I believe what they meant was that for many strings it really shouldn't matter if they were bytes or unicode. They would perform their function correctly either

I believe what they meant was that for many strings it really shouldn't matter if they were bytes or unicode. They would perform their function correctly either way. That's completely true, but you do still have to go through and find the cases where that doesn't work.