> Terms often lose their meanings as they become religions and more rigid choices and styles. The two in question here are certainly "object-oriented" and "functional programming"!
Absolutely, especially when they're vague. But I still think that the synchronous or temporal-logic based languages (which I love!) are very much distinct from FP (at least in the Haskell "tradition"), although they share with each other -- and not with the classical imperative approach -- the explicitness of state transitions: the synchronous/TL model with its clocks/steps, and the FP model with its (hideous) monads. They also "feel" very different (Esterel/Céu and even TLA+ feel more imperative than functional). It is not surprising that the killer-app for the FP approach is compilers (and other code analysis tools), while the killer-app for synchronous/TL approach is reactive systems, two application classes that are at the opposite, extreme ends of Pnueli's (I think) classification of programs (transformational, interactive and reactive).
I don't identify the general notion of "functional programming" with any particular language, including Haskell. I stay with the "idea of a function" as a reliable mapping. Similarly, I don't identify the general notion of "object-oriented" (as I characterized it long ago) with any particular language. Both of these terms have now been "colonized" and their meanings changed, etc.