Page 33 - FIGI: e-KYC use cases in digital financial services
P. 33

a)  Decentralization: DID  architecture should elimi-  a root of trust with neither centralized authority
                nate the requirement for centralized authorities   nor a single point of failure. In combination, DLTs
                or single points of failure in identity management,   and decentralized identity systems enable any
                including the registration of globally unique iden-  entity to create and manage their own identifiers
                tifiers, public verification keys, service endpoints,   on any number of distributed, independent roots
                and other metadata.                               of trust.
            b)  Entity control of identifiers: DID architecture   b)  The entities are identified by decentralized iden-
                should give  entities, both  human  and  non-hu-  tifiers (DIDs). They may authenticate via proofs
                man, the power to directly control their own digi-  (e.g., digital signatures, privacy-preserving bio-
                tal identifiers without the need to rely on external   metric protocols, etc.). DIDs point to DID Docu-
                authorities.                                      ments. A DID Document contains a set of service
            c)  Personal Identifiable Information Protection: DID   endpoints for interacting with the entity. Follow-
                architecture should enable entities to control the   ing the dictums of Privacy by Design, each entity
                identifiable data of their digital identities, includ-  may have as many DIDs as necessary, to respect
                ing minimal, selective, and progressive disclosure   the entity's desired separation of identities, per-
                of attributes or other identity data.             sonas, and contexts.
            d)  Security:  DID  architecture  should  enable  suffi-  c)  To use a DID with a particular distributed ledger
                cient security for relying parties to depend on   or network requires defining a DID method in a
                DID records for their required level of assurance.  separate DID method specification. A DID meth-
            e)  Proof-based: DID architecture should enable an    od specifies the set of rules for how a DID is reg-
                entity to provide cryptographic proof of authen-  istered, resolved, updated, and revoked on that
                tication and proof of authorization rights.       specific ledger or network.
            f)  Discoverability: DID architecture should make   d)  This design eliminates dependence on cen-
                it possible for entities to discover DIDs for oth-  tralized  registries  for  identifiers  as  well  as  cen-
                er entities to learn more about or interact with   tralized certificate authorities for key manage-
                those entities.                                   ment—the standard pattern in hierarchical PKI
            g)  Interoperability: DID architecture should use     (public key infrastructure). Because DIDs reside
                interoperable standards so DID infrastructure can   on a distributed ledger, each entity may serve as
                make use of existing tools and software libraries   its own root authority – an architecture referred
                designed for interoperability.                    to as DPKI (decentralized PKI).
            h)  Portability: DID architecture should be system
                and network-independent and enable entities to
                use their digital identities with any system that   7�4  DID Requirements and Authentication
                supports DIDs and DID Methods.                 Some requirements proposed for DIDs are as follows:
            i)  Simplicity: To meet these design goals, DID archi-
                tecture should be "as simple as possible but no   a)  Entities must be able to self-manage their own
                simpler".                                         DIDs on a distributed ledger, using a DID method.
            j)  Extensibility: When possible, DID architecture   b)  The DID method must conform to W3C working
                should enable extensibility provided it does not   draft "Decentralized Identifiers (DIDs) v1.0".
                greatly hinder interoperability, portability, or sim-  c)  the DID method must ensure the Holder/User
                plicity.                                          is the exclusive DID owner, (e.g., the private key
                                                                  corresponding to the public key used to authen-
               From the W3C Decentralized Identifier draft spec-  ticate the DID controller).
            ification:                                         d)  when  assisted by a  third party in the DID  cre-
                                                                  ation process, any party must always maintain
            a)  The emergence of distributed ledger technology    the DID's sole control.
                (DLT), sometimes referred to as blockchain tech-  e)  A DID shall be anchored on a distributed ledger.
                nology, provides the opportunity for fully decen-  f)  A DID must be universally resolvable to its corre-
                tralized identity management. In a decentralized   sponding DID document.
                identity system, entities are free to use any shared   g)  The DID document may be produced upon
                root of trust. Globally distributed ledgers (or a   demand when dereferencing the DID; and
                decentralized P2P network that provides sim-   h)  A DID document for a natural person must not
                ilar capabilities) provide a means for managing   be stored in a distributed ledger, except for its



                                                                          e-KYC use cases in digital financial services  31
   28   29   30   31   32   33   34   35   36   37   38