A feature hidden in a config file might as well not exist. Could a universal schema fix this?

I've been thinking about what still holds Linux desktop adoption back, and I don't think it's features anymore.

Linux already has an absurd amount of functionality. PipeWire alone supports things like advanced routing, crossovers, channel remapping, latency tuning, spatial audio, virtual devices, and much more. BlueZ has a ton of Bluetooth settings. NetworkManager, iSCSI, power management, storage, and networking tools all have capabilities most users never discover.

The problem is that most of these features are hidden behind config files, terminal commands, or documentation rabbit holes.

A feature that requires editing /etc/something.conf might as well not exist for 99% of users.

My idea isn't to replace existing configuration systems. Every project could continue using whatever it already uses internally. The only thing I'd like to see is a common way to expose configurable parameters and capabilities.

Imagine every project optionally publishing a simple schema describing:

• parameter name
• type
• valid values
• description
• complexity level

Then KDE, GNOME, or any other frontend could automatically generate configuration interfaces.

The concept already exists elsewhere. Home Assistant discovers entities and builds interfaces around them. OpenAPI can generate clients and documentation from schemas. Kodi has Basic, Standard, Advanced, and Expert modes.

The last one is especially interesting. One thing Linux often gets wrong is that applications either expose everything or almost nothing.

Imagine:

Basic

• Output device
• Volume
• Sample rate

Advanced

• Channel remapping
• Crossover
• Latency tuning

Expert

• Graph configuration
• Realtime scheduling
• Internal PipeWire parameters

The backend wouldn't change. The config files wouldn't change. The user would just gain a consistent way to discover what the system is capable of.

For software that doesn't implement the schema, adapters could be written. A BlueZ adapter could edit Bluetooth configs. An iSCSI adapter could expose targets, sessions and discovery through the same interface. Legacy software could still participate without upstream modifications.

I genuinely think Linux has reached a point where discoverability is a bigger problem than missing functionality.

A lot of developers spend months implementing powerful features that end up being used by maybe 0.1% of users simply because nobody knows they exist.

Am I missing something obvious, or has this already been attempted before?