I can only really speak to your first point. When imported my existing library, I did it using Sonarr/Radarr as applicable. They have a manual import method, here’s a description of Sonarr’s.

Unfortunately that’ll probably work best if they’re formatted in a way Sonarr can readily recognize, something like /Season ##/S##E## - .ext. It may take a little work to get there, I found a program called mmv which helps out a lot. It allows you to move files that match a pattern, capture parts of pattern, and use that captured part to name the output file. That allowed me for format entire seasons at a time, but that method does rely on most files having similar names to begin with.

Since Jellyfin v10.6.0, it’s had a feature called SyncPlay allowing multiple users to watch the same thing at the same time (coordinates pauses, fast forwarding/rewinding, and all that between clients). I’ve used it and it worked like a charm, although I did find that not all clients support it.

I doubt Reddit builds a decent search engine, that doesn’t actually help them at all.
If users can search, they find a previous post pertaining to what they want to see/know and they move on.
If there’s no search, users can’t find old posts or comments so they make new posts about a previously posted topic and more comments are made as other users react. That’s more content, even if low quality from a user perspective, that shows engagement which can be sold to advertisers.

That’s before considering the engineering effort it takes to make a good search engine, constantly fine tune that algorithm, and try to outpace those that are trying to game the search algorithm.

I like Czkawka for detecting and handling duplicate files, similar files, empty directories, and more.

Sunday afternoon, after careful evaluation of a significant security concern, we made the intentional decision to sever our ties to the internet.

I feel like most big announcements like this end up being Ransomware. Cutting off from the wider internet feels like a weird move to defend/mitigate that? Unless it’s to reduce exfiltration?

I assume you mean SSL/TLS certificates for internet accessable applications? I use a reverse proxy called Caddy in a Docker container, which handles requests from the internet and directs them to the proper docker container based on the subdomain. It also handles my certificates automatically, requesting a new Let’sEncrypt cert just before the old one expires using a community made plugin.

You may be able to use the CLI tool mmv, which can be installed through the apt package manager. It’s great at renaming files that are starting a similar naming convention and ending with a similar naming convention, you could use mmv to move your files. It also suppose sum links and hardlinks. It’s what I used to rename folders of tv shows when I need to do that.

It’s all about what you value, and supporting the things you love (or rely on, in a more utility sense). I’d value the speed, the lack of data collection that may be used against a user, the speed, the location options, and that same provider being in business for time to come. When I’d need a good VPN, nothing else will do. That seems worth the couple of USD per month to me, whether that VPN is for obscuring traffic I don’t want others to know about (whether it’d be because of those facilitating the connection or the other end of the connection).

I’m not sure, that’s a question for the Dev.

Some of the reason may be the hastle of rebranding, having two Openboards would be confusing so the fork would need to change names and icons and such. Some of it is also be this is for personal use, and we happen to find it, so they may not be interested in the expectation of maintaining it beyond their own useage. Some of it may be this is good enough, Openboard’s release cycle is pretty slow so the fork doesn’t need to be updated and released often, so an APK on Github downloaded twice a year is good enohgh for them.

I use a fork of OpenBoard with Swipe Typing libraries added to it.

Containers like Docker/Podman? I only use that for selfhosted services on my servers. I can’t speak to Docker/Podman applications used on desktop.

If you want to include more universal package methods such ad Snap, Flatpak, and AppImage, I use Flatpak installations for things I want to sandbox that don’t have a focus on Linux Development. These are applications like Discord and Spotify. I’ll also install applications through Flatpak that aren’t available through my package manager, after checking that the Flatpak is maintained by the developer. An example of that which I use is Czkawka which finds and offers solutions for duplicate and similar files.
The way I used Flatpak, I haven’t run into any show stopping issues although I do know using Flatpak for CLI applications can be difficult and annoying. I believe Flatpak CLI apps need to be run by using the full Flatpak package name instead of just the executable name.

If you’re open to things similar to Plex, I’d recommend Jellyfin! Plex has been making some decisions lately that aren’t necessarily selfhoster friendly. A selfhosted instance of Plex still authenticates using Plex’s central servers (if you’re internet is out or Plex is down and you want to stream your own movies or shows, that won’t work due to failed authentication). That’s compared to your Jellyfin instance handling authentication locally. If you can contact your server, you can watch your media. Plex has also announced a credit skipping feature, uploading credit timing to their central servers that can be restored on complete rebuild. While they say it’s anonymous, they need some way to associate you and the proper credit timings, to send that back to you.

Jellyfin is earlier days in development, and you should check to see what clients are available to see if that would work with your hardware. But Jellyfin is definitely catching up, I’ve been very happy with their server and applications.