What's everyone working on this week (31/2026)?

New week, new Rust! What are you folks up to?

Still working on some new database software, I got it to the stage of a browser interface where functions can be defined and altered. Quite a cool feature (IMO) is when a function is altered other functions do not need to be recompiled. There is a check that the return type and parameter types are unchanged. I also measured the time to respond to a simple query, which is around 200 micro-seconds (not counting IO overhead etc, just the database work ).

Built a small tool to query Hutrr42-c Keyboard extended attributes: keyboard-layout-info (needs better name).

If your keyboard reports extended attributes (and isn't based on my QMK pull request) I'd love to hear from you.

I want real world data on vendor+product ID's that make use of this AND how vendors map Keyboard layouts to the bcp 47 language tag.

I couldn't find any good crate providing a many-to-many bidirectional map, so I'm working on my own, multi_bimap. To not impose on the programmer how the underlying antiparallel multimaps should be implemented (there are many ways to do this, e.g. two hashmaps of hashsets, or two btreemaps of vecs, or mixed, etc.), it relies on generic container traits provided by another crate of mine, maplike, which I'm also improving.

Your readme still mentions undoredo, presumably another project of yours.

Thanks, fixed! As you can see, I'm still working on this.

Still miles better than the joke that's my readme, dw.

I'm working on Rimbun, a collaborative document creation platform, where versions from different authors can be easily seen and compared, instead of being buried in the version history.

After weeks of setting up a protocol crate for a middleware that is vaguely inspired by DCE, I just recently started implementing the actual client and server runtime libraries. Also decided to add a hybrid where an endpoint can act as both client and server at the same time.

Got unidirectional messages working, got request/replies working, and am currently working on streaming data to the server.

This is a project I've been passively working on since somewhere around 1995, as a result of growing a fascination for microkernels and messagepassing. I wrote large parts of a C implementation many years ago, but it would hang sometimes (threading bug) and I was kind of unhappy about the structure of the whole project -- I realized how I should have done it along the way, so I didn't feel like tracking the problem down. Never got around to rewriting it, until now.

Tracking & slewing input streams for on-device (GPU) DSP for a music visualizer. Discussion opened on repo.

Going to be publishing the first DFT code probably today. Audio ticks, dispatches DFT update. Video ticks, tracks & slews on the audio output ring.