The KISS version

• Keep the original files where possible.
• Use common formats such as PDF, FLAC, MP4 and JPEG.
• Keep one working copy and at least one backup.
• Store a second backup somewhere different.
• Check that backups can actually be opened.
• Review the setup once or twice a year.

Preservation is not about collecting for its own sake. It is about keeping reasonable access to things you bought, created or were given permission to keep.

Keep the design boring:

1. Use hardware you already own.
2. Run one service at a time.
3. Keep application data separate from the operating system.
4. Back up configuration and irreplaceable files.
5. Test restoring something before you need it.
6. Use friendly local names and HTTPS.
7. 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.

Sony removing or restricting access to disc-related features is a useful reminder of the wider trend. Hardware, software and media can be changed after purchase, delisted from stores, or made dependent on services that may not exist forever.

The answer is not to hoard everything or reject every modern service. It is to avoid a single point of failure:

• Buy DRM-free or physical copies when practical.
• Keep local backups of purchases and personal data.
• Prefer open formats and exportable libraries.
• Support creators directly.
• Learn which devices still work offline.
• Keep a simple inventory of what you own and where it is stored.

Convenience is valuable. It should not require surrendering all control.

The approach is deliberately simple:

• Keep a local copy where practical.
• Prefer open formats and exportable data.
• Use services that can be moved or restored.
• Back up before experimenting.
• Do not expose services unnecessarily.
• Choose boring, understandable technology.

This is not legal advice, and it is not a piracy download site. It is a place to discuss why access can disappear and how ordinary people can reduce unnecessary dependence on platforms.

My starting point

My homelab journey started with FreeNAS decades ago. At the time, the idea was simple: keep files on my own hardware instead of relying entirely on someone else’s computer.

In 2018, while transitioning into a new job, I came across Lawrence Systems and Novaspirit on YouTube. Their videos introduced me to a much wider homelab world: Raspberry Pis, containers, virtualisation, networking and the idea that a home server could be both useful and educational.

I have always been someone who keeps technology. I do not always do much with it immediately, but I keep it because it may become useful later. Over the years that has meant accumulating several switches, SFP connections and a 2.5-gigabit switch. The equipment is there, but the current setup is deliberately modest. A pile of hardware is not the same thing as a useful homelab.

The things I tried

I have experimented with several approaches, including:

• Xen Orchestra
• Proxmox
• Hyper-V
• Windows Server
• Different Linux distributions
• Containers and Docker-based services
• Network storage and dedicated NAS platforms

Trying different platforms was useful, but it also taught me that complexity is easy to add and hard to maintain. A more capable system is not automatically a better system.

These days I usually return to plain Debian or Fedora, rather than heavily customised derivatives. There is nothing wrong with derivatives, but a straightforward base makes it easier to understand what is installed, follow official documentation and recover from problems.

Where a new starter should begin

1. Start with one real problem

Choose one thing you genuinely want to improve. These are good first projects:

• Photos and phone backups → Immich
• Films and TV → Jellyfin
• Music → Navidrome
• Audiobooks and podcasts → Audiobookshelf
• Files between your own devices → Syncthing
• Passwords → Vaultwarden or the official Bitwarden self-hosted release
• Encrypted backups → Restic
• Notes and documents → Joplin or Simplenote

Do not start by installing every service you find interesting. Pick one, use it for a while and learn how it works. The Resources page contains more options and official project links.

2. Use hardware you already own

An old desktop, mini PC or Raspberry Pi is enough to begin. Start without buying a rack, enterprise switch or server-grade hardware. Upgrade only when you can explain the limitation you are solving.

A modest, quiet and reliable machine is often better than powerful hardware that uses too much electricity or becomes difficult to maintain.

3. Learn the basics before the advanced tools

The useful foundations are:

• Linux command line basics
• Files, permissions and ownership
• IP addresses and DNS
• SSH
• Storage and filesystems
• Docker Compose
• Backups and restores
• Reading logs

Virtualisation, VLANs, Kubernetes and advanced networking can wait. They are useful tools, but they are not prerequisites for running one good service.

4. Keep the network boring

You do not need to use every switch, SFP port or high-speed connection immediately. A simple wired network, a sensible DNS name and no unnecessary public exposure are a strong beginning.

Add faster links when they solve a real bottleneck, not simply because the hardware supports them.

5. Plan recovery before adding more services

Before installing the next application, know:

• Where its data lives
• How it is backed up
• How it is updated
• How it is restored
• What happens if the server dies

A homelab is successful when it is recoverable, not when its dashboard has the most icons.

My KISS rule

Keep it simple, keep it documented and keep a backup.

The best homelab is the one you understand well enough to repair on a tired Sunday morning. Start small, learn the boring parts and expand only when your existing setup is reliable.

Self-hosting and privacy

• Awesome Self-Hosted — a large catalogue of free and open-source self-hosted software.
• Privacy Guides: Self-hosting — practical privacy-focused guidance.
• AlternativeTo — find alternatives to commercial software.
• Docker documentation — containers and compose.

Media and personal data

• Jellyfin — personal media server.
• Immich — personal photo and video library.
• Audiobookshelf — audiobooks and podcasts.
• Navidrome — personal music server.
• Syncthing — direct device synchronisation.
• Restic — encrypted, verifiable backups.

Preservation and rights

• Library of Congress: Personal Digital Archiving
• Library of Congress: Recommended Formats
• Internet Archive
• EFF — digital rights and civil liberties.
• UK digital content rights

Piracy resources

• FMHY — a community-maintained index of free material and online resources.

FMHY changes frequently. Check links carefully and follow the law where you live. This site does not host or distribute copyrighted material, and linking to a resource is not an endorsement of every item in it.

Current focus

• A small self-hosted Docker environment
• Local media access with Jellyfin
• Personal photo storage with Immich
• Music and audiobook services
• Local DNS and HTTPS
• Encrypted backups and recovery checks
• Monitoring without automatic updates

The goal is not to replace every cloud service. It is to avoid making any one company the only place where an important service or copy of a file exists.

More detailed build notes will be added as each project is documented.