@thdxr@EffectTS_ But yeah, I think a lot of solutions are way too complicated to deploy atm, but I’m sure we’ll see simpler solutions, as it kind of feels like this is the future of development — a paradigm shift where the “unhappy” paths gets more consideration and patterns to handle them.
@thdxr@EffectTS_ is working on something in this space with multiple storage solutions (Postgres etc.). Combining X-State and Effect Cluster is something I want to dive into. I also think a combination of event sourcing and actors sound really promising for durable execution.
@fernandorojo When GraphQL incorporates “plan resolvers” a-la Grafast (or Hasura’s internal engine) in a way that is easy to work it will be hard do beat.
@rauchg@zeeg And with projects such as @wundergraphcom you can even federate multiple schemas and expose queries as JSON RPC endpoints. Stuff like this drives the web forward. Just like with RSC there will always be naysayers though…
@rauchg@zeeg I don’t get all this hate for #GraphQL these days. Someone always tries to post some “gotcha”, but there’s usually an agreed-upon solution to that particular issue. I find standing up GraphQL servers and clients to be very ergonomic, there’s amazing open-source tooling out there.