English language is like Chinese: while simple structurally (no cases, constructive verb forms, etc) has a terribly complicated writing and pronunciation system, where there sort of are rules, but you never know when you hit an exception. Despite that, people take it up, because the important communication happens in English. (People who study areas like ML, or who work a lot with industrial production, likely pick up some Chinese, out of the same necessity.)
https://www.vulture.com/2020/01/stand-up-comedy-english-lang...
For instance, Gad mentions in the article the expression "Got it" that doesn't have great equivalents in French. But French also has sentences that don't have great equivalents in English, or if they do have one it's not nearly as playful.
The one thing English has that most languages don't is a massive body of work and a dominant grip on international culture, and I think that's what the comics in the article are interpreting as higher overall playfulness.
I see we have met. My ideal computer language would be Scheme in Ido. :) I guess to do real work one needs a ~980 page spec like Common Lisp which consumes half of their available neurons--in English, which consumes the other half. The R7RS spec is ~98 pages, for reference. A few extensions do not remove that much cognitive availablity (see Jaffer's Slib and the SRFI suite). As a matter of fact, speaking of minimalism, I was using MIT Scheme Version 7.4 Edwin today, an Emacs compatible editor written at R4RS (with a few extensions).
Seriously though, I really wish like the universal translator in Star Trek that I could understand all human language with deep context--but my puny U.S.-public-school-trained brain limits me to a few.
"So stop ragging on C++, you meanies!"?
It would be interesting to see a Scheme-like underpinning to the semantics of any language -- maybe not to be used for communication in its own right, but to achieve things like more intelligent translation, or NLP machine-learning applications that extract meaning from a text. I don't see much interest in something like this emerging, however, with the current trend in AI being "throw more statistics at the problem".
English is a mess because it is a mixture of languages from different ethnic groups: Anglo-Saxon, Norse, Celtic, and Norman. It's been used for cross-cultural communication ever since it was identifiable as English.
Hackernews armchair linguists seem to think an ideal language for world communication can somehow be engineered from some small set of primitives, like Scheme -- resulting in suggestions like that we all start speaking Toki Pona -- but the reality is that human communication is messy and the most practical languages tend to be messy ones. Some linguists have observed that trade pidgins develop English-like morphologies, even when English is not one of the contributing languages.
(And even Scheme got messy; see R6RS and R7RS...)
As for Japanese, I love it, it's beautiful, but as even any weeaboo knows, Japanese language is very bound up with Japanese culture. You can't simply ignore or elide the politeness bits; where you stand in society very strongly influences what you say. The fact that English is largely free of this baggage helps make it an effective language that people around the world pick up and use for trade, especially when different ethnic groups are involved.