A post on the rather dismal adoption of Python 3.  It seems like there are two categories of options being presented:

1. Make it harder to use Python 2
2. Make it better to use Python 2, and in a manner closer to Python 3

I think the ultimatum approaches are bad (and more present in the comments on the article than in Alex's post).  It's a "let the beatings continue until morale improves" approach.

The whole "Python 2 is a dead end" notion was a bad approach from the beginning.  It supposes that there's some moral authority to Python 3, some intrinsic value that justifies making things harder for people.

I think Python 3 should roll back some changes, adding back some Python 2 syntax, even error-prone syntax.  Python 2 should continue to roll in Python 3 syntax and library changes.  Only when they are thoroughly blended will things move forward.  This of course is in contradiction to the entire idea of a big breaking 2->3 change, but the evidence is in, that wasn't the right path.  But that's a sunk cost, better to do the right thing now, which is gradual changes with all the necessary scaffolding to move things forward properly.  Python 2 and 3 can meet in the middle.
Shared publiclyView activity