), or that there is no way to obtain it at all and so new craft recipes have to be added alongside with the new mod so that it can be used. That's fine as well. That's then the job of whoever adds the mod to that game - be it a server admin or a user. But the mod as such can rely on "default:wood" beeing defined. It would be extremly bad design (as it is now!) if each modder would have to add his own version of compatibility just because a game creator got creative with node names instead of with content.
firefox wrote: or crafting is done in multiple steps, cutting and bending materials into specific shapes and then putting the pieces together?
That's someething I'd very much love to see in games. But in how far does that conflict with there having to be a default:wood defined for example? You can make it more difficult to obtain it, and the way the mod crafts are defined may be visually/gameplay-like game breaking - but that's still the job of whoever wants to integrate the game into the mod. It's just not the modder's job to build large amount of code around naming incompatibilities.

However, a practical way to deactivate all craft recipes in a mod could be very convenient and something we might want to aim for. That'd fascilitate integration into games.
firefox wrote: believe universal mods that add things to all games indiscriminately are quite game breaking. recipes aside, just adding a new metal and tools can totally mess up the game balance if the distribution of ores is different from the game that the mod creator used.
Could be, yes. But that's the job of whoever adds the mod to the game. Server owners and players ought to be free to choose! Game creators make suggestions (like modders do), but if someone wants to play a survival game in creative - so be it.
firefox wrote: as i see it, most mods were created to add things to MTG which didn't exist in MTG.
when (or if) new games are created, they most likely already include those or similar things, so they don't need to be compatible with those mods.
Why shall they re-invent the wheel? The idea is to use common libraries and to be able to configure things. If you want to for example use bees and honey and all the stuff in your game, it'd be best if the player could also choose to install another mod that does a very similar thing if the player happens to like that implementation more. Not all mods and games are maintained forever either.
firefox wrote: if someone makes a mod for another game, because the game creator didn't include something that the person wanted, then this mod would be for that game, specifically adjusted to that game's properties.
No. Modders don't create mods for particular games unless asked by the game creator to contribute something specific. It might not make much sense to add the mod to some games, but then - the user is free to do so if he likes.
A list of my mods can be found here.

User avatar
Worldblender
Member
Posts: 66
Joined: Tue Jun 30, 2015 22:12
GitHub: RQWorldblender
In-game: Worldblender

Re: Minetest Game now in maintenance/bugfix-only mode, no feature PRs accepted

by Worldblender » Post

It's somewhat disappointing that there won't be new features coming in at least for a while, but there likely will more positive things coming out of this change than negative. If/when I can get around to modding again, it means that I won't have to catch up with a moving target anymore, knowing that MTG will be stable and predictable. Maybe here could also be another opportunity to try other kinds of gameplay other than the ones provided by MTG.

In my particular case, once I finish with summer school by next week, I will have some time to revisit one of my mods at https://gitlab.com/Worldblender/chchar. I will try to provide some usable default models (most likely recycling SuperTuxKart kart assets) so that people can start using it. Hopefully, this is one step to doing something new, in this case, using arbitrary rigged 3D models instead of the blocky default character all the time.

User avatar
firefox
Member
Posts: 1709
Joined: Wed Jan 14, 2015 07:34
In-game: Red_Fox
Location: Xanadu

Re: Minetest Game now in maintenance/bugfix-only mode, no feature PRs accepted

by firefox » Post

Sokomine wrote:
Sun Aug 02, 2020 20:43
Game developers creating new games (RealTest and MineClone2 for example) tend to name the same thing diffrently. No matter if you call it default:wood, planks:wooden_planks, my_own_namespace:something_like_wood, w:1234 or whatever - if it mostly behaves as wood does, it ought to be named that way. Individual chairs, tables, houses and other things vary - yet we have common names for them to fascilitate communication and life. I want standard names for those things to fascilitate mod interoperability.
in my game, i have 4 variants of plain stone. they are exactly the same, except for their colour.
on the same principle, i will include 4 colours of gravel and sand, as well as cobble walls, bricks, sandstone, ect.
of course they must have different names, so for recipes that are not colour based i would use group names to identify "stone", "cobble" or "bricks" regardless of colour.
would this be what you wanted?
otherwise, if i had to the assign aliases based on MTG to support mods, then more than half of the nodes in my game would not be usable even though it's basically the same type of material...
✨🏳️‍🌈♣️✨

User avatar
Lone_Wolf
Member
Posts: 2576
Joined: Sun Apr 09, 2017 05:50
GitHub: LoneWolfHT
IRC: LandarVargan
In-game: LandarVargan

Re: Minetest Game now in maintenance/bugfix-only mode, no feature PRs accepted

by Lone_Wolf » Post

Sokomine wrote:
Sun Aug 02, 2020 20:43
Modders don't create mods for particular games unless asked by the game creator to contribute something specific.
Ahem.
None of those mods were requested by the game creator (Except for the ones created by the creator I guess).
Only two of them will work with another game, and good luck porting the others.
You could probably license a few of the ports as your own mod without legal issues since you'd have to change so much

That's a non-MTG example. If you want a MTG example I'll direct you to all mods with a `default` dependency (Unfortunately probably 80% of them)
My ContentDB -|- Working on CaptureTheFlag -|- Minetest Forums Dark Theme!! (You need it)

User avatar
Linuxdirk
Member
Posts: 3217
Joined: Wed Sep 17, 2014 11:21
In-game: Linuxdirk
Location: Germany
Contact:

Re: Minetest Game now in maintenance/bugfix-only mode, no feature PRs accepted

by Linuxdirk » Post

This sounds just wonderful!

I hope that will free some dev time for more important things! :)

Sokomine
Member
Posts: 4276
Joined: Sun Sep 09, 2012 17:31
GitHub: Sokomine
IRC: Sokomine
In-game: Sokomine

Re: Minetest Game now in maintenance/bugfix-only mode, no feature PRs accepted

by Sokomine » Post

firefox wrote: otherwise, if i had to the assign aliases based on MTG to support mods, then more than half of the nodes in my game would not be usable even though it's basically the same type of material...
Yes, that's certainly a problem if more than one node qualifies. Just having an alias for one would be enough to ensure general compatibility. But perhaps we need more, so that such situations can be covered as well? Recipes may use groups - or be provided with a list of alternate node names. Some nodes could also contain the information that they don't really exist in that game and are just provided for compatibility.

What I want is a mechanism for game creators to tell mods that their version of a default node is called so-and-so and that mods thus don't have to bother about each individual name for more or less the same thing in diffrent games. Recipes could be translated automaticly, and some mods might be intrested in which alternatives of a type of node are available. Wood/trees happen to come in diffrent flavors in diffrent games already, and dirt with grass is just one of its kind - there are lots of diffrent versions.
Lone_Wolf wrote: Ahem.
None of those mods were requested by the game creator (Except for the ones created by the creator I guess).
Only two of them will work with another game, and good luck porting the others.
Well, it's certainly nice if other games get support by mods as well and attract modders. Nodecore seems to have esacaped me so far. I joined two of its servers now and must say that it's not the kind of game I personally like to play. But that doesn't matter. It's fine if there are players and modders out there who enjoy it!

In some way, there ought to be no problem to allow other mods to be added to Nodecore and be able to work with it. The usual nodes are there - trees, leaves, dirt, water, dirt with grass, stone, ... Gameplay just seems to be so vastly diffrent that adding a standard mod to Nodecore wouldn't make much sense, and it seems unlikely that anyone would want to do so. It's not a game where there's placed value on nicely built structures (players are not given the means to do that) - all seems to be ultra-hardcore. Yet some of its nodes (i.e. that tree ladder and the shelves) are nice and might be of intrest for other games as well.
Lone_Wolf wrote: That's a non-MTG example. If you want a MTG example I'll direct you to all mods with a `default` dependency (Unfortunately probably 80% of them)
I do consider "default" to be a standard to which games ought to adhere as far as they can - so there's nothing wrong with that.

Perhaps there are enough mechanisms out there already and all it takes is some standarization and a few interface functions. Dirt with grass has the spreading_dirt_type group, there are tree and wood groups...However, relying on groups alone may not be sufficient as there's no sufficient guarantee that the order of registration will remain the same.
A list of my mods can be found here.

User avatar
paramat
Developer
Posts: 3700
Joined: Sun Oct 28, 2012 00:05
GitHub: paramat
IRC: paramat
Location: UK

Re: Minetest Game now in maintenance/bugfix-only mode, no feature PRs accepted

by paramat » Post

Sokomine,

You seem to be expecting that a mod you write should work in most or all games, which makes no sense. Different games are often completely different universes, with different ways of doing things. There should be no expectation that what works in one universe should work in another.

You seem to expect game makers to standardise for the nonsensical expectation of a mod working in most or all games. This is unreasonable.
It is good that new games are starting fresh and doing things in new and incompatible ways.

This attitude seems to have arisen from the current focus on MTG, where unfortunately most games are MTG plus a few random mods, in which case there is some compatibility between them and mods can be used in multiple games.
We need to get away from this way of thinking about games and mods, this is precisely the big problem here i am trying to address.

What you are requesting is fine if done unofficially and independently by whatever small groups of games want to do this, content makers are free to do what they want so there is nothing stopping this.
This might work for small groups of game makers whose games are similar.
Of course, the many games based on MTG are already standardised and have 'libraries' in the form of APIs.

However you seem to be suggesting this is 'officially decided and enforced for most or all games'. This is where your requests become ridiculous and completely unacceptable.
It is also unenforceable, because content makers are free and most would rightly refuse.
This would also be a lot of work, and imagine the endless arguments over what the standards or libraries actually are. The engine core devs would not have time for this.

Besides, a primary intention of freezing MTG is that MT is no longer officially involved in content creation. So official MT should have no say in any possible game standardisation or libraries, that is for content makers to organise if they want to.

The primary mistake in all this is the expectation that mods should work in most or all games. This attitude arises from the current lack of variety in many games and the unfortunate focus on MTG.
No modder should expect or even want a mod to work in most or all games, because, given a reasonable variety of games, many games will be so different they would naturally be unsuitable for that mod.

Mods should be written for the game they are intended for, and possibly also for any very similar games.
Each game is often a completely different universe.
Occasionally, it might be the case that small groups of games are similar or standardised, in which case it might be possible for a mod to work in multiple games. This should be seen as a lucky exception, not the norm.

I also want to say 'do not worry', because the MT community will still, unfortunately, have a heavy focus on MTG for a long time. Many servers, games and mods are based on it, so there will continue to be the ability for mods to work in multiple games.
MTG-based content already has standardisation and libraries (APIs), so you already have what you want in a large number of games, it is wrong to try to standardise games that are trying to be original.

Standardisation is the opposite of what MT games need, we need more diversification and more incompatibility, because maximising variety requires incompatibility.
Sokomine wrote:
Thu Aug 06, 2020 02:22
I do consider "default" to be a standard to which games ought to adhere as far as they can - so there's nothing wrong with that.
This is a very wrong statement.

The MTG 'default' mod is a structural disaster, the only reason it remains like that is because of how much depends on it remaining that way, and how much complexity would result from splitting it into logically divided mods.
As for its content, there is a lot of high-quality and reliable code there that is a good example. But there are also some very bad implementations we cannot remove, which games should avoid.
'default' is also a nonsensical name for a mod, no mod should ever be called 'default'.

User avatar
duane
Member
Posts: 1715
Joined: Wed Aug 19, 2015 19:11
GitHub: duane-r
Location: Oklahoma City
Contact:

Re: Minetest Game now in maintenance/bugfix-only mode, no feature PRs accepted

by duane » Post

I have no dog in this race anymore, but let me make a prediction, which will almost certainly turn out to be wrong in every important respect.

I predict that this move will increase fragmentation in minetest mods and make life harder for both players (new and old) and mod developers. There will be a lot of chaos, which some will thrive on, but most will just be annoyed by. Then, at some point in the future, someone or some group will put their collective foot down and enforce standards on mods -- which will basically turn into a different version of "minetest game as the default".

Have fun with that. : )
Believe in people and you don't need to believe anything else.

User avatar
rubenwardy
Moderator
Posts: 6972
Joined: Tue Jun 12, 2012 18:11
GitHub: rubenwardy
IRC: rubenwardy
In-game: rubenwardy
Location: Bristol, United Kingdom
Contact:

Re: Minetest Game now in maintenance/bugfix-only mode, no feature PRs accepted

by rubenwardy » Post

This decision will have practically no difference, because MTG was barely being developed before this
Renewed Tab (my browser add-on) | Donate | Mods | Minetest Modding Book

Hello profile reader

Sokomine
Member
Posts: 4276
Joined: Sun Sep 09, 2012 17:31
GitHub: Sokomine
IRC: Sokomine
In-game: Sokomine

Re: Minetest Game now in maintenance/bugfix-only mode, no feature PRs accepted

by Sokomine » Post

paramat wrote: You seem to be expecting that a mod you write should work in most or all games
Probably not in all - but at least in most and in as many as possible.
paramat wrote: Different games are often completely different universes, with different ways of doing things
To the player, gameplay may vary considerably, but behind the scenes most games share a lot of mechanisms.
paramat wrote: However you seem to be suggesting this is 'officially decided and enforced for most or all games'. This is where your requests become ridiculous and completely unacceptable.
Not necessarily decided by the core devs (those certainly do have enough work already!), but yes, decided as far as possible by consensus and provided as an easy way to access a collection of libs.
paramat wrote: This would also be a lot of work, and imagine the endless arguments over what the standards or libraries actually are.
Yes, I'm afraid so. But in the end that's life.
paramat wrote: The primary mistake in all this is the expectation that mods should work in most or all games. This attitude arises from the current lack of variety in many games and the unfortunate focus on MTG.
No modder should expect or even want a mod to work in most or all games, because, given a reasonable variety of games, many games will be so different they would naturally be unsuitable for that mod.
We seem to have a very diffrent opinon here.

Most games we see share a lot of concepts behind the scenes, i.e. trees, doors, dirt with grass, plants, farming, mobs, transparent blocks for windows, ores, processing ores, crafting, a player's inventory, storage systems (some form of chests), tools, blocks to build structures out of, vehicles etc. The actual gameplay may vary considerably, ranging from a purely creative server to a world where PvP is the main focus and players don't even change the world, from conventional farming to space exploration, from a world without farming and mobs to an rpg adventure game up to a puzzle game.

Some of these common concepts are implied or even enforced by the engine and the api. Some others may indeed be motivated by MTG.

That these common concepts exist doesn't mean that each game has to support all of them or even all of them in the same way. But there ought to be a reason for derivating from common concepts. "Oh, it'd be so cool to replace each second letter 'e' with '('" *might* be something a child might think, and you *may* get away with it by declaring it as art - but actually doing so would soon reveal that that's not such a good and helpful way of dealing with words as it needlessly makes communication more difficult. There's a reason standards exist in RL.

Standards also aren't everything. If a game would be a complete and true simulation of life at normal speed, what would be the point of it in the end?

A good game may derivate from RL in some parts, it ought to abstract things. Most of this is caused by computational and dev time limits, but that's not all. "What if (we change that common concept)" is also a huge factor in art and game development. "Why is that (common concept) so" is a question that may lead to new insights in sience.

The very way voxel games work - by turning the world in blocks of 1x1x1 m - that is an abstraction of RL and something that gave rise to a lot of games and created a lot of fun for many players.

But that same abstraction also was filled with things people know from RL. It is seldom realistic, yet players mostly fill the world with something they know about. Gameplay-wise, there's no reason to build a house at all in most MTG-based games. No matter what dangers you add, a cave or a hole in the ground or something similar will suffice. And likewise, on purely creative servers, you may occasionally find players who build abstract art.

Some games may derivate from common standards so much that they may be seen as art, but...they may not be the best games. It does take some balance between new, unexpected elements and common elements for the players to feel happy with beeing challenged by new challenges and similarily feel sufficiently at home to be able to interact with the game.

New textures may already change the way an MT world is percived by its visitors considerably.

Just renaming "default:dirt" to "blablub:something" isn't art. It doesn't enhance gameplay in any way. It is just a derivation from standards and ought to be frowned upon.

Games and mods still ought to be free to derivate from standards - and even encouraged to some degree - but there ought to be some thought behind it, a reason for doing so. No pointless derivating from all standards just to be diffrent without even appearing diffrent in gameplay to the player in the end!
paramat wrote: The MTG 'default' mod is a structural disaster, the only reason it remains like that is because of how much depends on it remaining that way, and how much complexity would result from splitting it into logically divided mods.
That's just how it is with almost anything that has developped over some time. The default mod from MTG certainly isn't ideal.
paramat wrote: 'default' is also a nonsensical name for a mod, no mod should ever be called 'default'.
That's right. default - that's usually a fallback if nothing else is specified. That's diffrent from what I want here.

In MT, each server owner and even each player who selects mods for his singleplayer game is to some very very limited degree a game developer. Players are free to choose which mods they want to be active in their world. That ought not to be hindered by game developers not liking any particular mod and making their games deliberately incompatible with it. It may not make much sense to have a "creative" mod in an extreme-survival-world - but then, it's up to the player to decide. This is open source - not some take-it-as-is!
A list of my mods can be found here.

User avatar
sorcerykid
Member
Posts: 1841
Joined: Fri Aug 26, 2016 15:36
GitHub: sorcerykid
In-game: Nemo
Location: Illinois, USA

Re: Minetest Game now in maintenance/bugfix-only mode, no feature PRs accepted

by sorcerykid » Post

paramat wrote:Standardisation is the opposite of what MT games need, we need more diversification and more incompatibility, because maximising variety requires incompatibility.
I fully agree with this point made by paramat. Dare I say, part of the reason why there is so little diversity across Minetest games and servers is because people expect an underlying game standard. It's good that we can finally begin to move away from a de facto "Minetest Game", and allow for many different and unrelated games to evolve and mature on their own.

User avatar
Lone_Wolf
Member
Posts: 2576
Joined: Sun Apr 09, 2017 05:50
GitHub: LoneWolfHT
IRC: LandarVargan
In-game: LandarVargan

Re: Minetest Game now in maintenance/bugfix-only mode, no feature PRs accepted

by Lone_Wolf » Post

Mods working across all games would also be annoying for grumps like me who might not want my mod to ever work with the game of someone who I don't like for some random reason
My ContentDB -|- Working on CaptureTheFlag -|- Minetest Forums Dark Theme!! (You need it)

User avatar
celeron55
Administrator
Posts: 532
Joined: Tue Apr 19, 2011 10:10
GitHub: celeron55
IRC: celeron55

Re: Minetest Game now in maintenance/bugfix-only mode, no feature PRs accepted

by celeron55 » Post

I don't think there's need for this much polarization.

Assuming different games don't want to call their nodes and items the same, we can (should?) define an API that games can implement that allows a mod to know the names of some very common things, like those that Sokomine listed. The API can also reveal that the game doesn't have something, in the case it doesn't have something. The simplest realization of this could just be a bunch of global constant variables.

Many libraries with functions can already be built on top of that. Of course libraries naturally get game specific more often as different games have different systems, so not all can be interoperable with everything. It wouldn't even make any sense.

User avatar
Dokimi
Member
Posts: 227
Joined: Mon Sep 19, 2016 03:52
GitHub: DokimiCU

Re: Minetest Game now in maintenance/bugfix-only mode, no feature PRs accepted

by Dokimi » Post

I have an idea: "Gold Standard" games.

These would be games that met certain quality criteria
e.g.
- developer approval
- popularity
- active development
- unique concept
- quality coding and content
-etc...

They would be labelled (e.g. on forums, ContentDB) like a Quality Assurance mark.

Why?
1) it is hard for new people to know what is good. Hard to know what games to bother making mods for, which to bother making pull requests etc.
2) engine development needs to be guided by what the engine gets used for. The "Gold Standard" games would be what Dev's have in mind when making decisions.
3) This directs effort, but avoids the "locked in" exclusive focus we had with Minetest Game.

Would this idea work???

ShadMOrdre
Member
Posts: 1118
Joined: Mon Dec 29, 2014 08:07
Location: USA

Re: Minetest Game now in maintenance/bugfix-only mode, no feature PRs accepted

by ShadMOrdre » Post

Assuming different games don't want to call their nodes and items the same, we can (should?) define an API that games can implement that allows a mod to know the names of some very common things, like those that Sokomine listed. The API can also reveal that the game doesn't have something, in the case it doesn't have something. The simplest realization of this could just be a bunch of global constant variables.
I have put thought into a mod that provides a table based ontology, (ie, common nodes, items, objects, tools with a simple name). The ontology can be as simple as providing a key of "stone", with the value pointing to the defined node, "default:stone" or "lib_materials:stone". Modders can then refer to the table using the key of "stone" which simply points to configured node. I have already used this in a couple of mods, to simplify schematics definitions that also provide runtime editing and loading of the schems from a .csv file, and for updating other standard mods for easier updating in general.

I do not have any particular mod available, but code can be found in this project of mine to show the potential.

The idea is not limited to my use case, but could easily be integrated into a "Game Abstraction Layer", which is another mod on which I am working, lib_gal, (code not yet released).

Just my thoughts. MTG is not dead, just less focus is being placed on it. Nothing wrong there, as it should encourage us modders and devs to think outside the cube....

Maybe a seperate dev group should be officially formed for the creation of a game development minimum. Not a test case, not a minipelli, but an official minimal development framework, something that provides only what the engine requires, with a simple API for accessing the world and the player. The craft grid and inventory, being a rather game specific concept, should be moved, along with formspecs, healthbar and the rest of the HUD, into a new Interface API, provided by the engine, and simplifying that particularly troublesome area into a singular point of access to all things related to the "game" interface.

Engine dev and "content" dev should be completely seperated. It cannot yet truly be called an engine, if it relies on any particular "user based" content, even if the "user base" providing the content is the core dev team. No offense to anyone there, just another unrequested observation from another IT "expert".

Shad

Post Reply

Who is online

Users browsing this forum: No registered users and 4 guests