Hacker News Reader

I didn't read it that way. He's seems to be claiming that getting good at a new language involves a lot of non-mind-expanding overhead---which, in the author's estimation, isn't worth the opportunity cost past one or two additional (sufficiently different) languages. I think this passage sums it up:

Aside from the immediate reasons, there's some merit to the mind expansion argument. I think being proficient in at least two languages is indeed important for boosting your ability as a developer. This resembles human language: learning a second one changes the way you think and your perception of the world. The third or fourth, not as much. But it can't be any two languages. If you know Portuguese and Spanish, your mind didn't have to expand much. Likewise, learning VB.net and C# doesn't count.

I've found that most languages I consider worth learning have fairly small overhead. The syntaxes of ML, Lisp, Haskell, Smalltalk, Prolog, Lua, Ruby, Qi and Python are small, simple and easy to learn[0]. Most of those have small standard libraries as well. Learning any one of those, assuming you don't already know a very similar language is likely to change how you think about programming. Even if you don't use that language much, it will alter how you code in other languages.

Is this worth it, from a strictly professional perspective? I have no clue. I've never met a strictly professional programmer who was any good.

[0] This should not be taken to mean that these are the only languages I consider worth learning - they're just the first ones that came to mind.

"I've found that most languages I consider worth learning have fairly small overhead...Lisp..."

Scheme yes, maybe not Common Lisp :P

Note that I said the syntax is small for all of (list of languages) and most of those have a small standard library. CL has a simple syntax and a large standard library. By extension, so does Qi.

There's more to the syntax than just the parenthesis. That you write

   (let ((x 4)) (print x))

instead of

   (let x 4 (print x))

is part of it. So is the order of arguments.

Sure, those are all parts of the syntax of a language, but I still maintain that CL's syntax is simpler and more regular than that of Ruby and vastly simpler than that of Java.

As an aside, the first form of the example you gave is what most Lisps use. CL, Scheme, Emacs and Goo all use this. The second, as far as I know is only used by Arc, which has a more complicated syntax than most Lisps.

My pragmatic advice is learn enough of a language from a different programming paradigm so you can apply that new way of thinking to your native language.

I think "enough" is a crucial point here. The author almost seemed to be arguing against switching to a new language every year, which I agree is ridiculous. The idea is to learn some new concepts, not to write significant programs in a new language every year. I only adopt new languages for regular work if they're the pragmatic choice.

That's exactly it. Saying that learning a new language is just a matter of learning a new syntax is like saying that an English-speaking, non-German speaker could translate a German text to English by taking a German-English dictionary and going through and translating the words one-by-one. It's not quite as simple as that.

Actually, I'd compare that to saying that two languages are the exact same except for some of the words -- that JavaScript is exactly like Java except System.out.print is renamed to document.write. Saying that learning a new language is just the syntax is more like saying that translation between English and German can be done with just a German-English dictionary and a full knowledge of the grammar.

However, just as lambdas cannot be directly translated into languages without lambdas, some languages have unique concepts. For example, the German language has a part of speech called the particle that conveys a general feeling into sentences. While the statement "Er wisst das." could be easily translated with a dictionary and knowledge of conjugation rules into "He knows that," adding a particle to make "Er wisst ja das" could be variously translated as "Of course he knows that," "He obviously knows that," "He most definitely knows that," etc.

Ja, genau.

http://www.toytowngermany.com/lofi/index.php/t971.html

I find such comparisons misleading. The differences between natural languages are not at all the same thing as the differences between programming languages. All natural languages have the same expressive power, or at least very close. The existence of the phenomenon of translation proves it. You can take a book in Classical Latin (according to my friend's girlfriend who is a linguist, a very terse language) and its translation into German (typically given as an example of a very verbose language) and the difference in length will be maybe 30% (and frankly, I don't even believe this difference can be accounted for by the greater expressive power of Latin; my guess is it's just that German words are longer on average.)