I'd be curious to know if your stance on PyPy has changed at all since 2011 (if indeed it's something that you've taken any new long looks at since) given their

I'd be curious to know if your stance on PyPy has changed at all since 2011 (if indeed it's something that you've taken any new long looks at since) given their progress in that time. I know that I would humbly submit at the least that my position has moved to believing that PyPy is a viable option for high-speed code (albeit in substantial part due to better interaction with C, nowadays).