[zzzcpan]

Just the other day I was thinking about how all this infix business is needlessly complicated and leads to subtle bugs and hard to understand code. Like every time I encounter an uncommon operator in some language I have to lookup its precedence. So much time wasted trying to satisfy this silly familiarity with math notation. Only imagine how much easier things could be if all infix operators were, for example, left associative and had the same precedence. No more parsing bugs, no implicit orders and behaviors to remember, consistent order, even more natural and familiar than the math notation allowing to focus on things that matter and forget about dealing with precedence and associativity.

[JadeNB]

If you're going to put all operators at the same level of precedence, you've accepted lots of parentheses anyway; so why not just go all the way and require parentheses everywhere, meaning that no associativity need be specified either?

[User23]

I've always thought it would be cool to have a language where functions were tagged as distributive, associative, symmetric/commutative, and monotonic and the optimizer (and editor!) used these properties to determine optimizations or simplifications. Note that this would be for all expressions, not just arithmetical ones.

I vaguely recall hearing about some fancy C++ template based techniques for accomplishing this, actually, but haven't looked into it.

[saagarjha]

Swift has operator precedence levels, and you are free to use them for your own operators: https://developer.apple.com/documentation/swift/swift_standa...

[sbmassey]

Mathematica/the Wolfram language does that. e.g the Flat attribute denotes associativity, etc. https://reference.wolfram.com/language/ref/Flat.html

[mamcx]

I also wonder that. However how much "dislike" will cause to force to use parents for math code? For me, I could live with that, but I'm not a math-heavy coder...

How much other stuff, apart of math, could be impacted?