The client-server relationship is distributed but the intra-browser stuff is just async IPC. Writing robust clients against remote APIs requires considerations

The client-server relationship is distributed but the intra-browser stuff is just async IPC. Writing robust clients against remote APIs requires considerations around retries, failover, idempotency and read/write consistency. It seems to me that distributed patterns in pure client code are thin on the ground and limited to things like CQRS and Sagas.