All the machinery of functional programming languages that's creeping into non-FP languages is great. I actually think that FP one-liners are usually _more_ transparent than the corresponding non-FP loop.
However if I'm not mistaken an important point of FPLs was to eliminate side-effects and that the ultimate goal was program verification (not validation). I.e. you could prove a program correct in the mathematical sense. That seems still a long way off.
The trend I'm seeing isn't towards verification by proof but leveraging the type system to ensure that the compiled program is correct and secure. Both Opa and Ur/Web take this line.
However if I'm not mistaken an important point of FPLs was to eliminate side-effects and that the ultimate goal was program verification (not validation). I.e. you could prove a program correct in the mathematical sense. That seems still a long way off.