Valadis

Publishing Events from Legacy Why? To give legacy bragging rights? You know, some of that latent coolness. The reason events are published out of a legacy system should be to provide a means of integration and to incrementally modernize/transform. The alternative is to jump into the existing legacy code and start teasing apart the tangle for the purpose of modularizing the monolith. I know that's possible because I've done it with very large codebases. Yet picking off small facts of happenings in the legacy is very effective and arguably much simpler. I've used both approaches, spoken and written about, and taught them. Some claim that using events at all is due to the influence of Kafka. Is that true? It wasn't for me. Although Kafka may be involved in enabling this approach, it has nothing to do with the motivation. Domain Events and my use of them significantly predate Kafka. Surprisingly, publishing events from legacy can require little to no legacy source code modification. Instead, you use a stream of database changes and reify transaction log entries to events, which then are published via a messaging mechanism. It may or may not be Kafka. If you are interested in learning more about this, see the OSS product Debezium. This will be the topic of my next Design Accelerator episode.