I kept seeing the same pattern.
A project starts open. Then something shifts. HashiCorp changed Vault's license. Docker put Desktop behind a paywall. Plex keeps pushing deeper into cloud dependency. Every time, people who built on these tools found out after the fact — after they were already dependent.
So I started asking: how do you evaluate a technology not just by what it does today, but by where it's heading?
I built The Autonomous Stack (TAS) — an open framework with 8 questions for any technology.
Three structural questions
1. Pause — Can you stop it without permanent damage?
If you shut it down for a week, does anything break irreversibly?
2. Exit — Can you leave with all your data?
Not a partial export. Everything, in a format you can use elsewhere.
3. Recoverability — Can you roll back if something breaks?
Snapshots, versioning, known restore path.
Five diagnostic questions
4. Personalisation — Does it build a behavioral model of you?
The more it knows about you, the harder it is to leave.
5. Urgency — Does it manufacture time pressure?
"Your trial ends in 3 days." A tool shouldn't need to panic you into staying.
6. Hidden cost — What do you pay besides money?
Attention, data, dependency, lock-in.
7. Transparency fragility — Would you still use it if you saw everything?
If full visibility into how it works would make you uncomfortable, that's a signal.
8. Trajectory — Is the project moving toward openness or closure?
License changes, acquisition rumors, VC funding rounds — direction matters more than current state.
The catalog
I rated 28 technologies on a scale from A0 (cloud-bound) to A3 (fully autonomous):
- Google Drive vs Nextcloud vs Syncthing
- Plex vs Jellyfin
- Notion vs Obsidian vs Paperless-ngx
- GitHub vs Gitea
- and more
The goal isn't to say "never use cloud." It's to make the choice conscious.
Try it
The framework is open source: The Autonomous Stack
There's also an 8-question Infrastructure Audit you can run on your current stack right now.
What tools in your stack would fail the Exit test?
Top comments (0)