Folia Maintenance Question

[R00tB33rMan]

Folia in its current state is effectively almost unmaintained. There is a dozen PRs that have sat there for months, over 50+ valid issues that have been left ignored, high priority/critical issues and typos manifesting across all parts of the codebase, with a maintainer than never communicates with his users (or someone with the experience to provide a fully thought-through response in said support channels). While admitting Spottedleaf is incredibly talented, these are issues that must be addressed that go beyond just allowing the platform to support newer versions of the game. Unless someone (e.g., Dueris, given his thorough knowledge on the Folia codebase, as provided by "a fork") steps in and continually makes an effort and has the capability to make modifications and improve the Folia repository, the various bugs, issues, and overall instability of Folia will only get worse.

Although this is being presented in light of us clarifying or signifying that we have a "better" fork, it is important to recognize that we want to ensure users not wanting to opt into a fork have the opportunity to at least stably run Folia. This fork has attempted to downstream various already-implemented features with extremely prolonged responses and delays that inevitably make it worthless to even consider migrating fixes into Folia.

To clearly iterate these "widespread issues," I've collected a list of evident concerns that substantially and/or minimally impact users' experience:

• End credits broken
• Ender pearls broken
• Waypoints broken
• Scoreboard broken
• World border not fully thread-safe
• Numerous crashes with entity AI
• Debug subscribers broken and cause crashes
• JsonRCP broken and ticked on all regions which isnt safe
• Players not autosaved correctly
• Only maps and chunks are autosaved
• ChunkPacketBlockControllerAntiXray not properly regionized
• swing command not regionized at all (and is still enabled by Folia)
• Broken command feedbacks on most commands
• Max player count is offset by -1 in Folia
• A few technical Minecraft mechanics are broken
• Minecraft GUI playerlist is broken
• Vanish API is broken (incurs unfixable crashes unless Folia is directly modified)
• Inconsistent event calls
• Flushing netty connection on every single packet send
• Broken world load/unload API
• Typo in RegionizedWorldData that always sets ticking block entities to true
• A lot of dead old code
• A lot of dirty code and typos in comments and inconsistent formatting
• Broken pitch/yaw in tp command
• Crash reports not generated on region crash
• Numerous commands still broken/disabled
• Lots of TODOs still existing with nothing done about them
• Lack of thread checks on important API
• Respawn anchor charges never decremented on use via respawning
• Still many usages of Bukkit.isPrimaryThread()
• Usages of MinecraftServer#isStopped() causing to issues like chat still being processed during shutdown
• Entities can leave region bounds if moved with high enoguh velocity, causing crashes with things like ender pearl cannons
• Riding statistics never updated
• Lots of entity crashes randomly
• Ender pearls removed on dimension switch when ender_pearls_vanish_on_death is enabled
• Subspace bubble advancement calls when respawning in the overworld rather than portaling out of the overworld via a nether portal
  Note: This list is incomplete, truncated, and many more issues follow

While we all understand that availability and time constraints may weigh into one's ability to maintain a project, it's important that PaperMC fulfills its promise of maintaining high quality code. If one or more contributors don't have time to maintain said project, that is their choice (because they aren't necessarily being paid to work on said project); however, that does in fact mean that PaperMC should instill and/or train someone to commit to the project so problems aren't left unchecked for years to come.

[R00tB33rMan]

@mja00, can you provide a description as to why this post was thumbs-downed? I feel like providing clarity would shed like into an alternative perspective.

[Dueris]

I do want to clarify this is not meant to attack SpottedLeaf, his work, or PaperMC, this is simply to make these issues known as we feel that this poses sorta an issue long-term for Folia.

We are simply just trying to help, no harm is intended by this

[mja00]

@mja00, can you provide a description as to why this post was thumbs-downed? I feel like providing clarity would shed like into an alternative perspective.

Inquiring about maintenance and appeared abandonment is fine. This is just not the way to do it.

[qnxt]

they're too busy "fixing" harmless things, like being able to place right up against the world border from the other side

[R00tB33rMan]

@mja00, can you provide a description as to why this post was thumbs-downed? I feel like providing clarity would shed like into an alternative perspective.

Inquiring about maintenance and appeared abandonment is fine. This is just not the way to do it.

It's important for users to have some technical knowledge of the Minecraft server; however, we've attempted to reach out to the maintainer (as instructed by kennytv; won't ping out of respect), but received no response. Providing a PSA via Discord doesn't really fit the bill. It'll be glanced over, ignored, and there simply isn't a place for something like that. The purpose of this issue report is simply to raise relevant points and enact change (to some degree), not to harm or disregard the substantial and already remarkably well-made aspects of Folia.

TL;DR: This is a "if nothing is said, nothing is done" sort of deal with respect to the author.

[Dueris]

@mja00, can you provide a description as to why this post w...
Inquiring about maintenance and appeared abandonment is fine. This is just not the way to do it.

Yeah that's fair TBH. In all honesty, we have tried reaching out via discord a few times to no avail, hence why we thought doing this may be a potentially constructive idea. How would u suppose we go about this? Like, we aren't rly trying to spark controversy or anything, we just sorta want to help the project since this is a REALLY talented and powerful project, but the lack of maintenance is an issue especially in a codebase like Folia.

[Toffikk]

I want to throw my 2 cents in and just say that directing people in the help channels to just fork folia if they want to use it isn't the solution. This has created the issue of every network and server using this software having their own fork fixing issues that could be fixed upstream. I do not think this is the proper way for such a project to go.

I strongly believe that Folia should either start being maintained and the directing of people to fork it if they want to use it should have an end put to it OR the project should be archived/clearly stated as a framework. not a high-performance fork promising multithreading when in fact it promises exploits and breaks every players' expectation of how a minecraft gameplay should look like

[mja00]

FWIW I do agree with points made, and that it's quite a pain for people to maintain multiple forks of various levels of stability. Just issues, and especially ones where you quite literally list all the pitfalls of a project a single person has put a lot of work into, don't usually make for a decent discussion space. Granted this is my take, and I have no stake in PaperMC (I just contribute from time to time).

I definitely get the want for a discussion around this. As for where it should be, that's up to those in charge. I'd personally stick to Discord but as this appears to be a last resort sorta deal I get it.

[mja00]

Maybe someone can be convinced to create a folia-contrib channel and role in the Discord to get people who want to see Folia prosper and/or want to help maintain Folia and that'd help out?

[Dueris]

Yeah this is def not to put any bad light on Spotteds work, his work is genuinely incredible and as someone who has reviewed the full patches in Folia line by line, it's genuinely impressive and beautiful work. Not at all trying to put bad light on his work, it's genuinely incredible. Def appreciate ur insight tho.

I like ur idea of a folia-contrib channel, that sounds awesome. Hopefully someone here with perms likes that suggestion and sees that lmfao

[R00tB33rMan]

FWIW I do agree with points made, and that it's quite a pain for people to maintain multiple forks of various levels of stability. Just issues, and especially ones where you quite literally list all the pitfalls of a project a single person has put a lot of work into, don't usually make for a decent discussion space. Granted this is my take, and I have no stake in PaperMC (I just contribute from time to time).

I definitely get the want for a discussion around this. As for where it should be, that's up to those in charge. I'd personally stick to Discord but as this appears to be a last resort sorta deal I get it.

When constructing this, I faced a dilemma: either baselessly claim there are issues or list a good handful of them to raise a valid level of awareness (i.e., the scientific method) and provide evidence of "what went wrong." Thank you for the hardened reviews, though. I do deeply appreciate and respect them! We're all just here wanting to make Folia better!

[JavierFlores09]

I want to throw my 2 cents in and just say that directing people in the help channels to just fork folia if they want to use it isn't the solution. This has created the issue of every network and server using this software having their own fork fixing issues that could be fixed upstream. I do not think this is the proper way for such a project to go.

I strongly believe that Folia should either start being maintained and the directing of people to fork it if they want to use it should have an end put to it OR the project should be archived/clearly stated as a framework. not a high-performance fork promising multithreading when in fact it promises exploits and breaks every players' expectation of how a minecraft gameplay should look like

The issue is that we can wish it would be maintained all we want, but the hard truth is that the number of people with both the knowledge and, more importantly, the time to do that right now is effectively zero.

It doesn't matter if 100 people make PRs or report 1,000 issues when there's a single person responsible for verifying those issues, validating fixes, and going back and forth with contributors. That person simply doesn't have, or doesn't want to (which is entirely valid, since it's a hobby project), spend their time doing all of that work.

That's why we direct people to fork Folia instead. However, out of all the people forking Folia, there's like one active fork, which I won't mention the name of but y'all probably know already, that actually follows the open-source approach and does all of the above within their own fork. The rest just work on their own because we all know that being an open-source maintainer is a tireless job with no gain.

So, while I understand the frustration, there's simply not much to be done here, the status quo won't change. If anything, it's good that Folia serves as a starting point for people trying to do the same, and it's probably best for people to lower their expectations of the project becoming anything more.

[PedroMPagani]

Isn't it funny how the people who complain most features aren't implemented have $$$ (networks) and capacity to implement them and help but instead chose to complain about them not being there, grow up ;)