Which programming languages should be retired in 2025, and why?
Can you list some non-obvious options?
9 comments
How do you propose “retiring” programming languages? People still write code in Fortran and COBOL. No oversight body exists to approve or retire languages. Either they get traction and eventually grow a lot of code to maintain, or they don’t.
If you ask rhetorically — king for a day kind of question — I would retire Python, because it enables too much slow, buggy, amateurish code.
If you ask rhetorically — king for a day kind of question — I would retire Python, because it enables too much slow, buggy, amateurish code.
It is more of a metaphorical question
python. its super bad for the environment really and there's lovely replacements for it which are not more difficult to learn, like Go.
Rust because it overpromises and confuses people with that. if it didnt brand itself as a systems programming language, it might be a replacement for python. (nice pkg manager, and quite nice 'high level' if you find the right packages.)
ofcourse, totally unrealistic and maybe a bit harsh judgements :p. personal taste more than anything i suppose. i code in C, and i guess a lot of ppl will actually mention that one as fit for retirement (despite there being no real reppacement for it)
Rust because it overpromises and confuses people with that. if it didnt brand itself as a systems programming language, it might be a replacement for python. (nice pkg manager, and quite nice 'high level' if you find the right packages.)
ofcourse, totally unrealistic and maybe a bit harsh judgements :p. personal taste more than anything i suppose. i code in C, and i guess a lot of ppl will actually mention that one as fit for retirement (despite there being no real reppacement for it)
> python. its super bad for the environment really
Browser based JavaScript is much worse for the environment. Lets go after that, too.
Browser based JavaScript is much worse for the environment. Lets go after that, too.
Is it? Python requires twenty times more computing time to solve the same problem. You could argue that there are much more browser JS users than python users. I don't have the numbers, but I guess with all the deep learning number crunching, the Python share is quite significant. Or what is your assumption based on?
Transmitting data across the internet requires energy, plus JS is JIT/compiled every time it's run. Popular websites load up on JS, so it uses quite a bit.
I was under the impression that Python's serious number crunching ultimately happens in modules written in another language like CUDA. If it couldn't be tied together with Python, it would be in another language.
I was under the impression that Python's serious number crunching ultimately happens in modules written in another language like CUDA. If it couldn't be tied together with Python, it would be in another language.
There was a peer-reviewed paper about energy efficiency of programming languages some years ago (see https://greenlab.di.uminho.pt/wp-content/uploads/2017/09/pap...). A JIT seem to have little influence on energy efficiency; there are even big differences between jitted languages; but Python is on the second last place. I have no numbers on what percentage Python itself contributes to the deep learning number crunching; but according to the paper, the overall use of JS could be twenty times higher than that of Python to cause the same energy consumption. Maybe someone has the numbers.
Probably recent ones with 0 adoption I would say. Carbon for example. Because there s no way anything with massive code bases can ever retire if that's what you mean.
you're not wrong, though it is possible to stop making 'new' things in them. maybe then in some generations it could be retired. the things with large codebases or widespread usage.