WebTorrent Desktop is a bit abandoned, but last time I tried it, it still worked despite some warts (I think it wouldn't work with newer Chromecasts).
My notes also have:
Not sure about how it handles video, but I've been meaning to take a look at https://getbananas.net/
I think Cloudflare Tunnels will require a different setup on k8s than on regular Linux hosts, but it's such a popular service among self-hosters that I have little doubt that you'll find a workable process.
(And likely you could cheat, and set up a small Linux VM to "bridge" k8s and Cloudflare Tunnels.)
Kubernetes is different, but it's learnable. In my opinion, K8S only comes into its own in a few scenarios:
Really elastic workloads. If you have stuff that scales horizontally (uncommon), you really can tell Amazon to give you more Kubernetes nodes when load grows, and destroy the nodes when load goes down. But this is not really applicable for self hosting, IMHO.
Really clustered software. Setting up say a PostgreSQL cluster is a ton of work. But people create K8S operators that you feed a declarative configuration (I want so many replicas, I want backups at this rate, etc.) and that work out everything for you... in a way that works in all K8S implementations! This is also very cool, but I suspect that there's not a lot of this in self-hosting.
Building SaaS platforms, etc. This is something that might be more reasonable to do in a self-hosting situation.
Like the person you're replying to, I also run Talos (as a VM in Proxmox). It's pretty cool. But in the end, I only run there 4 apps I've written myself, so using K8S as a kind of SaaS... and another application, https://github.com/avaraline/incarnator, which is basically distributed as container images and I was too lazy to deploy in a more conventional way.
I also do this for learning. Although I'm not a fan of how Docker Compose is becoming dominant in the self-hosting space, I have to admit it makes more sense than K8S for self-hosting. But K8S is cool and might get you a cool job, so by all means play with it- maybe you'll have fun!
Nextcloud is in EPEL 10. You'll get updates along with the rest of the OS.
I have been using EPEL 9 Nextcloud for a good while and it's been a smooth experience.
If you want specifically Docker, I would not choose an EL10 distro, really. I have been test driving AlmaLinux 10 and it's pretty nice, but I would look elsewhere.
Running LanguageTool locally is a bit of a pain, with some manual steps. Plus you have to fetch some data files. You can find around a few projects like this one to make it easier to run LanguageTool.
And yes, as the poster mentioned, LanguageTool keeps some code exclusive to their paid version. There's a bit of a tension because they ask people not to extend OSS LanguageTool with their paid features.
There's also this interesting clone, but it seems abandoned.
I like to live on the edge of time and therefore have the feeling that debian based distros (although being very stable) are too “old” for my liking.
Nowadays, with Flatpaks, so many software providing binaries, etc. this does not matter so much. If you want, you can even use something like Distrobox to have containers for tools using whatever bleeding edge distro you want, but still have a solid stable underpinning.
Debian also has more stuff than you would expect in backports. The main sticking point is yes, you'll be stuck in Debian 12's KDE until 13 comes out. But that might be sufficient for you?
(You could also use Debian Testing, which is basically a rolling release. But I'd consider stable first.)
Eh, my Nextcloud LXC container idles at less than 4.5% CPU usage ("max over the week" from Proxmox). I use PostgreSQL as the backend on a separate LXC container that has some peaks of 9% CPU usage, but is normally at 5% too.
I only have two users, though. But both containers have barely IO activity.
I was going to mention ZFS, but I suspect Raspberries are too weak for ZFS?
If you can use ZFS in both sides, send/receive is the bomb. (I use it for my backups.) However, I'm not sure how well encryption would work for your purpose. IIRC, last time I looked at it, if you wanted an encrypted replica, the source dataset should be encrypted, which did not make me happy.
I'd love to work on making NASes "great" for non-technical people. I feel it's key. Sending encrypted backups through peers is one of my personal obsessions. It should be possible for people to buy two NAS, then set up encrypted backups over the Internet with a simple procedure. I wish TrueNAS Scale enabled that- right now it's the closest thing that exists, I think.
The next TrueNAS Scale can do LXC containers using Incus. It's similar to a VM, but more lightweight. You can create a container for any Linux distro and install Borg on that. With previous versions, I googled and found some instructions to run Borg in a container with SSH, or you could use a VM.
Borg also supports dummy SSH targets, that TrueNAS can provide. Apparently, it's lower performance-
Why the choice of TrueNAS Scale? For just a Borg target, you could run any Linux distribution.
YunoHost is very nice to run on a VPS (or a box at home, or anything). It has good email hosting support, and I feel people without systems administration experience could get it running and host a couple of apps for a group without too much trouble.
TrueNAS Scale has awesome NAS capabilities. ZFS is the bomb. Plus, they are integrating Incus, which I'm a huge fan of. I think it hits a sweet spot for people with systems administration experience. Just install it and you get great NAS capabilities, the option of running a K8S instance, LXC/VM capabilities, and some "app catalog" (I test drove that briefly and it looked decent, but I think less hands-free than Yunohost.). My pet peeve (and I understand why they do this) is that you need separate drives for the OS and for data, so if you want redundancy you need 4 drives- which is likely fine for home use, but I'd like to run TrueNAS Scale on a Hetzner dedicated server, and that increases costs a lot.
If your primary desire is to run a few apps and you want to minimize your learning/effort, I'd check out YunoHost. If you want to do more, but also invest more time, TrueNAS Scale is awesome.
I'm still a huge fan of Ventoy, but lately I have been finding more and more issues with it.
So I decided to investigate using a Raspberry Pi Zero with a USB adapter to create a virtual drive:
It's very wonky and manual at the moment, but I have managed to boot all Linux ISOs successfully so far. Unfortunately, I think only ISOhybrid works OOB, so Windows ISO do not work. I have found some scripts to take Windows ISO and make them ISOhybrid, but haven't gotten around to doing that yet.
I think it should be doable to package this nicely.
You don't need to rebuild your server from scratch to use Ansible or any other configuration management tool. It helps, though, because then you can ensure you can rebuild from scratch in a fully automatic way.
You can start putting small things in control with Ansible; next time you want to make a change, do it through Ansible. If you stop making manual changes, you'll already get some benefit- like being able to put your Ansible manifests in version control.
(I still use Puppet for configuration files, installing packages, etc. It just does some stuff better than Ansible. Still, Puppet is harder to learn, and Ansible can be more than enough. Plus, there's stuff that Ansible can do that Puppet can't do.)
Dotfiles are a completely separate problem, tackle them separately. Don't use Ansible for that, use a dotfile-specific tool.