I generally haven't written much about software architecture. People make heuristics into religion. But here is something I thought about: how to build in self-correction into systems. This has been something just vaguely sitting in my head lacking a clear conceptual definition until a whole slew of things popped up today that all had the exact same issue at its core.

Erik Bernhardsson

State drift occurs when two components synchronize state via deltas and the observer's state diverges from the producer's due to lost messages or failures. The core problem is that without a self-correction mechanism, a single lost delta can permanently corrupt state. The recommended solution is to treat delta updates opportunistically while periodically performing full state reconciliation — similar to how Git uses Merkle trees, rsync compares directory listings, or video codecs use keyframes. Strong delivery guarantees (e.g., Kafka's durable commit log) are a complementary strategy. Webhooks are criticized for lacking redelivery guarantees, requiring inbound HTTP servers, and making debugging difficult, though a durable queue proxy service is proposed as a mitigation.

State drift

Why can’t you just write code with 100% uptime? :trollface: