I find the question really odd. It is like asking why people create new recipes or new music or whatever else - because they can. These are some of the best ways to introspect and think about patterns one might have been struck by and put forward a way to deal with them in their own ways.
People should always try new ways to deal with problems they see around them. Or generally be creative. I can not fathom what makes the author even ask this question.
So you think "why create music" is an odd question too? I dont, but that's another discussion.
Back to the original question, making programing language and making music are nothing alike in there purposes. Music are specifically made to be consumed while proraming languages or any kind of software/protocol/standard are meant to be productive.
You can argue that people make languages for their own amusement, those exist. However if a language rises to some sort of adoption there was definitely some amount of effort put into promoting it and convincing other people to use it. At that point a language ceases to exist for amusement purpose and become an investment that cost individuals and the whole industry a lot. It better has a good pitch.
No, that is your thought process that programming languages are meant to be productive. The person who creates, does so in many instances because they feel they want to create.
Which is why there is an order of magnitude more super-little-known programming languages out there than the ones that dominate the professional world of programming.
A few languages get traction and grow. Rest are just creations from someone.
Read again, I'm not disputing the fact that languages are created for fun. However when you start promoting a language and lure other people in contributing into your community, that (should) stops. Unless you specifically say smt like "this is a programming language for goofing around nothing serious", which is no PL author's pitch ever.
Making a new language is more like making a new instrument or a new cooking tool though. You've got quite a mountain to climb in getting people to use it, to the level where there's enough usage to to be worth the effort. Sure, you can make a new language as an exercise in creativity, but there's more than a few "serious" languages that all seem to have a mix of the same set of ideas (types, immutability, lambdas, etc).
I totally agree it is a climb, but only if that was the intention of the creator. We constantly forget that engineers can also be creators for their own sake.
There are many many many little know programming languages. Only a tiny % get traction. This is the natural pattern to everything around us. Newtown developed Infinitesimal Calculus in the same period that most students study the subject. We cherish these as accomplishments because they become famous, but most do not become famous or even useful. It does not mean people should stop experimenting or creating.
> I find the question really odd. It is like asking why people create new recipes
This analogy missed the point. It's about recipes written in a new language barely spoken by anyone, like elvish or klingon. Writing Romulan ale recipes in klingon is totally fine if you are targeting people on a star-trek convention, and want to start a romulan-klingon fight. But writing the only one documentation of a critical system in klingon at a public company sounds crazy, even if you can reduce the LOC (lines written) by 10% due to the nice and unique grammar of klingon (I mad this one up, I have no clue about klingon). Guess what, no sane person would want to work with that documentation, unique klingon grammar or not.
You just have to take a look at Esperanto, the "an easy and flexible language - created by Zamenhof - that would serve as a universal second language to foster world peace and international understanding, and to build a 'community of speakers', as he believed that one could not have a language without such a community." After 130 years still no one speaks it, and English is the common language in the western world. Which - considering its shortcomings - is like if everyone of us would mainly code in COBOL, just because everyone can code in COBOL.
tl,dr: normal human beings don't want to learn 2 new languages every year, just because [put a random argument here]. Why should we?
Very good points here, but you have to keep in mind this is not a small group of people pushing ideas down the throat of others. There are perhaps 10+ million programmers in the wild if you include those who are not professional or full-time.
Someone creates a language because they might simply want to learn. Their friend picks it up, has fun, spreads the word. It either dies after a few months or becomes an edge case language in the stack of a company. Someone writes a blog and then it gets more traction.
No one actually intends that everyone should learn 2 new languages. But we live on a large planet, people forget that frequently. The social consciousness that is easily accessible from within US might not be in India. So people here might create a language for some issues that already have a language in the US. What is wrong in that?
People should always try new ways to deal with problems they see around them. Or generally be creative. I can not fathom what makes the author even ask this question.