Post

Fellow #DDD folks out there, Can someone explain to me what is a _practical_ difference between a vertical slice and a bounded context?

For me Vertical Slice Architecture is rebranded CQRS where you slice your applications by behaviour top down. Having that I’d consider a slice per task. Which usually mean a single command/event handling (but can be more). That’s also aligned with task-based UI concept. 1/

question: wouldn't it be - technically - more like the following? Considering that a slice could live in multiple places in parallel (shared database - to some extent at least, separate UI codebase, etc.) 1/2

0 reposts0 likes

Things are shared when they need to be shared, and this is usually the DB, its data model in code, and parts of the UI like templates/layouts etc. The goal is not “shared nothing” but “share what needs to be”

My understanding is that the main point is to make each slice autonomous. And cohesive. And since db/be/fe are separate units (sep. deploys, sep. servers, whatever) they can't be really treated as a unit universally. Logically - yes. Deployment-wise - no, or at least it depends on many factors 2/2