Being fair, he is right. So grant him that. Python3 didn't really win. Lots of people jumped to TypeScript, Go and other stuff as a result. I wouldn't say you're writing better code, faster code either. Python2 supported unicode if that's the concern, and as of today the 2.7 interpreter is still faster than 3.6.
If you want to go on and use Python3, fine. But no need to lie about the facts on the transition and that Python3 is better. It's simply different, technical churn rather than innovation.
This is hilarious. How are you judging the quality or performance of my code again?
Being fair, you're wrong on all fronts.
Our code quality has increased dramatically if only by being able to replace swathes of thread and Twisted glue with asyncio. The very same upgrades saw a ×10 runtime speed-up in some claggy mixed-layer code.
But even just comparing runtimes, 3.6 isn't slow[1]. They're about the same.
If your only opinion of Python 3 is it does unicode and it's slow, you may want to read the first line of the comment you replied to. Your arguments may have applied to Python 3.3 but they hold little weight in 2017.
Unfortunately for you the facts aren't on your side. Python2.7 is still faster. People like you have said "your arguments hold little weight in 2013" too. Python3 was released in 2008, it's been 9 years and it's still a failure. The OP was correct and you were wrong. It's a shame that bothers you so much.
If you want to go on and use Python3, fine. But no need to lie about the facts on the transition and that Python3 is better. It's simply different, technical churn rather than innovation.