New release tag?

[sminnee]

I see that the latest release, v0.6.2, is a year old and that you've made lots of changes since then.

Do you think it would appropriate to tag a new release? I could make use of the master branch, but it's not clear to me how stable this is, and it's also less efficient for tools such as npm to use the branches directly.

Thanks, Sam

[bnolan]

It'd make the project look a bit more alive too - I love cannon and don't want people to overlook it because they think it's dead. 😸

[tjpalmer]

Cannon still seems dead. Which branch isn't behind master?

I like the engine. But I would also like rotation limits and a generic6dof constraint, among a few other things.

Several forks exist, but I'm not sure that any are trying to be a new lead.

[brundonsmith]

It's weird how dead this project feels given how many people seem to still use it. At the very least, the maintainers could do a new release tag/update NPM and maybe update some of the documentation (some of it is flat-out wrong from what I can tell) and the TypeScript types (also has some wrong stuff).

I've been keeping a list of inaccuracies in the docs and I'll probably make a pull request myself when I have time. Here's what I have so far:

• world.add() is now world.addBody()
• passing shape in the new Body() options silently fails; you have to call body.addShape().
• Heightfield data doesn’t mention anywhere that it’s a 2D array

Also,

• Vec3.UNIT_X, _Y, _Z don’t exist in typescript types

Of course, I'm using the version on NPM, so maybe these inaccuracies aren't accurate...

I'm no physics expert, but I'm in the process of building a major project on this library so I may start contributing where I can/when I have time. Really wish it had more backing; it seems like the best JS physics library out there, despite the lack of attention.

[dirkk0]

I am pretty sure that @schteppe didn't abandon Cannon. As far as I know he is ultra busy in a project he is launching right now and my guess is he in this phase where he has to focus 100% on it. In any way - thanks for the list!

[brundonsmith]

Apologies if I came off as demanding or impatient above.

Do you know him personally so as to state what his intents are? I only ask because the last commit was nearly two years ago... But then maybe that's just the nature of physics engines? Maybe they don't continue to change as much because their domain is inherently finite?

I guess my main surprise isn't that the creator's busy, but that other people haven't really taken over and kept things going

[dirkk0]

Five years ago I created this: http://dirkk0.github.io/fps0/ That's when I started to follow Stefan on Twitter. I don't know him personally, but I highly admire his skills. Check out his latest tweets on the very cool game he is working on: https://twitter.com/schteppe I wouldn't recommend taking over an open source 3D physics engine unless you know very well what you are doing, so I am not surprised there aren't any takers.

[brundonsmith]

By "take over" I didn't really mean taking over ownership or making a fork, just taking over some of the maintenance work and making commits.

But, yeah. Physics is hard and maybe there just really isn't anyone else out there with the interest, time, and experience to work on a project like this.

Also- I follow Stefan on twitter too and Jelly Mario is very cool. I just would really like to have a stable, maintained JS physics library for my project: https://github.com/brundonsmith/sovereignty.js

[dirkk0]

wow - cool! You have a star now.