I wrote a lot of assembler in my early days, so it always seemed strange to me that people spent so much time arguing about syntax of higher level languages bec

I wrote a lot of assembler in my early days, so it always seemed strange to me that people spent so much time arguing about syntax of higher level languages because it all ends up as machine code. There's no clear winner in the quest to create the best high level language, so it seems to me we could have saved a lot of time if we had just picked one and stuck with it. I've read hundreds of articles along the lines of "our project is written in X, but we rewrote it in Y" for marginal improvement. I get the utility of higher level constructs, but I wonder how much productivity is gained.