What he doesn't understand is that this is not an accidental feature of the existing programming languages; it's an inherent property of programming languages with enough power, i.e., Turing-complete languages.
Let's leave aside his claim that too many library functions is a bad thing. He also seems to think that such and such a language is difficult because the syntax is difficult, but we know that you can have a very simple language in terms of syntax -- a Turing machine, a bare-bones Lisp, or a C-like language with just assignment, increment, if and goto, for instance -- that will still give rise to very complex programs.
Of course, just because 2 programming languages are Turing-equivalent doesn't mean they're equally good or equally easy to learn or expressive. Most people would agree that for most purposes C is a better tool than Assembly, and that $FAVOURITE_LANGUAGE is better than C. Still, if you can really code Javascript then you should be able to code in Objective C. The only way that he can possibly use Javascript but not Objective C is if the limits his usage of Javascript so much that it is not really being used as full programming language.
What he wants (but doesn't know) are non-Turing-complete DSLs. Just as he can do stuff in HTML/CSS, he might be able to use another DSL to do whatever it is that he wants.
PS: also see this comment and the linked short essay by Dijkstra:
http://www.leavesofcode.com/2011/06/we-need-programming-lang...
'On the foolishness of "natural language programming"'
http://www.cs.utexas.edu/users/EWD/transcriptions/EWD06xx/EW...