I maintained same attitude for years.
I've changed my mind now. Anything feature I want already exists in some programming language. The only distinguishing feature I can offer when designing a new programming language is "very readable", so syntax matters more than I used to think.
Whether you get traction or not depends a lot on syntax - if your syntax is too much of an outlier compared to mainstream languages, your features don't matter.
Because I got tired of writing my htmx stuff as a tag tree.
I should probably post a show HN here.
Python's syntax inspires imitations that you will never hear about in spite of their syntax.
C doesn't have beginner-friendly syntax, yet it succeeded over languages like Pascal and Modula. C++ enjoys great popularity in spite of a mind-numbingly large grammar full of arcane syntax, which grows worse as time passes. There is a lot interest in Rust, which has tons of bizarre syntax to learn.
Poorly-considered, simplistic hypotheses connecting syntax with popularity simply don't hold up to even a modest amount of scrutiny.
> C++ enjoys great popularity in spite of a mind-numbingly large grammar full of arcane syntax, which grows worse as time passes. There is a lot interest in Rust, which has tons of bizarre syntax to learn.
C++ becomes less popular as its syntax becomes more arcane. Rust is a recent language and its long term survival remains to be seen. Languages are like memes: they come and they go. BASIC used to be the beginner language, now it's Python. Lua is popular due to its syntax and ease of embedding. We'll see what simplistic whitespace syntaxes come out within the next 25 years. I predict many language which exist today will fall by the wayside while newer ones replace them
Development on Python started in 1989 and it was evidently released in 1991. The popularity didn't take off until well past Y2K. During all those intervening years, it had a small number of users, characteristic of an unpopular language.