I can hardly agree more with this. I program to solve problems. Changes in a language or worse, "how it is done" mean that I must keep up with something else

I can hardly agree more with this. I program to solve problems. Changes in a language or worse, "how it is done" mean that I must keep up with something else in addition to solving the problems before me. Waves of increasing syntactic sugar in a language begin to look like layers of frippery and frosting, and I can only remember that "fashion is a form of ugliness so intolerable that we have to alter it every six months." I do not want to look at some bits of upper-row "executable line noise" that is somehow called syntactic sugar and try to have to mentally project what surface lies under that frosting. What I want is to be able to look at ten year old code and it immediately be working and understandable. I don't want it to be out to style or "uncool" or to even have to think about these things. By all means, fix bugs. Add libraries as new formats, protocols, and technologies emerge. Huge mistakes can be corrected in rare major version updates, which may have to count as actual forks in the language. When I first looked at Python, somewhere around 2004 or 2005 if I had to guess, I remember seeing that whole floor division bit and thinking, "Yeah, they're going to have to change that eventually and digging in their heels on that one was foolish." That's a rare exception, and it counts in the "why did this make it to 2.x?" category. I feel a bit lonely in this, but I just cannot seem to embrace the churn.