I like the idea of proof of personhood, and proof of location. These are both concepts that I have explored extensively. I looked at retina scans, and found it they're both not unique enough, and not secure enough to act as a basis for a currency.
My proof of location was based on the 500 mile email story. Many millions of round trips between two devices could allow someone to prove on a blockchain that two secret keys were stored arbitrarily close to one another for some amount of time.
The paper mentions key signing parties which I believe must be part of any web of trust that proves personhood. Proving uniqueness is more difficult. There's nothing that would stop me from having several different identities verified at different key signing parties, then using those to generate even more.
Using genealogy seems to be the only way to truly verify a unique identity if the system becomes widespread enough that parents often verify their children at birth. Then the age of the identity matches the age of the individual.
Farther into the future, the possibility of creating multiple copies of myself makes it much more difficult to tie an identity to a particular group of cells. There's nothing truly unique about a copy, especially if we have the ability to change our DNA.
Philosophically, I started to wonder if there's anything particularly special about any arrangement of cells. Maybe we should all just do our best, share resources, and create a society where abundance makes keeping track of all this stuff unnecessary.