It sure does, but I don’t log my family and friends’ queries so I’ll probably MITM myself using a travel router.

Good point, I’ll be on the lookout for that.

I’m blocking primarily with my self-hosted, non-logging DNS server (Unbound).

I might just use my travel router to MITM myself while Tailscale is disabled on the iPhone to glean more information that way.

I agree. Most of my duplicates came from the raw disc files. I too dump some content to MKV (mainly TV episodes) but those files likely have much less duplication, though I do recall some of the duplicates coming from The Office in MKV.

(I do wonder if those The Office duplicates were something like the opening title, or scenes from the episode showing clips from previous episodes because it seems highly unlikely that the raw video streams were similar.)

I did for a few years when the network started, but it became increasingly difficult to do so from a residential IP with slow upload speeds (cable internet).

I use Storj, it’s been my favorite for years.

I’m in almost the exact same situation as OP, 8 TB of raw Blu-ray dumps except I’m on XFS. I ran duperemove and freed ~200 GB.

This is very related to the SNAT option for subnet routers on Tailscale. Though it’s enabled by default, I ran into issues with some services when I’d left it turned off by accident at one point.

In theory the “clean” way to do is to not use SNAT but then the network router needs to do some extra work to bridge the gap in the connection. Personally I was a dealing with a strict service on a device that wouldn’t accept regular non-SNAT traffic (the service was smart enough to say “no, I’m only running on 192.x.x.x and refuse to send traffic to Tailscale”).

deleted by creator

When did I say it’s open-source? And why would it need Plex? I’ve used it with Samba, S3, WebDAV but never Plex

I use Firecore Infuse, never understood the hype with Plex

I have a cheap $20 1byone scale from Amazon but it’s actually fantastic.

I think the simplest way is using Docker: https://codeberg.org/rimu/pyfedi/src/branch/main/compose.yaml

I’d say this comes down to your experience with Docker (or whatever you use to containerize).

Generally speaking updates are as easy as pulling the updated image, but if something goes wrong you should know how to run commands inside the container, access the database, etc. Containerization can be painful if you don’t work with it everyday, but at the same time it brings so many advantages and it’s not hard to learn.

That seems to be the consensus but what keeps me from picking it up quickly is it always looks very low-level (I’ve been in Ruby land for far too long 😆)

I’ve looked at Piefed and I kind of actually wish it were compiled. I come from the Ruby on Rails world and while I love Ruby (and Hanami too!) I’m just tired of interpreted languages (and Python always feels clunky coming from Ruby).

I wish I had the free time to learn Rust or Go, but I would be ecstatic to find a Crystal-based Lemmy server. Maybe when I’m 80 I’ll have the free time to work on that.

looks at community I hope so?

I’d almost go through the trouble of getting the content out of Wordpress. The nice thing about static site generators is you can completely switch out the framework, runtime, base Docker image and/or OS at any time.

Your router probably does have one, but your end devices should too. If your router is some piece of trash ISP-supplied one, it might not even have a firewall for IPv6 (if it even supports IPv6 at all).

I would add from an end-user privacy perspective, they might want HTTPS. If I hit a website not using HTTPS, I pretty much immediately back out. Bad actors like hostile governments and hackers can use seemingly meaningless data against you.

I can’t remember exactly what happened but I remember back when WebMD was fighting against rolling out TLS hackers were able to find medical weaknesses against people.