I'm with you. I'll even go a step further, since I haven't heard anybody else say this:
My hot take -- There are already too many programming languages in use and every engineers life would be easier and productivity would be higher if we standardized.
There is almost no situation where making an entirely new language is the optimal solution. And the worst reason of all to make a language is anything aesthetic (e.g. spacing, parentheses, notations) -- all of those cases should be handled via native transpiling in the IDE.
Hmm, maybe we could come up with a standard syntax? We could call it “Standard-expressions”, or perhaps “S-expressions”? That way an entire family of languages could use the same syntax. Now how do we get people to follow this standard?
I'm not sure if you meant this as evidence for or against my point, but this is exactly what I mean.
Every year dozens of people think they're going to make some holy-grail language and it's 1% different from the prior version but now every single library needs to be rewritten, a dozen years of patches/bugs/security-holes must happen, whole resumes get shined up rewriting tech stacks that were perfectly capable, all for it to be thrown for the next shiny toy in less than a decade.
It sounds like you're trying to draw some equivalence between multiple spoken languages (largely seen as something brag-worthy-ish) and multiple programming languages. Like if culture thinks it's sexy to know french that somehow that means it's sexy to know Cobol [It's not].
The two are wildly different of course, a programming language can be learned in weeks and a spoken language takes years.
But if your question is -- would the world be a better place if we all spoke the same language? Then yes, absolutely.
Fortunately AI will help continue to reduce these artificial barriers between languages.
I'd consider a person that has only studied programming for weeks extremely junior.
Edit: And no, "AI" is not going to affect any such barriers. It's already trivial to learn new natural languages, and some people do, some don't. Those who don't, generally don't care about people that speak other languages than the one they learned in childhood.
If we were forced to use only selected, most popular languages to communicate and ban all others, everyone would have to speak Mandarin, Spanish, and English. And the world would never know the Quran, the Bible, Arabian Nights, the Epic of Gilgamesh, War and Peace, the Ramayana, or the Divine Comedy.
There are some good examples of applications that are specifically written in Clojure and would be quite difficult to replicate in other PLs - Roam Research and Logseq, XTDB and Datomic, Nextjournal and Precursor.
> And the worst reason of all to make a language is anything aesthetic
Interestingly, people avoid learning Lisp, because it doesn't look "sexy enough" the first time they see it.
My hot take -- There are already too many programming languages in use and every engineers life would be easier and productivity would be higher if we standardized.
There is almost no situation where making an entirely new language is the optimal solution. And the worst reason of all to make a language is anything aesthetic (e.g. spacing, parentheses, notations) -- all of those cases should be handled via native transpiling in the IDE.
reply