I love the concept of Supabase, I only worry about performance in this case. Almost all API calls in my applications execute some SQL to gather data from the database. Having the edge function execute several select queries on the database on the other side of the world is a few times slower than having the API server near the database. Basically, the distance between API and database is much more important than the distance between user and API. I guess the only solution is to have read-copies of the database on the edge as well..?