@Mithrandir24601
"If you define it as the unitary process before the information is 'neglected' (not entirely sure what this actually means, but anyway)," ... I think by information your saying your able to say something definitive about the system say an eigenvalue from the states of the system??
"to include the 'information being neglected' then it's not" ... I'm guessing this is the essensce to way $3$ .. the measurement is a non-unitary operation. So maybe a better question to ask was what is Zurek achieving by trying to model "the non-unitarity of measurement is an effective descript…