Well, the title's a bit of a tautology, but it does seem that a bit of the glow has worn off of Java (to those that were captivated by it in the first place anyway). Functional programming seems to be trending upwards too.
I agree with the general assertion that hard core OO developers (Java and .Net folks) will need to start learning some new mental models of programming to keep adding relevance. I think we will start solving some harder problems that are intractable with traditional OO mechanisms.
Yes, perhaps "are intractable" was too strong of an assertion.
I think as we migrate our data towards more flexible structures and look more towards asynchronous parallel processing, we may find that splitting out data and behavior serves us better than tightly binding them as OO methodology does.
I think ultimately we will treat our code and data as the same thing, just from different perspectives and levels of abstraction. I'm not sure most OO programming environments today support this notion easily.
I do think we will be solving harder problems, but what exactly they are falls into my unknown unknowns category.
I agree with the general assertion that hard core OO developers (Java and .Net folks) will need to start learning some new mental models of programming to keep adding relevance. I think we will start solving some harder problems that are intractable with traditional OO mechanisms.