🚀 Colyseus is fully independent again!
“Colyseus Arena” has been shut down on 31 March 2023.
“Colyseus Arena” was a managed hosting service developed and provided by the company Lucid Sight, Inc. My relationship with the company was limited to a consulting contract that has also been terminated on 31 March 2023.
In case you missed it, here are the contents of the open letter published on the colyseus.io website on 5th April 2023:
A brief summary of the project
- 2016 — The first version was released as a hobby/personal project
- 2018 — Quit my job to focus on Colyseus and freelancing
- 2021 — A LA-based startup acquired Colyseus
- 2023 — The LA startup failed —the ownership has been transferred back
- 2023-onwards — The sustainability issue, Colyseus Cloud, and lots of work ahead!
In February 2021, VentureBeat announced that a startup based in Los Angeles acquired Colyseus.
Behind the acquisition
The proposal was made in December 2020, when I was feeling overwhelmed and frustrated. I was spending countless hours on the project while having very little in return.
Letting go of ownership of the project was one of the hardest decisions I’ve ever made. I had to go through tremendous mental gymnastics to detach myself from it while also considering new project(s) to start before the contract with that company would expire.
The papers were signed in January 2021.
After the acquisition
I can’t share much due to contract agreements, but I’ve joined the startup’s team and they were lovely with me. It felt surreal that they would refer to “Colyseus” as a real business the whole time. It felt much bigger than it was. I was really excited in the beginning.
The team managed to launch “Colyseus Arena” (the managed hosting service) a few months after the acquisition. It worked well and was growing steadily.
By the end of 2021, the upcoming version 0.15 of the framework was ready to be released — which could reduce server costs considerably. Its release was being postponed until the “Arena” service could support it to avoid user friction.
“Arena” never supported the upcoming version 0.15. I felt that the work I was doing on the open-source framework wasn’t being appreciated enough, and I started to slowly lose my enthusiasm. Eventually, I decided to focus most of my energy on a side project.
“Colyseus Arena” shutting down
Fast forward to March 31, 2023 — Colyseus Arena had to be abruptly discontinued due to a lack of funds. My contract with the company was also terminated.
(I feel really sorry for everyone who had their games and projects affected by this. It wasn’t meant to be this way.)
With the “Arena” service discontinued, Colyseus would serve no purpose to the startup anymore, so they were nice enough to return the ownership of the project to me.
Having the ownership back
The feelings of attachment to the project were quick to come back after they told me they’d give me back ownership. I immediately began working on my own hosting service.
I have a new challenge now. Only receiving donations as before won’t pay my bills.
A business — “Colyseus Cloud”
For Colyseus as an open-source framework to continue existing and evolving, it must have a viable business.
The main purpose of Colyseus Cloud is to offer dead simple and effective Colyseus hosting.
No serverless, no edge-computing, no auto-scaling, no Kubernetes. (At the time of this writing.)
I have ambitious plans for Colyseus and even more ambitious plans for a side project I’ve been working on recently.
Colyseus Version 1.0
There are many long-lasting bugs, and feature requests that must be taken care of. Version 1.0 must include most of them.
I don’t have a public roadmap yet, but I’m expecting to have version 1.0 released somewhere between the year of 2024 and 2025.
After the 1.0 release, I’m expecting to explore tighter integration between Colyseus Cloud and user-facing features. It’s too soon to share any plans here, but I don’t wanna simply copy what the competition is doing.
An ambitious side-project [?]
During this period between 2021 and 2023, I was researching and prototyping a QUIC implementation in the Zig programming language. It is so primitive at the moment it doesn’t even worth sharing.
Having a set of portable lower-level networking libraries would serve a much broad audience by being reusable mostly anywhere.
Anyway, this is way more ambitious than Colyseus — the focus should be 100% on making Colyseus sustainable right now.
Thank you for caring for Colyseus ❤️
The project has only come this far because of the constant feedback the community has given. I will continue doing my best to listen to you and improve wherever possible.
Yours truly, Endel.