Although he named it the Python paradox and wrote mostly about Python being chosen by Java programmers, he doesn't think of Python too highly. He mentions Perl, Ruby, Icon, Joy, J, Lisp and Smalltalk. He also speaks about esoteric languages, but based on his other essays, he probably means any language better than the average.

Since the average language depends on the type of the program, it depends:

    - if Java is the average: Clojure, Scala, Groovy, Python, Ruby, Go, Erlang
    - if Javascript: CoffeeScript, TypeScript, ClojureScript
    - if C/C++: Rust, D

That's not exhaustive and highly subjective of course.

And since programming languages are such an important topic of his, any language that is considered the best overall language by clever hackers would fit the description.

I also would throw Scala in.

It has it drawbacks because it has to run on the JVM. But that is one of it's strengths as well.

If you once have programmed Scala for a while, you do not want to bend your mind around java anymore ;)

> If you once have programmed Scala for a while, you do not want to bend your mind around java anymore ;)

I'm having that exact experience with Clojure.

IMHO Scala would be the present day equivalent. Go would be a close second.

Go? Hah, you almost had me there, until you said "Go".

I think clojure/ocaml/haskell.

Clojure. It's funnel is filled with people who solve the seriously complex real world problems that live on the JVM. Because learning Clojure requires the sort of investment that professional programmers make because they are professional and because Clojure attracts people who are attracted to lisps, it tends to filter out people who are just interested in lines of code for a paycheck.

Big pipeline through a highly selective filter.

I think it still would be Python. Can't think of any language that has leap-frogged the Python that Paul Graham described back 11 years ago.

From how I interpret the linked article, it's not about language features, but the willingness of people to embrace/challenge better ideas that might not be mainstream.

    And people don't learn Python because it will get them a job; they learn it because they genuinely like to program and aren't satisfied with the languages they already know.

I might suggest Haskell (re-emergent of pure functional, but statically typed) and Go (improved type system but without resorting to language lawyering and re-emergence of coroutines (besides Lua)).

That pretty much covers why I elected to pick up Go and Clojure lately. I actually love programming, and my old tools were feeling rusty.

> And people don't learn Python because it will get them a > job; they learn it because they genuinely like to program > and aren't satisfied with the languages they already know.

Today people very much learn Python to get a job. It is listed as the top language for machine learning / data science (or at least top 5, depending on who you ask) and that is a very hot field job wise currently.