Debating a "transitional" Python 2.8

trying to do a survey of python version usage will only get responses from the early adopter crowd that is paying attention to such things.

There will be no feedback from the people who have python code they are running that is 2.x only because they don't know or care about python 3

most people who use a language are not involved with discussions about changing it (or if they are, the language is either very new or breaking a lot)

Getting to where one binary can run both python 2 and python 3 code would be a very good starting point. After that the debate can begin about disabling features. That will be an interesting debate, as most of the time the people trying to turn them off are justifying doing so 'for their own good', which won't make people who have code that gets broken very happy.