Any objections to dropping 2.6 support?
- 2.7 was introduced in 2010
@Rollin_King's made the switch
- more important to work on 3.5 compatibility than 2.6
- like with C++11, it'd be nice to use new-ish language features (cleaner printing, dictionary comprehensions, etc.)
While I'm at it, among 2to3, futurize, six, ad hoc redefinition of fns, etc., does anyone have preferences for maintaining 2- & 3-compatible code?