I’m surprised this isn’t talked about WAY more than it is. I have heard a few podcasts talking about the possibility that PPM doesn’t handle spoken word (and some other formats) well. I am no engineer but holy cow if that turns out to be even the slightest bit true.
“We had concern about various types of formats,” a former Arbitron engineer said in an email. “It was thought that classical and some talk formats would be challenging to encode and possibly decode reliably.”