Spot on. Self-hosting is the most effective form of quiet, material protest we have. Every time your family uses Syncthing instead of OneDrive, you’re starving the machine of the telemetry it needs to function.

Running that stack for your inner circle is essentially building a “digital mutual aid” node. You’re taking the burden of surveillance off their backs and putting it on your own hardware where you can actually defend it. That’s the work.

Quick question. Home assistant.

We are hooked on “Hey Google turn off the lights”

Is there a way to remove the Google from that but still use the voice aspect?

It is a skill much like maintaining a car yourself, or your own lawn/garden.

It’s pretty easy to get started, and there are certain ways of doing things that keep it pretty simple forever, at the cost of some flexibility.

But no matter how you do it, there will be a non-zero amount of work involved indefinitely. Just like you need your cars oil changed, your garden mulched and weeded, or your server patched and cleaned up once in awhile.

I feel this deeply. I used to volunteer at a library teaching “Cyber Seniors” digital literacy, and the biggest hurdle was always the fear of “breaking” something. The truth is, the big tech companies want you to think it’s too hard so you’ll keep paying them with your data.

You don’t need to be a sysadmin to start. It’s not about days of fixing errors; it’s about taking one small win at a time; like setting up a password manager first. If you can follow a recipe, you can build a node. We’re working on better, no-jargon guides to make sure the “thousand little errors” don’t stand in your way. You don’t have to be an expert to be part of the resistance.

Maybe I’m mistaken, maybe it’s all really easy even for noobs

I’ll be the first to admit, shit is complicated, especially networking, but it’s not insurmountable. Do you already have a server deployed? How familiar are you with Linux?

See what you think: https://linuxupskillchallenge.org/

You’re right. We’ve been traded convenience for our autonomy for way too long, and it’s created this massive power imbalance where a few tech bros basically own the digital roads we walk on. Voting with your wallet is a huge first step, but like you said, the real work starts when we actually take responsibility for our own data.

That’s exactly why I’m moving toward helping local businesses and groups build out their own nodes. It’s one thing to stop paying for a subscription, but it’s another thing entirely to stand up your own infrastructure that doesn’t report back to a corporate mother-ship. Every person who rejects the “default” and builds a private alternative is a small win for the rest of us, it’s about making the corporate extraction model fail by simply making it unnecessary.

You nailed it, the “I’m not techy” thing is often just a shield people use because they are simply exhausted by this economy, and time is the one resource Big Tech steals that we can’t ever get back. I’ve spent a lot of time teaching seniors at a library program, and I’ve seen firsthand how that “convenience” is a trap designed to keep people from even looking under the hood to see what’s actually happening to their data.

You are right about the remote admin headache too, that’s exactly why the movement needs to shift from just “hobbyist favors” to actual, reliable infrastructure that doesn’t break every time an adult in the house clicks a link. If we don’t make these sovereign nodes as easy as a light switch, people will always fall back into the arms of a corporation just to get through their Monday. We have to be the ones who put in the work to make the “resistance” feel like less of a chore and more like a utility.

Wireguard is stupid easy.

I run a docker container using docker compose. Put in my bits of info on the compose file…

Launch the container and scan a QR code with my phone app.

Done.

Openvpn was out to door when I saw how easy wireguard is

As OP said, it’s fine if you still use some corporate services, I think this one should be in the bottom of the list

Wireguard can easily replace simple Tailscale usages, like if you only have 2 nodes to connect and have a static IP address. One thing Tailscale is good at is creating a mesh network, where if you have more than 2 nodes, you only need to configure each one to connect to the central server which will allow the nodes to connect to each other (internally it uses a wireguard connection). With plain wireguard if you have 4 nodes, you need to configure on each one the configuration to the 3 other. Another thing Tailscale is good at is Nat hole punching, if your ISP provider doesn’t give you a static IP address or if you don’t want to open a port in the firewall of your home router, Tailscale will allow you to access services hosted on your local network (another commercial solution for this is cloudflare tunnel), wireguard doesn’t provide this

When you’re using tailscale, they get a lot of metadata about your hosts, but the data transfered between your nodes is encrypted (by wireguard)

By replacing the tailscale servers which are ran by the tailscale company with headscale which is the self hostable open source solution, tailscale won’t be able to get the metadata of your nodes. Tailscale clients are oss and compatible with headscale, but headscale is not on par for features (like tailscale serve or funnel).

For headscale to really make sense it usually needs to run on a pubicly accessible host like a vps, and not in your home network. Another selfhosted alternative to tailscale, which have totally different approach, is pangolin

Hope this helps

I need to get a backup of some encyclopedias and/or get wikipedia somewhere hosted. That would be fun and informative.

I selfhost the full Wikipedia in Kiwix, plus a decent amount of IT, Political, History reading material in my Calibre library. I’m not much for fiction, novels, or movies.

On 1: Autoseeding ISOs over bittorrent is pretty easy, helps strengthening the commumity distribution, and makes sure you already have the latest stable locally when you need it.

While a bit more resource intensive, running a full distribution package mirror is very nice if you can justify it. No more waiting for registry sync and package downloads on installs and upgrades. apt-mirror if you are curious. While not as resilient there is also apt-cacher-ng to at least get a seamless shared package cache on the local network.