Reactive Programming

Reactive programming is a declarative programming paradigm centered on data streams and the propagation of change: when a data source changes, every computation that depends on it is automatically recomputed, and every output that depends on those computations is automatically updated. The paradigm inverts the traditional imperative model. Instead of the program polling for changes or explicitly updating state, the program declares relationships — 'Y is always the sum of X and Z' — and the runtime maintains those relationships as the underlying data changes.

The conceptual roots lie in functional programming and dataflow architectures, but reactive programming has become the dominant paradigm for modern user interfaces, sensor networks, and real-time data dashboards. Frameworks like React, RxJava, and Vue.js implement reactive programming through dependency-tracking systems that rebuild or re-render only the parts of the application affected by a change. The performance challenge is not recomputation but change detection: determining which dependencies have changed and in what order recomputation must occur to maintain consistency.

Reactive programming is not without criticism. The implicit propagation of change can produce cascading updates that are difficult to debug, and the loss of explicit control over execution order can lead to performance pathologies — update storms, glitches, and memory leaks from forgotten subscriptions — that are invisible in the source code. The programmer trades control for convenience, and the runtime's update scheduler becomes a hidden source of complexity.