I never use the function in question, only because my pedantry for the optimal ensures I'd near never try update values that do not need it, nor depend on such a function to inform me of the success/incidence of it, but if someone asked me to guess at what it did, I confess I would be with Simon on that... For a function labelled "Changes" to report a positive value where absolutely Zero changes happened, seems very counter-intuitive. It probably is more useful, as the use-case in this thread suggests, but still feels wrong. Either way, my knowledge-base on the matter just got updated - thanks.