Design patterns and architecture · How changes travel
Data flow patterns-style app
Illustrative system design for a how changes travel app like Data flow patterns: 3 topics to read, from requirements to deep dives.
Illustrative design, not Data flow patterns’s actual implementation. Arrowbox is not affiliated with Data flow patterns’s owner. Disclaimer
Start with Props down, events up Sign in to start readingUpdated 2 Oct 2026
Core
Data flows from its owner down through props; children ask for changes by calling callbacks. Shared data is lifted to the nearest common owner.
To reach components far apart, publish the change once and let consumers subscribe. The difference is who re-renders: every context consumer, only store subscribers whose selected slice changed, or whoever happens to listen on a bus.
Workers, iframes and other tabs share no memory with your page, so state moves between them only as copied or transferred messages. Design those messages like a small protocol.