In any reactive systems, to achieve reactivity, we can follow different approaches. One can follow the poll and diff mechanism, but this is not cool because systems have to constantly poll and compute differences. This makes the client snappy at times. Another approach is eventing. Eventing is good, but can become messy very easily. Reactive programming is another approach, not much used, but a powerful solution for bigger applications.
MeteorJS follows a reactive programming approach. It is declarative. In the sense that, when a user is logging in it, changes the login button to display the user's name if the authentication succeeds. Why did MeteorJS choose reactive programming instead of other approaches? Ultimately, applications are going to display some sort of data via interfaces. Updating interfaces is going to be mandatory for most of the applications.
MeteorJS wants to offer an easy but powerful, less time-consuming way to write interfaces that can update themselves...