20 comments

[ 3.0 ms ] story [ 53.5 ms ] thread
One thing that Minecraft has been missing, and that servers such as Bukkit tried to implement is an API for plugins. Looking through the docs, the LUA API[1] makes it seem as easily extendable as Garry's Mod.

[1] http://mc-server.xoft.cz/LuaAPI/

This is an interesting project. There DOES exist Minetest, which is c++ based, and uses lua entirely for running the worlds it serves. It's GPL for server + client.

See http://www.minetest.net/

Minetest isn't compatible with minecraft though. This project lets players connect using the Mojang proprietary client (and currently that's the only client).
It's a nice project but the server doesn't work for Minecraft clients whereas MCServer does and Minecraft is more popular than minetest.

But it's nice for you to share this so we know there are similar options avaliable

Nice. A bunch of friends of mine play Minecraft, but I never really got into it. So I decided to look into the protocol, and coded a chat-only minecraft server in dlang, which the Mojang client can connect to. (I never released it, even though I meant to as OSS)
Why yet another one? There is already https://github.com/fador/mineserver which could use more people working on it.
Is that one better? Why?
Dunno, but it has been under development for longer and seems to have the same goals so I was wondering why the creator made a new one or if he/she was unaware of the existence of mineserver. From the downvotes I've got I take it that such a discussion isn't welcome around here though.
I'd sure welcome the discussion.

Not that I know, but perhaps the downvotes were due to it coming across like you think this project shouldn't exist, without any justification beyond that another project exists that is similar.

Some might disagree with you because they believe, contrary to your apparent position, that diversity in options outweighs the fragmentation of effort that competing projects bring.

Others might be wary of how you seemed dismissive of this project without any explanation as to why the other is more deserved of attention.

I personally would love to hear about any thoughts you have comparing the two, particularly around technical underpinnings, development community, adoption, and feature completion. It's been a while since I dove into the belly of the MC server and am very interested in how current open source efforts are tracking.

MCServer is actually an older project than mineserver, it was originally created in 2011 by FakeTruth.
Both projects seem to be quite mature, but according to the git history, mineserver is older. The first commit of mineserver was on Oct. 12, 2010.

https://github.com/fador/mineserver/commit/399bca07f66b2882b...

The first MCServer commit was about a year later, on Oct. 3, 2011.

https://github.com/mc-server/MCServer/commit/cc2b15a2335db9a...

Though I'm not completely sure which one is older, MCServer was closed source for a while, so there are no commits before that time.
It isn't relevant which is older, or who was aware of what. There isn't some kind of rule which says that people can't start projects which are similar to other projects.
I tried out MCServer for a while. I put it back on the shelf after discovering that the redstone simulator was not as functional as the official server, and there are subtle behavioral differences in the things that do work. Between the two, some complex device designs that work on the official server break on MCServer.

But with respect to system resources, it's very frugal.

I'd imagine MCServer to be extremely valuable to server hosts, no? As far as I know, the official server uses a ton of resources
They should pick a working donation medium though. Gratipay seems broken beyond recognition and I'm desperately throwing my cc to my monitor for the last twenty minutes.