Flux forces us to think about the information architecture of our application in new and interesting ways. Rarely does adopting a new approach like this happen in a vacuum. The ideas tend to spread to other parts of the technology stack. With Flux, it's the architectural principles of data-flow direction and feature-driven information that stand out as having a positive impact. If these things can have a positive impact on the frontend code, why couldn't they influence the design of the system as a whole?
The unidirectional flow of data through a Flux architecture is probably the key aspect that enables it to scale. By itself, unidirectional data-flow makes the code we write easy to reason about. In places, this approach can be a little more verbose, but this is a conscious trade-off that we make in order to facilitate predictability. For example, with the two-way data binding capabilities found in some frameworks, we can get away with writing...