In my opinion, Scala and Clojure aren't new "kings", they're hybrid languages: bridges from the old paradigm to the new one, the same way C++ was around 2 decades ago. That's why the "tail wags the dog" and sitting on top of the Java runtime (or .NET for the Microsoft world) is what matters first.
That doesn't mean they won't have longevity and staying power (see: C++), but they aren't the "kings", the Java runtime is (the same way C is for C++). If you're looking for a new "king" you need to be looking for a language/platform that is as different from the JRE as the JRE is from C. I don't think we know what that would be yet (or why we'd invest the effort in that transition).
I've had a difficult time getting into either Clojure or Scala - bought a book on each, and had fun playing, but neither language felt great to me. I am stuck on (J)Ruby, Common Lisp, and Java when I need it.
BTW, Kenny is about as close to a "pure Common Lisp" purist as you can get. I worked with him for about half a year - cool guy,and very much focused on Lisp.
Here is my prediction: it will take academia 10 years to stop teaching Java 1.2 and start teaching, say, Scala. It will then take industry about 20 years to realize that Java is not "the best" anymore.
So for 30 years, everything other than Java 1.6 and C# will be "niche".
(Sure, there are plenty of non-Java shops around. But anywhere where software is not the primary concern, it's the default. It got that way because of marketing and an easy word to Google, not due to any technical merits. So the new languages that are competing on technical merits are probably not going to win in industry, only among the handful of programmers with a clue. Just sayin'.)
I think you've missed how the computer world works, sure people still program in fortran but that doesn't mean it is still at the same working extent.
cobol -> c++ -> java actually happened pretty quickly and I'd expect when a useful alternative proves itself in a few niche senses on large scale a new language will be adopted and everyone will have something else to complain about.
I think you're wrong; we will have to wait and see.
In the mean time, you shouldn't let the industry's stupidity affect you. Make your own language decisions. (I work with someone who went to a hard-core Java school. He is a great Perl programmer. You can hire "Java programmers" to use real languages too :)
"Pretty quickly" in this case was at least 15 years. COBOL was the main language of the early 80s and Java of the late 90s - and it's not as if C++ has disappeared either.
Not necessarily; Steele is a programming polyglot and can hack in any language thrown at him. Kenny is a Lisp weenie; you know, one of those guys who can only code in Lisp.
I would take Kenny lisp aesthetic over Steele's, but Steele's technical competency over anyone else's.
That doesn't mean they won't have longevity and staying power (see: C++), but they aren't the "kings", the Java runtime is (the same way C is for C++). If you're looking for a new "king" you need to be looking for a language/platform that is as different from the JRE as the JRE is from C. I don't think we know what that would be yet (or why we'd invest the effort in that transition).