In addition to the ones listed:
Matrix client (Element, Shildichat, Fluffy chat, etc.)
XMPP client (Conversations, Monocles Chat, Blabber, etc.)
SimpleX Chat
Briar
None of them are tied to your phone number if that's what you mean by non voip.
Tried ElementX but ran into issues while getting it working with a selfhosted matrix server. After setting up dendrite (matrix server) and sliding-sync server (asynchronous message history syncing; which it requires) I still wasn't able to get past the initial sync screen. Are you hosting your own matrix server, did you use it with a public server, or have you not tried using it yet?
Just looking at the TypeMatrix for the first time. Looks like a pretty interesting layout. Manufacturing a mechanical TM(TypeMatrix) clone is likely going to be pretty difficult, but becomes more possible the more you diverge from the design.
At a high level though, the main issues I see at first glance is that TM is using laptop switches, custom legends, and 1/2 size function row keys.
The only analogue to laptop switches in the mechanical keyboard world is Kalih X switches, but they're hard to come by and a bit more difficult to design for, but they just might work with the TM keycaps, but don't quote me on that.
Choc switches are possible, but you're not going to get legends (the letters and symbols on the keycaps) unless you have them custom made, which is next to impossible if using traditional methods. Same story with full size MX switches. You might be able to fill out the alpha-numerics, but the secondary legends in blue aren't going to be there, not to mention the vertical shift, back space, enter, and control.
As for the 1/2 size function row, you'll have to compromise and go with full 1u keys because there aren't any switches out there that will fit in a 0.5u size, not to mention lack of keycaps. And speaking of odd-sized keycaps, you might run into issues with Choc keycaps for the spacebar in particular, as I don't think anyone manufactures a 5u spacebar in choc form factor, and the 2u shift, enter, and control keys will need choc stabilizers, which are pretty hard to find IIRC.
So, with that all said, these are the compromises you're probably going to have to make to get a completed keyboard:
There's a few routes you can go down to get his done.
Pay someone else to do it.
Learn how to do it yourself.
If you take route 1, it's going to be much more costly than route 2 most likely, but far less time and effort invested to get a working product. I'm sure there are plenty of people who have designed their own PCB/Keyboard that would be willing to design yours for the right price.
If you take route 2 then you're going to be putting a lot of time into learning different software packages, how a keyboard works at the circuit level, and how to solder at the very least; but you'll have done all the work yourself, which is pretty gratifying IMO.
A good place to start learning is "ergogen". It's like a configuration language and accompanying software for designing keyboards programmatically. It's not especially complicated and has plenty of resources on YouTube and in written form. There are online viewers that make it easy to visualize the layout you're configuring, and other projects on Github that can take ergogen output files and convert those to pcb design files that are ready to send off to manufacturing, 3d printed case files, and and interactive BOM (bill of materials) that can help you when it comes time to solder the switches, controller, diodes, etc.
I see now, that makes sense why you are building the image since it was set up that way. I don't know why projects set up the compose file to build the image when they already have a publicly available image to use; it just creates unnecessary friction for people who just want to test out the software. Anyway, using that image should work for you, but feel free to ask if you run into any issues.
Why are you building the image yourself? Not that there's a problem with that necessarily, but it seems a bit wasteful of your resources unless you have a specific reason to do so. There's a docker image (quay.io/invidious/invidious:latest) built by the developers that gets updated pretty frequently. I've been using it for years now and it's been working perfectly fine for me the whole time.
Even if you need something just once, just install it and then uninstall it, takes like 10 seconds.
apt install foo && apt remove foo
That's essentially what nix-shell -p does. Not a special feature of nix, just nix's way of doing the above.
Actually using it though is pretty convenient; it disappears on its own when I exit the shell. I used it just the other day with nix-shell -p ventoy to install ventoy onto an ssd, I may not need that program again for years. Just used it with audible-cli to download my library and strip the DRM with ffmpeg. Probably won't be needing that for a while either.
The other thing to keep in mind is that since Nix is meant to be declarative, everything goes in a config file, which screams semi-permenant. Having to do that with ventoy and audible-cli would just be pretty inconvenient. That's why it exists; due to how Nix is, you need a subcommand for temporary one-off operations.
If you're ok with just file storage sftpgo has been solid for me for years now. Does sftp ftp and WebDAV (like nextcloud). Webui isn't as pretty but it's fast. Mobile apps will be various sync apps with sftp or WebDAV support. On Android folder sync pro is pretty good for keeping documents and pictures backed up
They did. Its called airmessage. Has been around for almost 3 years now
Ah you're probably right about the mobile clients. I'm not a mobile watcher really. I can say though that the jellyfin desktop app and jellyfin mpv shim both have skip-intro integration, though I've only tested it with jellyfin MPV shim.
Skip-intro is an unofficial plugin ATM but can vouch that it works decently well. Can't compare it to the Plex implementation since it has been quite a while since I've had Plex deployed.
Just this one https://wallhaven.cc/w/2y2wg6
Have it hard-coded in my config so I'm less likely to faff around with my wallpaper instead of doing something productive. Less cognitive load IMO
No problem, glad to help point you in a useful direction. If you're not especially concerned with labels, you might be able to get around needing to have a blank set of keycaps laser etched or something like that.
I also got bored yesterday and had your idea of a TypeMatrix 2030 kicking around my head, so I decided to mock-up a layout similar to what I described in the previous post, and created a Github project you can use as a jumping off point if you want to proceed that direction.
https://github.com/pyunramura/mechamatrix2030
Keep in mind, the layout probably needs some more thought and the pcb generation part of it has not even been started, but it should be possible to get a completed board with some learning, some funds, and some work.
One of the major issues I didn't think of before is the fact that the keyboard has quite a lot of keys, so many in fact that the matrix is too large for a standard pro-micro type microcontroller. The keyboard has 7 rows and 15 columns which needs 22 pins, which is unfortunately 4 too many for a pro-micro. There are other microcontrollers with a similar footprint that might be suitable instead though. Elite-C, Blackpill, and some of the RP2040 boards come to mind.