The Hidden Unity of APIs and Event Streams
summary
Claim: APIs and event streams are two expressions of the same state model and work best together. Uses a rideshare example and Uber webhooks (now deprecated) to show commands as REST for commands and queries and state changes as publish/subscribe events, recommends Kafka for event streams and notes tooling like AsyncAPI and CloudEvents. Explains tradeoffs—latency and strong consistency for queries, eventual consistency and fan‑out for events, and complexity/race conditions for large choreographies.
AI-written summary · about
chapters (5)
related talks
29mAPIs and Event Streaming Dualty
38mEvery Event, Everywhere, All at Once
43mUnraveling the Threads: Event-Sourcing, CQRS, and Event-Driven Architecture
54mSo You Want to Build An Event Driven System?
45mEvent-based system: Uklon practice
59mEffective Microservice Communication and Conversation Patterns
43mStructure and Interpretation of Stream Processing
52mNavigating complexity in event-driven architectures: A domain-driven approach