Wow, and I’m here and thinking who needs this, what problems are solved here by adding that complexity and don’t see any real benefit from the producer side perspective and that’s where the costs are.

Using a multi-tenant SQLite setup helps ensure user data is separated into their own databases and won't be inadvertently queried by other users. Users can then choose to share the data they want to either via peer-to-peer or by aggregating to a postgres database.

But the main benefit to me would be that user data will be more private and protected and configurable.

This gives someone much more fine-grained control over their data and how their data is stored. I think it ultimately makes the attack vector space smaller for any client data.

Me as a service provider I would just want to keep the data seperated at my backend. I guess Ill need it anyways for some usecases. And after that why should I bother adding complexity to ship that data to the client side?

Right, but then you’re not building a local-first app. of course the tools that are useful for building a local-first app won’t be that relevant to you if that’s not what you’re doing!