DISTRIBUTED SYSTEMS PROGRAMMING HAS STALLED

Summary from the TLDR Web Dev newsletter

Programming for distributed systems hasn't seen fundamental improvements in the last decade. Current approaches rely on band-aids over existing paradigms like external-distribution, static-location, and arbitrary-location, each with limitations in performance, control, or semantic clarity. There's a need for a new programming model that is native to distributed systems that has things like semantic locality and can surface potential issues from network delays and faults.