I don't agree with the author. There're plenty of use cases which aren't covered by machine learning and won't be covered by existing machine learning method in

I don't agree with the author. There're plenty of use cases which aren't covered by machine learning and won't be covered by existing machine learning method in the future. Yes, we've came far with ML, but until we really have Software 2.0, a lot of time will pass.