That it also applies to things "such as" RFID tags isn't really that interesting. The salient part is identifiers. Because fingerprinting turns that into a mess.
Is your browser user agent string an "identifier"? It generally isn't unique, and requiring explicit consent to process it would cause a lot of trouble, but that and a few other things you could say the same thing about are collectively enough to be uniquely identifying.
Which is something different which they apparently hadn't considered and it's not clear how it's supposed to work. Do they become an identifier as soon as you have enough of them to uniquely identify someone? How do you even know when that threshold is passed? Does it require you to actually use them as an identifier, or is it enough just to have them because then they could be used retroactively? What if you provide a non-identifying subset of them to a third party in another jurisdiction who collects others from someone else and then combines them without explicitly notifying you?
Is your browser user agent string an "identifier"? It generally isn't unique, and requiring explicit consent to process it would cause a lot of trouble, but that and a few other things you could say the same thing about are collectively enough to be uniquely identifying.
Which is something different which they apparently hadn't considered and it's not clear how it's supposed to work. Do they become an identifier as soon as you have enough of them to uniquely identify someone? How do you even know when that threshold is passed? Does it require you to actually use them as an identifier, or is it enough just to have them because then they could be used retroactively? What if you provide a non-identifying subset of them to a third party in another jurisdiction who collects others from someone else and then combines them without explicitly notifying you?
They made a hash of it.