Anyone wanna help with an idea [cancelled]

User avatar
PolySaken
Member
Posts: 817
Joined: Thu Nov 09, 2017 05:18
GitHub: PolySaken-I-Am
In-game: PolySaken
Location: Wānaka, Aotearoa
Contact:

Anyone wanna help with an idea [cancelled]

by PolySaken » Post

I've been gone for a while (around two months), mostly because I've been frustrated with minetest's impracticality and lack of direction.
While I've been away I've been doing a lot of thinking and minetest, and about programming and game/engine design in general.
I want to make a fork of minetest that functions as a general-purpose voxel engine, which minetest is clearly not intended to be and never will. My main concerns are these:
  • Minetest is specialized. It has far too many game-specific features, all of which need to go.
  • Minetest is poorly managed. It has no clear goal and currently exists as a list of ideas, nothing more.
  • Minetest is dearly missing many essential features. most of which will never be added.
  • Minetest is too entrenched in 'tradition' to fix many things, for example the very limiting lua-only API.*
  • Minetest is horribly niche and old fashioned in terms of branding and appeal. It's unpolished in terms of the application, the sites, and its entire presence. (except for the CDB of course.)
  • Minetest has horrible settings management, and user friendliness in general. Setting up a game and creating a world with mods and settings feels eerily like managing an SQL database, which after doing so for two years I can tell you no-one should ever have to do.
There are many more specific issues that I won't cover here, but when I get the fork set up there will be a priority master list.

So, now we get to the point of this topic. I can do this alone, but it will be slow and very annoying because of all the things I have yet to learn. So I'd like to know if anyone's interested in this proposal, before I even start planning properly so I can account for that in the roadmap.

Lastly, this project will not be non-profit. Any team members will receive a cut from whatever ads or donations the project receives, when I get that set up.
I love Lua. Coming from a background of classic maths and logic notation, lua is very close to the ideal scripting language from my point of view. However, I also love Python, JS, and many other scripting languages.
It's obviously one of the most long-term goals, but getting one or more other modding languages working would be amazing. The main downside of lua for me at least is the lack of huge module population that python and others have. Being able to write the main content for a mod/game in lua and then the menus and forms in JS, and then the file control and data management in python would be a dream come true.

(something even crazier that I've thought about doing before, and even attempted once, is developing a new syntax format to streamline developing minetest content. I've gotta say, it looks promising.)
Last edited by PolySaken on Mon Aug 02, 2021 20:57, edited 1 time in total.
Guidebook Lib, for in-game docs | Poly Decor, some cool blocks | Vision Lib, an all-purpose library.

User avatar
philipbenr
Member
Posts: 1897
Joined: Fri Jun 14, 2013 01:56
GitHub: philipbenr
IRC: philipbenr
In-game: robinspi
Location: United States

Re: Anyone wanna help with an idea (yep its another fork)

by philipbenr » Post

Definitely not interested in joining as a developer, as I don't have the time or desire, but I wish you luck. Putting some much needed polish is great, and the possibility of using multiple other programming languages for a different API is both wonderful and a bit insane. ;)

Don't forget to clearly set up a framework for how you want to develop with your team, as structure will be your success or downfall. But it sounds like you have that in mind with your 2nd bullet. And as with any software project that is open to more than 2 people, I strongly recommend a robust documentation suite, as that will really get people coming to you and ready to get messing with the code from both an engine and modding perspective.

The idea is an interesting one, so I hope you and whoever joins you find success and happiness with the project (and don't burn out into a steaming mess).

User avatar
PolySaken
Member
Posts: 817
Joined: Thu Nov 09, 2017 05:18
GitHub: PolySaken-I-Am
In-game: PolySaken
Location: Wānaka, Aotearoa
Contact:

Re: Anyone wanna help with an idea (yep its another fork)

by PolySaken » Post

hehe thanks. Team structure and documentation are the most important of all my points, considering the eldritch ways of the MT core dev team, and the heinously bad documentation of the lua api.

A wiki that's just plain wrong, a list of all the docs in a single messy file, and many user-made attempts at docs with varying degrees of success are all minetest issues that i intend to do away with.
Guidebook Lib, for in-game docs | Poly Decor, some cool blocks | Vision Lib, an all-purpose library.

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

Re: Anyone wanna help with an idea (yep its another fork)

by Wuzzy » Post

Sorry, not interested in that. I'm just not convinced by this.

First of all, I think the idea of supporting multiple scripting languages is terrible. Maintaining Lua alone is already effort enough. Supporting two or even more will significantly . It also increases difficulty for modders because suddenly you will see mods in different languages, depending on different APIs that might be mostly the same but will probably still have that one quick that will trip people up.

This sounds more like a personal preference than anything else. Lua is an acceptable scripting language which gets the job done. I don't see the need to add another one. If you look hard enough, there is always some minor nitpick you will find about the programming language, but I am extremely bored by these kinds of debates. Just learn the damn language. I don't understand how the Lua API is limited because it's Lua. This doesn't make sense.

What this will do is just exponentionally increase the complexity of both API and mods themselves for absolutely no good reason.
Minetest is specialized. It has far too many game-specific features, all of which need to go.
I actually do think this is a good thing.

While I think that some features are WAY too specific, I wouldn't say that all of them have to go. The problem is IMO more that many things are a bit too hardcoded and can't be changed. This can be fixed tho.

Anyway, a game engine that is specific towards a certain genre only is still perfectly valid, game engines don't need to be 100% generic to count as game engines. Have you heard of FLARE? However, I still think the long-term goal for Minetest should be to generalize away needlessly hardcoded features.
Minetest is poorly managed. It has no clear goal and currently exists as a list of ideas, nothing more.
Yes. But how do you want to fix that? When I think about it, the implementation of a clear roadmap should be the number 1 priority for Minetest right now. Everything else depends on it. Nobody disagrees we need a roadmap.

On the other hand, I think the current PR pipeline has ONE benefit. Minetest is stable. I can't remember when Minetest segfaulted the last time. When I think about the last major fork, Freeminer, it constantly crashed, it wasn't funny anymore. They went the other extreme, in just allowing everything into the codebase. Do not underestimate the importance of QA!
Minetest is dearly missing many essential features. most of which will never be added.
That's pretty vague tho.
Minetest has horrible settings management, and user friendliness in general. Setting up a game and creating a world with mods and settings feels eerily like managing an SQL database, which after doing so for two years I can tell you no-one should ever have to do.
I agree. IMO the entire main menu needs to be nuked and be done from scratch … mostly. I think at least the FSTK might be reusable, idk. The main menu should be redone in one go, ideally, instead of incrementally. This is not something that can be fixed with minor tweaks with tiny PRs.

What is your plan to redo the main menu?
A wiki that's just plain wrong, a list of all the docs in a single messy file, and many user-made attempts at docs with varying degrees of success are all minetest issues that i intend to do away with.
Are you referring to the Dev Wiki? Because then you're correct. The fix is actually simple in princinple: Delete everything that was just ripped from lua_api.txt because of redundancy. That's the one big mistake in Dev Wiki right now. The thing is, we haven't figured out yet how to mass-delete (or at least mass-add-a-warning-sign) pages. Apart from that, I don't see where the wiki is wrong. If you have specific complaints, head over to the Wiki Complaint Department: viewtopic.php?f=6&t=23734

I think you have missed one of the biggest problems with MT IMO: The lack of reviewers with approval rights. This is why development is so slow. There are so many PRs that would greatly benefit Minetest, it's just the approval process is painfully slow. Minetest does NOT have a shortage in contributors.


To summarize: Your proposal is not convincing. Many of the problems you want to solve are already in Minetest's interest to be solved. We all know about these problems, it's not exactly news. But multi-scripting support is a definite no-go for me.
Additionally, you didn't say much about how your fork will be different, especially with regards to organization.


Finally, I also think that you can work on Minetest (or a fork of it) all you want, we should never forget that the REAL end goal is to actually end up with nice games. I think that's very important to keep in mind.

EDIT: Relevant post: viewtopic.php?f=3&t=24742

User avatar
PolySaken
Member
Posts: 817
Joined: Thu Nov 09, 2017 05:18
GitHub: PolySaken-I-Am
In-game: PolySaken
Location: Wānaka, Aotearoa
Contact:

Re: Anyone wanna help with an idea (yep its another fork)

by PolySaken » Post

Wuzzy wrote:
Thu Apr 08, 2021 00:21
First of all, I think the idea of supporting multiple scripting languages is terrible. Maintaining Lua alone is already effort enough. Supporting two or even more will significantly . It also increases difficulty for modders because suddenly you will see mods in different languages, depending on different APIs that might be mostly the same but will probably still have that one quick that will trip people up.

This sounds more like a personal preference than anything else. Lua is an acceptable scripting language which gets the job done. I don't see the need to add another one. If you look hard enough, there is always some minor nitpick you will find about the programming language, but I am extremely bored by these kinds of debates. Just learn the damn language. I don't understand how the Lua API is limited because it's Lua. This doesn't make sense.

What this will do is just exponentionally increase the complexity of both API and mods themselves for absolutely no good reason.
Lua is limited by its lack of libraries, its uselessness with unicode, and its horrible file management.
Its a great language, but it has drawbacks.
Having multiple is obviously not an easy thing to do, but with a well thought out management system that is actually followed most of those issues evaporate.
Multi scripting is not a priority for me, but itd be nice to at least modify the lua version to better fit the engine. That's what lua is good for, after all: being adapted.
Wuzzy wrote:
Thu Apr 08, 2021 00:21
I actually do think this is a good thing.

While I think that some features are WAY too specific, I wouldn't say that all of them have to go. The problem is IMO more that many things are a bit too hardcoded and can't be changed. This can be fixed tho.

Anyway, a game engine that is specific towards a certain genre only is still perfectly valid, game engines don't need to be 100% generic to count as game engines. Have you heard of FLARE? However, I still think the long-term goal for Minetest should be to generalize away needlessly hardcoded features.
This is just preference. Minetest is a more specialized engine, mine will not be.
Wuzzy wrote:
Thu Apr 08, 2021 00:21
Yes. But how do you want to fix that? When I think about it, the implementation of a clear roadmap should be the number 1 priority for Minetest right now. Everything else depends on it. Nobody disagrees we need a roadmap.

On the other hand, I think the current PR pipeline has ONE benefit. Minetest is stable. I can't remember when Minetest segfaulted the last time. When I think about the last major fork, Freeminer, it constantly crashed, it wasn't funny anymore. They went the other extreme, in just allowing everything into the codebase. Do not underestimate the importance of QA!
Yes, a roadmap and good pr management are both key ideas in my planning document. However i plan to address all existing prs before accepting any new ones. In essence I'm just going to give more attention than the core devs currently do.
Wuzzy wrote:
Thu Apr 08, 2021 00:21
That's pretty vague tho.
It is indeed, because there are hundreds and hundreds of such missing features.
I can give a few examples: colored lighting, keyframe animations, metadata/paramtype textures for nodes.
Wuzzy wrote:
Thu Apr 08, 2021 00:21
I agree. IMO the entire main menu needs to be nuked and be done from scratch … mostly. I think at least the FSTK might be reusable, idk. The main menu should be redone in one go, ideally, instead of incrementally. This is not something that can be fixed with minor tweaks with tiny PRs.

What is your plan to redo the main menu?
My plan is to get rid of it. Give me one example of a real game where the engine defines the menu. Instead there will be a small game selector, and a way to access global settings and a content manager. The rest will be up to the game, including everything from managing worlds to changing game settings.
The servers tab will be replaced with an online play menu where you can sort by games, mods, etc. And actually pick favorites, accessed from the top of the game selector.
Wuzzy wrote:
Thu Apr 08, 2021 00:21
I think you have missed one of the biggest problems with MT IMO: The lack of reviewers with approval rights. This is why development is so slow. There are so many PRs that would greatly benefit Minetest, it's just the approval process is painfully slow. Minetest does NOT have a shortage in contributors.
This is an issue, yes. Essentially I'm going to have a constant backup system so it's easy to quickly revert any bad changes, allowing me to be more liberal with who can approve and merge prs.
Wuzzy wrote:
Thu Apr 08, 2021 00:21
To summarize: Your proposal is not convincing. Many of the problems you want to solve are already in Minetest's interest to be solved. We all know about these problems, it's not exactly news. But multi-scripting support is a definite no-go for me.
Additionally, you didn't say much about how your fork will be different, especially with regards to organization.
Most of these issues are talked about in minetest but wont be fixed for years, if ever. My fork will be different in that its a voxel game engine, not a copy of minecraft engine. Some games may not use map generation, or an fps style control scheme, or crafting. These things are annoying to work around in minetest and sometimes impossible.
Wuzzy wrote:
Thu Apr 08, 2021 00:21
Finally, I also think that you can work on Minetest (or a fork of it) all you want, we should never forget that the REAL end goal is to actually end up with nice games. I think that's very important to keep in mind.
Of course, and its hard to find a game nice when theyre all so similar.
Guidebook Lib, for in-game docs | Poly Decor, some cool blocks | Vision Lib, an all-purpose library.

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

Re: Anyone wanna help with an idea (yep its another fork)

by Wuzzy » Post

Don't get me wrong: I appreciate the effort, but I have my doubts. For this to be going anywhere, several conditions need to be met:

- Drop the multi-scripting, this will end up in desaster or at least will just be a huge time-waster with no real benefit. This whole idea is basically just to give devs something to do, instead of doing something that gives results. Players really couldn't care less about this.
- You need to convince enough people to join your project since you probably don't want to do it all alone
- A solution to the PR bottleneck problem
- A roadmap with clearly defined goals and non-goals
- Think about compability: Do you want it or not?
- Think about how it will differ from Minetest: What will be your unique "selling point" that will move developers and players over? What will be your "killer features"?
- Infrastructure (a website and a master server at minimum)
- Do you have any idea for a concrete game that you would want to make?
- And finally, what the heck is even your end goal with this?

Seeing that there is still no solid plan and it's all very vague, I'm not convinced at all, sorry.


One of the reasons why Freeminer crashed and burned is because it never attraced much interest from the Minetest community; a lot of Freeminer's features were so obscure that players and game developers couldn't care less about them.
Yes, a roadmap and good pr management are both key ideas in my planning document. However i plan to address all existing prs before accepting any new ones. In essence I'm just going to give more attention than the core devs currently do.
Ooff. That's a huge commitment, because we have >100 open PRs right now.

If you're THAT committed, have you considererd to become a Minetest core dev yourselves? Minetest REALLY needs some help with the PR reviewing process right now, currently only 2-3 core devs are actively reviewing. Even just ONE new official PR reviewer would be invaluable to the Minetest project. I heard celeron55 is pretty willing in accepting new core devs.
My plan is to get rid of it. Give me one example of a real game where the engine defines the menu. Instead there will be a small game selector, and a way to access global settings and a content manager. The rest will be up to the game, including everything from managing worlds to changing game settings.
The servers tab will be replaced with an online play menu where you can sort by games, mods, etc. And actually pick favorites, accessed from the top of the game selector.
Yeah, my idea for a new main menu is more or less like that. But it's not like the core devs are opposed to a main menu rework ...

User avatar
PolySaken
Member
Posts: 817
Joined: Thu Nov 09, 2017 05:18
GitHub: PolySaken-I-Am
In-game: PolySaken
Location: Wānaka, Aotearoa
Contact:

Re: Anyone wanna help with an idea (yep its another fork)

by PolySaken » Post

Wuzzy wrote:
Thu Apr 08, 2021 22:19
- Drop the multi-scripting, this will end up in desaster or at least will just be a huge time-waster with no real benefit. This whole idea is basically just to give devs something to do, instead of doing something that gives results. Players really couldn't care less about this.
As I said, the Idea of multi-scripting or a custom modding syntax is not at all a priority, and If it's implemented it will be done on the side as perhaps even a separate branch.
Wuzzy wrote:
Thu Apr 08, 2021 22:19
- You need to convince enough people to join your project since you probably don't want to do it all alone
- A solution to the PR bottleneck problem
I'm prepared to do it alone, as was my initial plan. I'm not worried about convincing people to join as I'm promising a lot of things that people have been asking for for years and will likely take a long time to make their way into minetest.
Wuzzy wrote:
Thu Apr 08, 2021 22:19
- A roadmap with clearly defined goals and non-goals
I'm currently developing one now. I'm not going to start the project for a while yet, I'm going to spend at least a month setting everything up and just planning.
Wuzzy wrote:
Thu Apr 08, 2021 22:19
- Think about compability: Do you want it or not?
- Think about how it will differ from Minetest: What will be your unique "selling point" that will move developers and players over? What will be your "killer features"?
In an ideal world a developer would be able to drop an API compatibility package into their game and have it work perfectly in my engine, however this is wishful thinking. MT games will probably work most of the time, and it shouldn't be too difficult to fix most compatibility issues.
As of yet I haven't come up with a 'selling point', but if I had to pick one at this point it would likely be the greater flexibility. I don't intend for anything to be hardcoded that doesn't need to be; for example the player object is set up by default to be exactly what is needed for MTG, with inventories and such.
Wuzzy wrote:
Thu Apr 08, 2021 22:19
- Infrastructure (a website and a master server at minimum)
- Do you have any idea for a concrete game that you would want to make?
- And finally, what the heck is even your end goal with this?
First one is pretty easy, I can have that set up soon.
A game I plan to make is a story-driven combat focused rpg set in one of my fictional worlds.
The end goal is to have the best general purpose voxel game engine freely available.
Wuzzy wrote:
Thu Apr 08, 2021 22:19
Seeing that there is still no solid plan and it's all very vague, I'm not convinced at all, sorry.
That's fine, I didn't expect you to want to join.
Wuzzy wrote:
Thu Apr 08, 2021 22:19
Ooff. That's a huge commitment, because we have >100 open PRs right now.

If you're THAT committed, have you considererd to become a Minetest core dev yourselves? Minetest REALLY needs some help with the PR reviewing process right now, currently only 2-3 core devs are actively reviewing. Even just ONE new official PR reviewer would be invaluable to the Minetest project. I heard celeron55 is pretty willing in accepting new core devs.
I have before. It seemed like a tricky process and I don't have a reason to be that invested. (I plan to develop a commercial game for my engine, while that's impossible with MT)
Guidebook Lib, for in-game docs | Poly Decor, some cool blocks | Vision Lib, an all-purpose library.

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: Anyone wanna help with an idea (yep its another fork)

by rubenwardy » Post

PolySaken wrote:
Thu Apr 08, 2021 23:00
I have before. It seemed like a tricky process and I don't have a reason to be that invested. (I plan to develop a commercial game for my engine, while that's impossible with MT)
It's certainly possible, and has been done before - see kidscode. A fork won't change anything about the license, as for development you have the benefit of being able to break compat and tie the engine to you particular game and the features you need I guess
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: Anyone wanna help with an idea (yep its another fork)

by Wuzzy » Post

Well then. I guess if you're prepared to do it all alone and already have a specific game in mind, this changes everything. I guess I better take back my complaints. Instead, I wish you the best of luck on your journey and I'll just watch it unfold from a distance. I still don't really know what precisely you want to end up with but maybe you'll sort it all out eventually. It's nice to hear you're already working on a roadmap.

It is still interesting to see Minetest side projects.

User avatar
PolySaken
Member
Posts: 817
Joined: Thu Nov 09, 2017 05:18
GitHub: PolySaken-I-Am
In-game: PolySaken
Location: Wānaka, Aotearoa
Contact:

Re: Anyone wanna help with an idea (yep its another fork)

by PolySaken » Post

rubenwardy wrote:
Thu Apr 08, 2021 23:29
PolySaken wrote:
Thu Apr 08, 2021 23:00
I have before. It seemed like a tricky process and I don't have a reason to be that invested. (I plan to develop a commercial game for my engine, while that's impossible with MT)
It's certainly possible, and has been done before - see kidscode. A fork won't change anything about the license, as for development you have the benefit of being able to break compat and tie the engine to you particular game and the features you need I guess
The licensing isnt the issue. With 'vanilla' minetest, games can only be played through a visually poor engine that can only be safely obtained either by downloading it from a website that google flags as unsafe, or by getting it from github. This is obviously extremely unconducive to a commercial game.
I intend to create a single-game version of the engine, allowing developers to package it with their game and treat it the same way proper standalone games are treated.
I.E. the game boots up to a proper main menu without any weird built in engine stuff like a game selector.
Guidebook Lib, for in-game docs | Poly Decor, some cool blocks | Vision Lib, an all-purpose library.

DOOM_possum
Member
Posts: 172
Joined: Sat Mar 27, 2021 22:06
In-game: DOOM_possum

Re: Anyone wanna help with an idea (yep its another fork)

by DOOM_possum » Post

sure, everyone likes MAIN MENU re-works, just as long as it doesn't branch off into some sell out project, if creators want to make a departure from the Engine, and brand their own Game, that's up to them, but with little to no backing from big companies, and larger ones too for That matter, the Game, will always remain, a behind the scenes follower gem, i'd want to work on other Things, first, like outlining blockscaping without making them look too fuzzy, unless the Code calls for It

cuthbertdoublebarrel
Member
Posts: 348
Joined: Tue Apr 14, 2020 16:03
GitHub: cuthbert

Re: Anyone wanna help with an idea (yep its another fork)

by cuthbertdoublebarrel » Post

PolySaken wrote:
Wed Apr 07, 2021 21:43
[*] Minetest is specialized. It has far too many game-specific features, all of which need to go.
[*] Minetest is horribly niche and old fashioned in terms of branding and appeal. It's unpolished in terms of the application, the sites, and its entire presence. (except for the CDB of course.)
[*] Minetest has horrible settings management, and user friendliness in general.
I was searching specifically for a specialized engine but I am as mythed as you as to why they are trying to push it as a general purpose engine and insist to continue branding it as a game .
I was hoping someone would suggest a soft split with a legacy engine and alpha engine the latter totally removing focus from minetest and the exsisting mods . is there any reason the we can not have more the one engine version ? other projects do not restrict themselves to just one engine version .
Project BrutalTest...hide your Petz

User avatar
PolySaken
Member
Posts: 817
Joined: Thu Nov 09, 2017 05:18
GitHub: PolySaken-I-Am
In-game: PolySaken
Location: Wānaka, Aotearoa
Contact:

Re: Anyone wanna help with an idea (yep its another fork)

by PolySaken » Post

After messing around with the engine code for the better part of 2 months, I've concluded that there's no point trying to change minetest so drastically. So many things are ingrained toward creating MTG derivatives and clones that I'd basically have to rewrite the entire codebase, and at that point I'd rather just create an entirely new but related project (that has been a plan of mine for years, but it's not a thing I can just do on a whim).
Irrlicht is pretty horrible too, I'd much rather use raylib.
Guidebook Lib, for in-game docs | Poly Decor, some cool blocks | Vision Lib, an all-purpose library.

User avatar
philipbenr
Member
Posts: 1897
Joined: Fri Jun 14, 2013 01:56
GitHub: philipbenr
IRC: philipbenr
In-game: robinspi
Location: United States

Re: Anyone wanna help with an idea (yep its another fork)

by philipbenr » Post

PolySaken wrote:
Thu Jul 01, 2021 21:46
So many things are ingrained toward creating MTG derivatives and clones that I'd basically have to rewrite the entire codebase, and at that point I'd rather just create an entirely new but related project (that has been a plan of mine for years, but it's not a thing I can just do on a whim).
That's what I've come to notice with current Minetest. Despite it being geared around being a game engine, it is surprisingly constricted by the... for lack of a better word "genre" it is targeting. It does lower the barrier to creation within this genre, but it does have tradeoffs

If you do create an alternative, keep us updated, as I'd personally be interested in a lighter-weight, more barebones cubic-voxel engine.

User avatar
PolySaken
Member
Posts: 817
Joined: Thu Nov 09, 2017 05:18
GitHub: PolySaken-I-Am
In-game: PolySaken
Location: Wānaka, Aotearoa
Contact:

Re: Anyone wanna help with an idea (yep its another fork)

by PolySaken » Post

I'll definitely post about it in this topic if/when I put something together, but the degree to which it's 'on-topic' might be debatable if it's just 'inspired' by MT.
However:
Since I consider Lua the gold standard for API languages, I'll probably use Lua with my future endeavours. This might mean that using some sort of compatibility signifier, my engine can detect minetest mods/games and supply a dummy 'minetest' object for them, allowing compatibility.

Edit: I'd rather have OOP as an inherent feature of whatever language I use for API (if not Lua, probably Wren, which is like python but readable), so it'll either be modified lua or something else with a compatibility system.
Guidebook Lib, for in-game docs | Poly Decor, some cool blocks | Vision Lib, an all-purpose library.

User avatar
philipbenr
Member
Posts: 1897
Joined: Fri Jun 14, 2013 01:56
GitHub: philipbenr
IRC: philipbenr
In-game: robinspi
Location: United States

Re: Anyone wanna help with an idea (yep its another fork)

by philipbenr » Post

PolySaken wrote:
Thu Jul 01, 2021 22:25
I'll definitely post about it in this topic if/when I put something together, but the degree to which it's 'on-topic' might be debatable if it's just 'inspired' by MT.
That's what the Minetest-related-projects section is for, if you feel the need to be appropriately on-topic.

User avatar
PolySaken
Member
Posts: 817
Joined: Thu Nov 09, 2017 05:18
GitHub: PolySaken-I-Am
In-game: PolySaken
Location: Wānaka, Aotearoa
Contact:

Re: Anyone wanna help with an idea (yep its another fork)

by PolySaken » Post

Compatibility with minetest would be very useful in that sort of project, especially if it extends to the whole game<mod<texturepack hierarchy, because it prepopulates content rather than starting with none.
Guidebook Lib, for in-game docs | Poly Decor, some cool blocks | Vision Lib, an all-purpose library.

User avatar
philipbenr
Member
Posts: 1897
Joined: Fri Jun 14, 2013 01:56
GitHub: philipbenr
IRC: philipbenr
In-game: robinspi
Location: United States

Re: Anyone wanna help with an idea (yep its another fork)

by philipbenr » Post

PolySaken wrote:
Thu Jul 01, 2021 23:52
Compatibility with minetest would be very useful in that sort of project, especially if it extends to the whole game<mod<texturepack hierarchy, because it prepopulates content rather than starting with none.
I would be careful, as that is a fairly limiting proposal. Alternatively, I would suggest having some form of conversion to your specific form instead of full compatibility. In the case of converting a mod, it would be easier to start with just converting over registered nodes and entities. This way, you have an easier roadmap to work with, and don't have to deal with any of the flaws Minetest may have in the way it does things.

User avatar
PolySaken
Member
Posts: 817
Joined: Thu Nov 09, 2017 05:18
GitHub: PolySaken-I-Am
In-game: PolySaken
Location: Wānaka, Aotearoa
Contact:

Re: Anyone wanna help with an idea (yep its another fork)

by PolySaken » Post

philipbenr wrote:
Fri Jul 02, 2021 01:32
PolySaken wrote:
Thu Jul 01, 2021 23:52
Compatibility with minetest would be very useful in that sort of project, especially if it extends to the whole game<mod<texturepack hierarchy, because it prepopulates content rather than starting with none.
I would be careful, as that is a fairly limiting proposal. Alternatively, I would suggest having some form of conversion to your specific form instead of full compatibility. In the case of converting a mod, it would be easier to start with just converting over registered nodes and entities. This way, you have an easier roadmap to work with, and don't have to deal with any of the flaws Minetest may have in the way it does things.
True. Perhaps I could make a minetest mod that, when a certain command is run, takes the registered content and converts it into a mod for my engine. This would limit the functional aspects of certain things like machines, etc. and only allow recipes, visuals, and basic interaction, however wouldn't require any of minetest's limitations.

Another option is a standalone mod converter software which runs the supplied mod in a fake mt environment and builds a compatible mod from it.
Guidebook Lib, for in-game docs | Poly Decor, some cool blocks | Vision Lib, an all-purpose library.

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

Re: Anyone wanna help with an idea (yep its another fork)

by Mineminer » Post

Wuzzy wrote:
Thu Apr 08, 2021 00:21
Sorry, not interested in that. I'm just not convinced by this.

First of all, I think the idea of supporting multiple scripting languages is terrible. Maintaining Lua alone is already effort enough. Supporting two or even more will significantly . It also increases difficulty for modders because suddenly you will see mods in different languages, depending on different APIs that might be mostly the same but will probably still have that one quick that will trip people up.

This sounds more like a personal preference than anything else. Lua is an acceptable scripting language which gets the job done. I don't see the need to add another one. If you look hard enough, there is always some minor nitpick you will find about the programming language, but I am extremely bored by these kinds of debates. Just learn the damn language. I don't understand how the Lua API is limited because it's Lua. This doesn't make sense.

What this will do is just exponentionally increase the complexity of both API and mods themselves for absolutely no good reason.
Minetest is specialized. It has far too many game-specific features, all of which need to go.
I actually do think this is a good thing.

While I think that some features are WAY too specific, I wouldn't say that all of them have to go. The problem is IMO more that many things are a bit too hardcoded and can't be changed. This can be fixed tho.

Anyway, a game engine that is specific towards a certain genre only is still perfectly valid, game engines don't need to be 100% generic to count as game engines. Have you heard of FLARE? However, I still think the long-term goal for Minetest should be to generalize away needlessly hardcoded features.
Minetest is poorly managed. It has no clear goal and currently exists as a list of ideas, nothing more.
Yes. But how do you want to fix that? When I think about it, the implementation of a clear roadmap should be the number 1 priority for Minetest right now. Everything else depends on it. Nobody disagrees we need a roadmap.

On the other hand, I think the current PR pipeline has ONE benefit. Minetest is stable. I can't remember when Minetest segfaulted the last time. When I think about the last major fork, Freeminer, it constantly crashed, it wasn't funny anymore. They went the other extreme, in just allowing everything into the codebase. Do not underestimate the importance of QA!
Minetest is dearly missing many essential features. most of which will never be added.
That's pretty vague tho.
Minetest has horrible settings management, and user friendliness in general. Setting up a game and creating a world with mods and settings feels eerily like managing an SQL database, which after doing so for two years I can tell you no-one should ever have to do.
I agree. IMO the entire main menu needs to be nuked and be done from scratch … mostly. I think at least the FSTK might be reusable, idk. The main menu should be redone in one go, ideally, instead of incrementally. This is not something that can be fixed with minor tweaks with tiny PRs.

What is your plan to redo the main menu?
A wiki that's just plain wrong, a list of all the docs in a single messy file, and many user-made attempts at docs with varying degrees of success are all minetest issues that i intend to do away with.
Are you referring to the Dev Wiki? Because then you're correct. The fix is actually simple in princinple: Delete everything that was just ripped from lua_api.txt because of redundancy. That's the one big mistake in Dev Wiki right now. The thing is, we haven't figured out yet how to mass-delete (or at least mass-add-a-warning-sign) pages. Apart from that, I don't see where the wiki is wrong. If you have specific complaints, head over to the Wiki Complaint Department: viewtopic.php?f=6&t=23734

I think you have missed one of the biggest problems with MT IMO: The lack of reviewers with approval rights. This is why development is so slow. There are so many PRs that would greatly benefit Minetest, it's just the approval process is painfully slow. Minetest does NOT have a shortage in contributors.


To summarize: Your proposal is not convincing. Many of the problems you want to solve are already in Minetest's interest to be solved. We all know about these problems, it's not exactly news. But multi-scripting support is a definite no-go for me.
Additionally, you didn't say much about how your fork will be different, especially with regards to organization.


Finally, I also think that you can work on Minetest (or a fork of it) all you want, we should never forget that the REAL end goal is to actually end up with nice games. I think that's very important to keep in mind.

EDIT: Relevant post: viewtopic.php?f=3&t=24742
Well said, reinventing the wheel is not a good thing if problems can be solved as straightforward as getting more PRs passed with an additional QA person on deck or so. :)

User avatar
ywwv
Member
Posts: 299
Joined: Mon Jan 18, 2021 11:51

Re: Anyone wanna help with an idea (yep its another fork)

by ywwv » Post

my fork of minetest will include smooth terrain with marching cubes. I have already implemented a lense flare shader. so tell me. why should I help you?

this is an opportunity. I can be your first core developer and right hand man. you just have to sell me . give it your best shot and you just might get a real star <<<--- ME

User avatar
PolySaken
Member
Posts: 817
Joined: Thu Nov 09, 2017 05:18
GitHub: PolySaken-I-Am
In-game: PolySaken
Location: Wānaka, Aotearoa
Contact:

Re: Anyone wanna help with an idea (yep its another fork)

by PolySaken » Post

ywwv wrote:
Fri Jul 09, 2021 17:44
my fork of minetest will include smooth terrain with marching cubes. I have already implemented a lense flare shader. so tell me. why should I help you?

this is an opportunity. I can be your first core developer and right hand man. you just have to sell me . give it your best shot and you just might get a real star <<<--- ME
Read back, I've decided to abandon this in favor of making a new engine.
Guidebook Lib, for in-game docs | Poly Decor, some cool blocks | Vision Lib, an all-purpose library.

User avatar
runs
Member
Posts: 3225
Joined: Sat Oct 27, 2018 08:32

Re: Anyone wanna help with an idea (yep its another fork)

by runs » Post

No more forks of a single solitary person. A lot of dream, but little resolution. They lead to a dead end...

I recommend you to join the Crafter project, it is already somewhat advanced.

Realistic solutions, please.

User avatar
PolySaken
Member
Posts: 817
Joined: Thu Nov 09, 2017 05:18
GitHub: PolySaken-I-Am
In-game: PolySaken
Location: Wānaka, Aotearoa
Contact:

Re: Anyone wanna help with an idea (yep its another fork)

by PolySaken » Post

runs wrote:
Sat Jul 10, 2021 22:57
No more forks of a single solitary person. A lot of dream, but little resolution. They lead to a dead end...

I recommend you to join the Crafter project, it is already somewhat advanced.

Realistic solutions, please.
Minetest itself is pretty much a dead end, this is nothing new
Guidebook Lib, for in-game docs | Poly Decor, some cool blocks | Vision Lib, an all-purpose library.

User avatar
runs
Member
Posts: 3225
Joined: Sat Oct 27, 2018 08:32

Re: Anyone wanna help with an idea (yep its another fork)

by runs » Post

PolySaken wrote:
Sun Jul 11, 2021 01:07
runs wrote:
Sat Jul 10, 2021 22:57
No more forks of a single solitary person. A lot of dream, but little resolution. They lead to a dead end...

I recommend you to join the Crafter project, it is already somewhat advanced.

Realistic solutions, please.
Minetest itself is pretty much a dead end, this is nothing new
NO, look at the new shadow thing. Minetest may be dying, but never dead.

Post Reply

Who is online

Users browsing this forum: Semrush [Bot] and 16 guests