> Thus, hashing completely fails to address the underlying privacy concerns I don't understand their argument against pseudonomous identifiers (well, part of t

> Thus, hashing completely fails to address the underlying privacy concerns I don't understand their argument against pseudonomous identifiers (well, part of the problem is they present it without a lot of argument). Are they arguing that companies will reverse the hash, or that they will de-anonymise it using additional data? Otherwise it seems harmful to me to tell people that using a different identifier per web site is useless (a bit like telling everybody that locking your car is useless because a determined thief would break in anyway...)