Most of the big data science libraries are sunsetting support by 2020, so I don't think supporting Python 2 is a great idea. It used to be true that there were

Most of the big data science libraries are sunsetting support by 2020, so I don't think supporting Python 2 is a great idea. It used to be true that there were lots of libs stuck on Python 2, but I'm struggling to think of any major ones that'd be relevant to Excel that haven't moved forwards nowadays. No way it'd be worth the confusion of having two different versions of Python and the documentation headache that'd be. There's a case to be made for keeping Python 2 around for stuff like internal software, but you're not gonna be running that in Excel. http://www.python3statement.org/ http://www.python3statement.org/ <- look at the list of projects dropping support... NumPy, MPL, Pandas, SymPy etc.