Leaving Rust upstream
I'm not really sure how it happened but for some reason I got interested in contributing to the rustc compiler. Rustc was the first and only compiler I've ever contributed to.
The Rust compiler is under constant change. I've seen the Makefile build system being replaced by a Rust written one. I've seen that build system being almost rewritten from scratch. I've seen MIR being introduced. And that's just the past stuff. There are plans to get typecheck happen on MIR, for a new borrow checker, to parallelize rustc using rayon, to switch off LLVM entirely in some scenarios. And all of this while every six weeks, a new release of Rust is being churned out. I don't think that many other languages can afford to do something like that.
I've learned a ton of stuff about how compilers work, and many interesting ideas about development of open source projects in general.
If you wonder about which contributions I've done, I've prepared a document listing all the relevant changes. Most of them of the smaller kind. Note that I don't consider myself to be relevant. Even if you restrict yourself to my main contribution phase, there are still 19 people who have contributed more than me during that time. I'm mainly writing a post because I've done similarly for a project I've contributed to prior to Rust upstream, Minetest. This is just my way of parting with things.
It seems to me that I have contributed enough. From now on, I'll try to avoid to send contributions, that is PRs, reviews, comments, or issues to any official Rust projects. I'll do an exception for bug reports for things I encounter personally. Things that are blockers, etc.
Looking forward, I want to focus on other, more interesting projects. Most importantly: projects that are still in need for help, where I can learn exciting new stuff. I'll still write most of my stuff in Rust. There is a whole ecosystem to explore.
It's exciting to have been part of one of the most cutting edge compilers of recent days!