Reliable defaults beat clever recovery
A predictable first path prevents more incidents than a complicated fallback can repair.
Small services often accumulate recovery logic before their normal path is fully understood. Every retry, alternate endpoint, and compatibility branch adds another state that must be observed and tested.
Make the common path boring
A useful default is explicit, easy to inspect, and safe to repeat. It should fail in one recognizable way and leave enough evidence to explain what happened.
Recovery still matters, but it works best when it restores a single known state instead of choosing among several nearly equivalent ones.
Prefer fewer authorities
When configuration, runtime state, and documentation disagree, repair becomes guesswork. Keeping one authoritative source reduces both operational work and surprise.