About the future

[damagefilter]

Hello people and thank you so very much for sticking around and enduring our technical shenanigans.

We are aware that it is far away from optimal.

However, for all of you who are still around and interested, I wish to kick off a good discussion about the future of the project.
As you'll all know by now, the resources of the former core team are very very limited and we are on a very basic code maintenance mode.
We would like to change that of course but then we have to put thought into the project.

In my own opinion, we cannot continue like we did. There is no way to keep up with the changes made to Minecraft
and dissect the code and re-apply our changes each update, keep up with development and community and then some.

However, there are alternatives on the horizon and denying them seems to be stupid.

Okay, I will get to the point now and here are some questions I want to discuss.

What would be a good alternative base of implementation?
Technically that means, what can or should we use instead of self-decompiled NMS?
What are the implications?

In case we the choice falls on a platform that supports plugins by itself and we would implement CanaryLib as such a plugin, what will we do with our own plugin system? Throw away? Keep it? Change it to fit into the implementation?
You know, keeping it would introduce a form of redundancy but throwing it away will destroy all existing Canary-Plugins and Canary is reduced to a mere (but pretty damn good) administration software.

Who is interested in developing (actually coding) such a new implementation?
Because the former core team can not put nearly enough resources into this and if nobody is interested, we might as well just let it be.
There will have to be a lot of changes to CanaryLib to adjust to "modern modding". And then such an implementation takes time and good knowledge of the framework we're implementing on.

And that's that.

Personally, I think implementing a Forge Core Mod sounds like a good idea but neither of us has good knowledge of Forge.
So that looks like a problem there.

[jamierocks]

What would be a good alternative base of implementation?
I recommend using the Mixin technology to implement CanaryLib.

Who is interested in developing (actually coding) such a new implementation?
I have been developing Neptune for a couple months now, which has both a Vanilla and Forge implementation.

[damagefilter]

Yes, I was hoping you'd pitch in.

Because what you do looks very promising.

For the others: https://github.com/NeptunePowered

But then, I suppose you'd need some backup too.
A bunch of reliable people with 2-3 days a week to spare for development?

[jamierocks]

Yeah, development is slow because I am the only developer.

A couple of people would probably be sufficient to start picking up some pace.

[damagefilter]

Well. I suppose rocks won't start to roll all on their own.
Ping me on IRC the next time you're on.
We should see what can be done to clear up the mess.

[sammi4]

Nice to see you again.
Unfortunately I ran into an UUID-conversion problem with Canary, I tried to have help, Jason (DarkDiplomat) and Brian McCarthy could not fix it and the website was then down, so I had to switch to Spigot.

I hope that Canary will be up again, but it is a huge task. Not just the core coding, the plugins are also important to be able to run a server. I often mentioned that the website did not have complete information about the project, hence too much time was spent on answering basic questions from the users. I offered my help in writing a comprehensive manual and I started making a list of commands, but I did not get any response. I remember that other users offered a fine guide on how to set up a complete Canary server, which was really needed.

In essence, restarting the Canary project would need careful preparation - and ofcourse dedicated staff maintaining software and website.