In this talk, the author discusses the traditional database model as global, shared, and mutable state, challenging developers to consider a new approach. Rather than viewing databases as static collections of facts that can be queried at a specific time, the focus is on processing streams of immutable facts in real-time. The talk introduces Apache Samza, a distributed stream processing framework that aims to revolutionize traditional database architecture by leveraging a distributed, durable commit log implemented by Apache Kafka. The benefits of this approach include simpler code, better scalability, robustness, lower latency, and increased flexibility for data manipulation. This talk encourages developers to rethink their application architecture and embrace a new way of handling data.
https://martin.kleppmann.com/2015/11/05/database-inside-out-at-oredev.html