If they are going to push edge computing, they need to have a first-party answer to the major application needs. Having a third-party ecosystem of offerings is extremely welcome, but I don't think Cloudflare should be banking on these. They need to develop an in-house solution - why should I be confident that edge solutions are better when they cannot offer me a competitive storage solution?
Durable Objects is awesome, but it's complementary. They need to offer a datastore.
I think this is exactly right, but the truth is that developers are slow to adopt new databases and databases are slow to support a broad array of use-cases. While we're also working on that piece, we think we can make relational databases easier to use and faster by caching them at our edge. To cache them, you have to connect... hence this announcement :)
Databases are a tough challenge, but I'm happy to hear you are trying to tackle it head on - I've been following the developments in storage from Cloudflare quite closely because the potential is massive.
My reasoning is that all their offerings are geared toward the edge, and while it's nice to support relational databases as they currently are, it seems to counter why you'd build on Cloudflare's edge to begin with.
This seems more to be suggesting that I should not pursue an edge-based architecture and stick with containers instead.
If I'm going to be using workers, I'm already giving up a ton of capabilities possible on a native OS. In return, I'd hope for something wild, like a fire and forget global relational database.
If my original ask was too tall an order, I think a global GraphQL service would fit nicer with their current stack and offer most of the same benefits as direct database access (i.e. schema as a service accessed over http).