On Sat, 26 May 2018 23:54:51 +1000, Chris Angelico wrote: > Seems pretty non-controversial, which means it's almost guaranteed to > reach 100+ posts debating what the name should be.
Including: * 27 posts declaring that using such singleton symbols is completely un-Pythonic and is sure to doom Python to irrelevancy; * 13 posts stating that whether it is spelled "Symbol(name)" or "Atom(name)" it will be completely confusing; * 4 posts proposing spelling it "~~name!" instead; * 25 posts agreeing that the functionality is important, but should only be allowed inside functions whose name starts with a vowel; * 11 posts insisting it should only be functions starting with a consonant; * 3 posts saying it shouldn't be allowed in functions at all, only in module-level try...except statements, and only so long as there are no list comprehensions used in the module; * 18 posts demanding a complete list of every use-case for the feature and why using object()/strings/enums/ints isn't sufficient; * 7 posts stating that the author would personally use it if it were a built-in, and they wouldn't bother installing a third- party package from PyPI for it, but nevertheless it should be put on PyPI first to see if there is any community interest; * 12 posts cherry-picking programming languages invented between August 2003 and May 2015 by people whose name starts with "B", proving categorically that the modern trend is to avoid this feature; * and 17 posts discussing the history of Forth and Intercal. Good thing I didn't propose a short-hand syntax for it as well :-) -- Steven D'Aprano "Ever since I learned about confirmation bias, I've been seeing it everywhere." -- Jon Ronson -- https://mail.python.org/mailman/listinfo/python-list