Practical notes about keeping control of data, choosing what is worth self-hosting, and making important files recoverable without turning everyday technology into a second job.
In 2011, Valve co-founder Gabe Newell said something about piracy that has stuck with me:
“Piracy is almost always a service problem and not a pricing problem.”1
His point was fairly simple. If an unauthorised service is easier to use than the one selling the product, the seller has a problem.
At the time, that could mean waiting months for something to reach your country, dealing with region locks or restrictive DRM, or hunting down a physical copy while an unauthorised version was already online.
Installing a media server is the easy part.
It may not feel easy the first time. There are unfamiliar settings, folders that refuse to appear and devices that behave differently for no obvious reason. Eventually, though, the application starts. A film plays on the television. An album appears on a phone. The small thrill of having built something useful is real.
For a while, that can feel like the finished project.
A self-hosted music library is good at playing music I already know. Spotify and Apple Music are good at answering a different question: what should I listen to next? The usual bargain is that they learn my habits, keep the recommendation system on their servers, and make discovery part of a subscription catalogue I do not control.
I wanted to separate those things. Navidrome could remain the player for music I keep locally, while an open recommendation service could learn from my listening and suggest unfamiliar artists. Most importantly, a recommendation would still be a suggestion rather than an instruction to collect another album.
I keep a media library, and I try to be honest with myself about what it is for: making the films and shows I actually want to watch available in my own home, without turning the hard drives into a warehouse of everything I once thought I might watch.
A library without a policy drifts. Every new thing looks like something worth keeping, the disks fill, and I end up curating storage instead of watching films. So the watching policy is small enough to remember:
A copy of a file is not automatically a backup, and a backup is not automatically a preservation system. Self-hosting becomes much easier to reason about when these three jobs are kept separate.
Synchronisation keeps copies convenient # Synchronisation keeps data available on more than one device. Syncthing is useful for this kind of direct device synchronisation, and it makes working files available where I need them.
Self-hosting is often presented as an answer to every problem. I do not think it is.
A service can be open source, privacy-friendly and technically interesting while still being the wrong thing for me to operate. Every service adds updates, storage, monitoring, backups, permissions and another failure mode. The aim is not to recreate every cloud product at home. The aim is to keep control where control is valuable without turning the homelab into a second job.
A useful project is not automatically worth adding just because it is new, popular or active on GitHub. I look for a clear problem, sensible deployment, useful documentation, a workable backup path and enough maintenance history to justify the time spent learning it.
This list deliberately excludes tools already used here, including Beszel, Dockhand and youtube-dl workflows. Karakeep is also excluded as a recommendation because it is already in use and has proved to be good in practice.
Self-hosting is becoming more interesting for a reason that has little to do with fashionable infrastructure. People are discovering that services they relied on can become worse while they are still using them.
Features disappear. Prices change. Limits arrive. Advertising takes over. A product that began as useful becomes harder to use, less private or dependent on an account that can be closed.
Cory Doctorow calls this process enshittification. In Enshittification: Why Everything Suddenly Got Worse and What to Do About It, he argues that platforms can first attract users with convenience, then lock in users and businesses before shifting value away from both groups.1 The Creative Salon article supplied alongside this book is another useful reference point for the discussion.2
A disc on a shelf is not perfect, but it is understandable. I can see it, lend it, play it without signing in and decide when to replace it. A digital item tied to an account, region, server or changing licence is different: I may have paid for access without receiving lasting control.
For a long time, that distinction felt mostly theoretical to me. Then I had to recover years of family photographs.
Self-hosting does not have to mean building a data centre. Start with one service that solves a real problem: a media library, photo backup, password manager or notes system.
Keep the design boring:
Use hardware you already own. Run one service at a time. Keep application data separate from the operating system. Back up configuration and irreplaceable files. Test restoring something before you need it. Use friendly local names and HTTPS. Document the few commands that matter. The goal is not to replace every cloud service. The goal is to make sure a company closing an account does not erase your only copy of your life.