Really strong take on the directive vs reactive orchestration distinction. Most discussions gloss over this, but in practice the differnce between commanding systems versus observing them fundamentaly shapes architecture decisions. The insurance quoting example nails it - realtime supervision of intent beats batch processing every time. One thing worth adding: the hidden cost of cloud orchestrators isnt just vendor lock-in, its the cognitive overhead of translating business logic into state machine DSLs that nobody outside the team can read.
Absolutely. That point about state-machine DSLs is spot on. Cloud orchestrators do not only create platform dependency, they often force business meaning into a mechanical notation that only the implementation team understands. The irony is that orchestration exists to improve clarity but expressed poorly it obscures it. The most successful teams I have seen treat orchestration as supervision logic first, notation second.
Really strong take on the directive vs reactive orchestration distinction. Most discussions gloss over this, but in practice the differnce between commanding systems versus observing them fundamentaly shapes architecture decisions. The insurance quoting example nails it - realtime supervision of intent beats batch processing every time. One thing worth adding: the hidden cost of cloud orchestrators isnt just vendor lock-in, its the cognitive overhead of translating business logic into state machine DSLs that nobody outside the team can read.
Absolutely. That point about state-machine DSLs is spot on. Cloud orchestrators do not only create platform dependency, they often force business meaning into a mechanical notation that only the implementation team understands. The irony is that orchestration exists to improve clarity but expressed poorly it obscures it. The most successful teams I have seen treat orchestration as supervision logic first, notation second.