Decentralized Identifiers (DIDs) v0.11

Decentralized Identifiers (DIDs) are a new type of identifier for verifiable, "self-sovereign" digital identity. DIDs are fully under the control of the DID subject, independent from any centralized registry, identity provider, or certificate authority. DIDs are URLs that relate a DID subject to means for trustable interactions with that subject. DIDs resolve to DID Documents — simple documents that describe how to use that specific DID. Each DID Document contains at least three things: cryptographic material, authentication suites, and service endpoints. Cryptographic material combined with authentication suites provide a set of mechanisms to authenticate as the DID subject (e.g. public keys, pseudonymous biometric protocols, etc.). Service endpoints enable trusted interactions with the DID subject.

This document specifies a common data model, format, and operations that all DIDs support.

Comments regarding this document are welcome. Please file issues directly on GitHub, or send them to ( subscribe, archives).

Portions of the work on this specification have been funded by the United States Department of Homeland Security's Science and Technology Directorate under contracts HSHQDC-16-R00012-H-SB2016-1-002 and HSHQDC-17-C-00019. The content of this specification does not necessarily reflect the position or the policy of the U.S. Government and no official endorsement should be inferred.

Work on this specification has also been supported by the Rebooting the Web of Trust community facilitated by Christopher Allen, Shannon Appelcline, Kiara Robles, Brian Weller, Betty Dhamers, Kaliya Young, Kim Hamilton Duffy, Manu Sporny, Drummond Reed, Joe Andrieu, and Heather Vescent.

Conventional identity management systems are based on centralized authorities such as corporate directory services, certificate authorities, or domain name registries. From the standpoint of cryptographic trust verification, each of these centralized authorities serves as its own root of trust. To make identity management work across these systems requires implementing federated identity management.

The emergence of distributed ledger technology (DLT), sometimes referred to as blockchain technology, provides the opportunity for fully decentralized identity management. In a decentralized identity system, entities are free to use any shared root of trust. Globally distributed ledgers (or a decentralized P2P network that provides similar capabilities) provide a means for managing a root of trust with neither centralized authority nor a single point of failure. In combination, DLTs and decentralized identity systems enable any entity to create and manage their own identifiers on any number of distributed, independent roots of trust.

The entities are identified by decentralized identifiers (DIDs). They may authenticate via proofs (e.g. digital signatures, privacy-preserving biometric protocols, etc.). DIDs point to DID Documents. A DID Document contains a set of service endpoints for interacting with the entity. Following the dictums of Privacy by Design, each entity may have as many DIDs as necessary, to respect the entity’s desired separation of identities, personas, and contexts.

To use a DID with a particular distributed ledger or network requires defining a DID method in a separate DID method specification. A DID method specifies the set of rules for how a DID is registered, resolved, updated, and revoked on that specific ledger or network.

This design eliminates dependence on centralized registries for identifiers as well as centralized certificate authorities for key management—the standard pattern in hierarchical PKI (public key infrastructure). Because DIDs reside on a distributed ledger, each entity may serve as its own root authority—an architecture referred to as DPKI (decentralized PKI).

Note that DID methods may also be developed for identifiers registered in federated or centralized identity management systems. For their part, all types of identifier systems may add support for DIDs. This creates an interoperability bridge between the worlds of centralized, federated, and decentralized identifiers.

DIDs have a foundation in URLs, so it's important to understand how the W3C clarified the terms [[URI]] (Uniform Resource Identifier), [[URL]] (Uniform Resource Locator), and [[URN]] (Uniform Resource Name) in September 2001. The key difference between these three categories of identifiers are:

  1. [[URI]] is the term for any type of identifier that identifies an abstract or physical resource. It may or may not be resolveable to a resource (e.g. URNs, URLs, etc.).
  2. [[URL]] is the term for any type of URI that can be resolved or de-referenced to locate a representation of a resource on the Web (e.g., Web page, file, image, etc.).
  3. [[URN]] is the term for a specific type of URI intended to persistently identify a resource, i.e., an identifier that will never change no matter how often the resource moves, changes names, changes owners, etc. URNs are intended to last forever.

The growing need for decentralized identifiers has produced two specific requirements for a new type of URL that still fits Web Architecture and has a few additional requirements that more traditional URLs, like HTTP-based URLs, do not have:

  1. The new type of URL SHOULD NOT require a centralized authority to register, resolve, update, or revoke the identifier. The overwhelming majority of URIs today are based on DNS names or IP addresses that depend on centralized authorities for registration and ultimate control. DIDs can be created and managed without any such authority.
  2. A URL whose ownership and associated metadata, including public keys, can be cryptographically verified. Authentication via DIDs and DID Documents leverage the same public/private key cryptography as distributed ledgers.

DIDs achieve global uniqueness without the need for a central registration authority. This comes, however, at the cost of human memorability. The algorithms capable of generating globally unique identifiers automatically produce random strings of characters that have no human meaning. This demonstrates the axiom about identifiers known as Zooko's Triangle: "human-meaningful, decentralized, secure—pick any two".

There are of course many use cases where it is desirable to discover a DID when starting from a human-friendly identifier—a natural language name, a domain name, or a conventional address for a DID owner such as a mobile telephone number, email address, Twitter handle, or blog URL. However, the problem of mapping human-friendly identifiers to DIDs (and doing so in a way that can be verified and trusted) is out-of-scope for this specification.

Solutions to this problem (and there are many) should be defined in separate specifications that reference this specification. It is strongly recommended that such specifications carefully consider: (a) the numerous security attacks based on deceiving users about the true human-friendly identifier for a target entity, and (b) the privacy consequences of using human-friendly identifiers that are inherently correlatable, especially if they are globally unique.

The first purpose of this specification is to define the generic DID scheme and a generic set of operations on DIDs that can be implemented for any distributed ledger or network capable of supporting DIDs. The second purpose of this specification is to define the conformance requirements for a DID method specification—a separate specification that defines a specific DID scheme and specific set of DID record operations for a specific distributed ledger or network.

Conceptually, the relationship of this specification and a DID method specification is similar to the relationship of the IETF generic URI specification ([[RFC3986]]) and a specific URI scheme ([[IANA-URI-SCHEMES]] (such as the http: and https: schemes specified in [[RFC7230]]). It is also similar to the relationship of the IETF generic URN specification ([[URN]]) and a specific URN namespace definition (such as the UUID URN namespace defined in [[RFC4122]]). The difference is that a DID method specification, in addition to defining a specific DID scheme, must also specify the methods for reading, writing, and revoking DID records on the network for which it is written.

For a list of DID Methods and their corresponding specifications, see the DID Method Registry [[DID-METHOD-REGISTRY]].

This section summarizes the design goals and principles of DID architecture.

Goal Description
Decentralization DID architecture should eliminate the requirement for centralized authorities or single points of failure in identifier management, including the registration of globally unique identifiers, public verification keys, service endpoints, and other metadata.
Self‑Sovereignty DID architecture should give entities, both human and non-human, the power to directly own and control their digital identifiers without the need to rely on external authorities.
Privacy DID architecture should enable entities to control the privacy of their information, including minimal, selective, and progressive disclosure of attributes or other data.
Security DID architecture should enable sufficient security for relying parties to depend on DID Documents for their required level of assurance.
Proof-based DID architecture should enable an entity to provide cryptographic proof of authentication and proof of authorization rights.
Discoverability DID architecture should make it possible for entities to discover DIDs for other entities to learn more about or interact with those entities.
Interoperability DID architecture should use interoperable standards so DID infrastructure can make use of existing tools and software libraries designed for interoperability.
Portability DID architecture should be system and network-independent and enable entities to use their digital identifiers with any system that supports DIDs and DID Methods.
Simplicity To meet these design goals, DID architecture should be (to paraphrase Albert Einstein) "as simple as possible but no simpler".
Extensibility When possible, DID architecture should enable extensibility provided it does not greatly hinder interoperability, portability, or simplicity.

This is a simple example of a DID:


Following is an example of a DID Document that describes the DID above. This example assumes that the entity that controls the private keys for this identifier is authoritative for the DID Document.

{ "@context": "", "id": "did:example:123456789abcdefghi", "publicKey": [{ "id": "did:example:123456789abcdefghi#keys-1", "type": "RsaVerificationKey2018", "owner": "did:example:123456789abcdefghi", "publicKeyPem": "-----BEGIN PUBLIC KEY...END PUBLIC KEY-----\r\n" }], "authentication": [{ // this key can be used to authenticate as DID ...9938 "type": "RsaSignatureAuthentication2018", "publicKey": "did:example:123456789abcdefghi#keys-1" }], "service": [{ "type": "ExampleService", "serviceEndpoint": "" }]

This specification is dependent on a number of base specifications. The dependencies and their purpose are listed below.

The JSON specification provides the base data format that this specification uses.
The JSON-LD specification enables the layering of data semantics on top of JSON data.

The concept of a globally unique decentralized identifier is not new; Universally Unique Identifiers (UUIDs) were first developed in the 1980s and later became a standard feature of the Open Software Foundation’s Distributed Computing Environment. UUIDs achieve global uniqueness without a centralized registry service by using an algorithm that generates 128-bit values with sufficient entropy that the chance of collision are infinitesimally small. UUIDs are formally specified in [[RFC4122]] as a specific type of Unified Resource Name (URN).

A DID is similar to a UUID except: (a) like a URL, it can be resolved or dereferenced to a standard resource describing the entity (a DID Document—see Section ), and (b) unlike a URL, the DID Document typically contains cryptographic material that enables authentication of an entity associated with the DID.

The generic DID scheme is a URI scheme conformant with [[RFC3986]]. It consists of a DID followed by an optional path and/or fragment. The term DID refers only to the identifier conforming to the did rule in the ABNF below; when used alone, it does not include a path or fragment. A DID that may optionally include a path and/or fragment is called a DID reference.

Following is the ABNF definition using the syntax in [[RFC5234]] (which defines ALPHA as upper or lowercase A-Z).

did-reference = did [ "/" did-path ] [ "#" did-fragment ]
did = "did:" method ":" specific-idstring
method = 1*methodchar
methodchar = %x61-7A / DIGIT
specific-idstring = idstring *( ":" idstring )
idstring = 1*idchar
idchar = ALPHA / DIGIT / "." / "-"

See Sections and for the ABNF rules defining DID paths and fragments.

A DID method specification MUST define exactly one specific DID scheme identified by exactly one method name (the method rule in Section ). Since DIDs are intended for decentralized identity infrastructure, it is NOT RECOMMENDED to establish a registry of unique DID method names. Rather the uniqueness of DID method names should be established via human consensus, i.e., a specific DID scheme MUST use a method name that is unique among all DID method names known to the specification authors at the time of publication.

A list of known DID method names and their associated specifications is provided in Appendix .

Since the method name is part of the DID, it SHOULD be as short as practical. A method name of five characters or less is RECOMMENDED. The method name MAY reflect the name of the distributed ledger or network to which the DID method specification applies.

The DID method specification for the specific DID scheme MUST specify how to generate the specific-idstring component of a DID. The specific-idstring value MUST be able to be generated without the use of a centralized registry service. The specific-idstring value SHOULD be globally unique by itself. The fully qualified DID as defined by the did rule in Section MUST be globally unique.

If needed, a specific DID scheme MAY define multiple specific specific-idstring formats. It is RECOMMENDED that a specific DID scheme define as few specific-idstring formats as possible.

A generic DID path (the did-path rule in Section ) is identical to a URI path and MUST conform to the ABNF of the path-rootless ABNF rule in [[RFC3986]]. A DID path SHOULD be used to address resources available via a DID service endpoint. See Section .

A specific DID scheme MAY specify ABNF rules for DID paths that are more restrictive than the generic rules in this section.

A generic DID fragment (the did-fragment rule in Section ) is identical to a URI fragment and MUST conform to the ABNF of the fragment ABNF rule in [[RFC3986]]. A DID fragment MUST be used only as a method-independent pointer into the DID Document to identify a unique key description or other DID Document component. To resolve this pointer, the complete DID reference including the DID fragment MUST be used as the value of the id key for the target JSON object.

A specific DID scheme MAY specify ABNF rules for DID fragments that are more restrictive than the generic rules in this section.

For the broadest interoperability, DID normalization should be as simple and universal as possible. Therefore:

  1. The did: scheme name MUST be lowercase.
  2. The method name MUST be lowercase.
  3. Case sensitivity and normalization of the value of the specific-idstring rule in Section MUST be defined by the governing DID method specification.

A DID MUST be persistent and immutable, i.e., bound to an entity once and never changed (forever). Ideally a DID would be a completely abstract decentralized identifier (like a UUID) that could be bound to multiple underlying distributed ledgers or networks over time, thus maintaining its persistence independent of any particular ledger or network. However registering the same identifier on multiple ledgers or networks introduces extremely hard entityship and start-of-authority (SOA) problems. It also greatly increases implementation complexity for developers.

To avoid these issues, it is RECOMMENDED that DID method specifications only produce DIDs and DID methods bound to strong, stable ledgers or networks capable of making the highest level of commitment to persistence of the DID and DID method over time.

NOTE: Although not included in this version, future versions of this specification may support a DID Document equivID property to establish verifiable equivalence relations between DID records representing the same identifier owner on multiple ledgers or networks. Such equivalence relations can produce the practical equivalent of a single persistent abstract DID. See Future Work (Section ).

If a DID is the index key in a key-value pair, then the DID Document is the value to which the index key points. The combination of a DID and its associated DID Document forms the root record for a decentralized identifier.

A DID Document MUST be a single JSON object conforming to [[RFC7159]]. For purposes of this version of the DID specification, the format of this JSON object is specified in JSON-LD, a format for mapping JSON data into the RDF semantic graph model as defined by [[JSON-LD]]. Future versions of this specification MAY specify other semantic graph formats for a DID Document such as JXD (JSON XDI Data), a serialization format for the XDI graph model.

The following sections define the properties of this JSON object, including whether these properties are required or optional.

JSON objects in JSON-LD format must include a JSON-LD context statement. The rules for this statement are:

  1. A DID Document MUST have exactly one top-level context statement.
  2. The key for this property MUST be @context.
  3. The value of this key MUST be the URL for the generic DID context:

Example (using an example URL):

{ "@context": ""

DID method specifications MAY define their own JSON-LD contexts. However it is NOT RECOMMENDED to define a new context unless necessary to properly implement the method. Method-specific contexts MUST NOT override the terms defined in the generic DID context.

The DID subject is the identifier that the DID Document is about, i.e., it is the DID described by DID Document. The rules for a DID subject are:

  1. A DID Document MUST have exactly one DID subject.
  2. The key for this property MUST be id.
  3. The value of this key MUST be a valid DID.
  4. When this DID Document is registered with the target distributed ledger or network, the registered DID MUST match this DID subject value.


{ "id": "did:example:21tDAKCERh95uGgKbJNHYp"

DID Method specifications MAY create intermediate representations of a DID Document that do not contain the id key, such as when a DID Resolver is performing resolution. However, the fully resolved DID Document MUST contain a valid id property.

Public keys are used for digital signatures, encryption and other cryptographic operations, which in turn are the basis for purposes such as authentication (see Section ) or establishing secure communication with service endpoints (see Section ). In addition, public keys may play a role in authorization mechanisms of DID CRUD operations (see Section ); This may be defined by DID Method specifications.

The primary intention is that a DID Document lists public keys whose corresponding private keys are controlled by the entity identified by the DID ("owned" public keys). However, a DID Document MAY also list "non-owned" public keys.

If a public key does not exist in the DID Document, it MUST be assumed the key has been revoked or is invalid. The DID Document MAY contain revoked keys. A DID Document that contains a revoked key MUST also contain or refer to the revocation information for the key (e.g. a revocation list). Each DID Method specification is expected to detail how revocation is performed and tracked.

The rules for public keys are:

  1. A DID Document MAY include a publicKey property.
  2. The value of the publicKey property should be an array of public keys.
  3. Each public key must include id and type properties, and exactly one value property.
  4. Each public key may include an owner property, which identifies the entity that controls the corresponding private key. If this property is missing, it is assumed to be the DID subject.
  5. The value property of a public key may be publicKeyPem, publicKeyJwk, publicKeyHex, publicKeyBase64 or similar, depending on the format and encoding of the public key.
  6. A registry of key types and formats is available in Appendix .


{ "@context": ["", ""], "id": "did:example:123456789abcdefghi", ... "publicKey": [{ "id": "did:example:123456789abcdefghi#keys-1", "type": "RsaVerificationKey2018", "owner": "did:example:123456789abcdefghi", "publicKeyPem": "-----BEGIN PUBLIC KEY...END PUBLIC KEY-----\r\n" }, { "id": "did:example:123456789abcdefghi#keys-2", "type": "Ed25519VerificationKey2018", "owner": "did:example:pqrstuvwxyz0987654321", "publicKeyBase58": "H3C2AVvLMv6gmMNam3uVAjZpfkcJCwDwnZn6z3wXmqPV" }, { "id": "did:example:123456789abcdefghi#keys-3", "type": "Secp256k1VerificationKey2018", "owner": "did:example:123456789abcdefghi", "publicKeyHex": "02b97c30de767f084ce3080168ee293053ba33b235d7116a3263d29f1450936b71" }], ...

A key may be embedded or referenced in a DID Document. For example, the authentication property may refer to keys in both ways:

... "authentication": [{ // this key is referenced "type": "RsaSignatureAuthentication2018", "publicKey": "did:example:123456789abcdefghi#keys-1" }, { // this key is embedded "type": "RsaSignatureAuthentication2018", "publicKey: { "id": "did:example:123456789abcdefghi#keys-2", "type": "Ed25519VerificationKey2018", "owner": "did:example:pqrstuvwxyz0987654321", "publicKeyBase58": "H3C2AVvLMv6gmMNam3uVAjZpfkcJCwDwnZn6z3wXmqPV" } }], ...

The algorithm to use when processing a publicKey property in a DID Document is:

  1. Let value be the data associated with the publicKey property and initialize result to null.
  2. If value is an object, the key material is embedded. Set result to value.
  3. If value is a string, the key is included by reference. Assume value is a URL.
    1. Dereference the URL and retrieve the publicKey properties associated with the URL (e.g. process the publicKey property at the top-level of the dereferenced document).
    2. Iterate through each public key object.
      1. If the id property of the object matches value, set result to the object.
  4. If result does not contain at least the id, type, and owner properties as well as any mandatory public cryptographic material, as determined by the result's type property, throw an error.

While the owner field may seem redundant in some of the examples above, keys may be expressed in a DID Document where the owner is described in another DID Document. Linked Data Proof libraries typically expect the owner field to always exist and may throw an exception if it is missing. Futhermore, per the requirement that DID Documents be interpretable as either a graph or a tree, a default owner field cannot be inferred by using a key's position in a tree.

Caching and expiration of the keys in a DID Document is entirely the responsibility of DID resolvers and other clients. See Section .

Authentication is the mechanism by which an entity can cryptographically prove that they are associated with a DID and DID Description. See Section . Note that Authentication is separate from Authorization because an entity may wish to enable other entities to update the DID Document (for example, to assist with key recovery as discussed in Section ) without enabling them to prove ownership (and thus be able to impersonate the entity).

The rules for Authentication are:

  1. A DID Document MAY include an authentication property.
  2. The value of the authentication property should be an array of proof mechanisms.
  3. Each proof mechanism must include the type property.
  4. Each proof mechanism MAY embed or reference a public key (see Section ).


{ "@context": "", "id": "did:example:123456789abcdefghi", ... "authentication": [{ // this key can be used to authenticate as DID ...fghi "type": "RsaSignatureAuthentication2018", "publicKey": "did:example:123456789abcdefghi#keys-1" }, { // this key can be used to authenticate as DID ...fghi "type": "PseudonymousBiometricAuthentication2018", "biometricTemplate": "did:example:123456789abcdefghi#bio-1" }], ...

Authorization is the mechanism by which an entity states how one may perform operations on behalf of the entity. Delegation is the mechanism that an entity may use to authorize other entities to act on its behalf. Note that Authorization is separate from Authentication as explained in Section . This is particularly important for key recovery in the case of key loss, when the entity no longer has access to their keys, or key compromise, where the owner’s trusted third parties need to override malicious activity by an attacker. See Section .

Since Authorization and Delegation are typically implemented by the ledger, each DID Method specification is expected to detail how authorization and delegation are performed for the ledger.

There are at least two suggested methods for implementing Authorization, Delegation, and the concept of Guardianship:

  1. A ledger could implement a coarse grained guardian pattern by re-using the same cryptography suite pattern used by the authentication property, or more preferably
  2. A ledger could implement a Capabilities-based approach and provide more fine-grained control of authorization, delegation, and guardianship.

In addition to publication of authentication and authorization mechanisms, the other primary purpose of a DID Document is to enable discovery of service endpoints for the entity. A service endpoint may represent any type of service the entity wishes to advertise, including decentralized identity management services for further discovery, authentication, authorization, or interaction. The rules for service endpoints are:

  1. A DID Document MAY include a service property.
  2. The value of the service property should be an array of service endpoints.
  3. Each service endpoint must include id, type, and serviceEndpoint properties, and MAY include additional properties.
  4. The service endpoint protocol SHOULD be published in an open standard specification.
  5. The value of the serviceEndpoint property MUST be a JSON-LD object or a valid URI conforming to [[RFC3986]] and normalized according to the rules in section 6 of [[RFC3986]] and to any normalization rules in its applicable URI scheme specification.


{ "service": [{ "id": "did:example:123456789abcdefghi;openid", "type": "OpenIdConnectVersion1.0Service", "serviceEndpoint": "" }, { "id": "did:example:123456789abcdefghi;vcr", "type": "CredentialRepositoryService", "serviceEndpoint": "" }, { "id": "did:example:123456789abcdefghi;xdi", "type": "XdiService", "serviceEndpoint": "" }, { "id": "did:example:123456789abcdefghi;agent", "type": "AgentService", "serviceEndpoint": "" }, { "id": "did:example:123456789abcdefghi;hub", "type": "HubService", "serviceEndpoint": "" }, { "id": "did:example:123456789abcdefghi;messages", "type": "MessagingService", "serviceEndpoint": "" }, { "id": "did:example:123456789abcdefghi;inbox", "type": "SocialWebInboxService", "serviceEndpoint": "", "description": "My public social inbox", "spamCost": { "amount": "0.50", "currency": "USD" } }, { "id": "did:example:123456789abcdefghi;authpush", "type": "DidAuthPushModeVersion1", "serviceEndpoint": "" }]

See Sections and for further security considerations regarding authentication service endpoints.

Standard metadata for identifier records includes a timestamp of the original creation. The rules for including a creation timestamp are:

  1. A DID Document MUST have zero or one property representing a creation timestamp. It is RECOMMENDED to include this property.
  2. The key for this property MUST be created.
  3. The value of this key MUST be a valid XML datetime value as defined in section 3.3.7 of W3C XML Schema Definition Language (XSD) 1.1 Part 2: Datatypes [[XMLSCHEMA11-2]].
  4. This datetime value MUST be normalized to UTC 00:00 as indicated by the trailing "Z".
  5. Method specifications that rely on DLTs SHOULD require time values that are after the known "median time past" (defined in Bitcoin BIP 113), when the DLT supports such a notion.


{ "created": "2002-10-10T17:00:00Z"

Standard metadata for identifier records includes a timestamp of the most recent change. The rules for including an updated timestamp are:

  1. A DID Document MUST have zero or one property representing an updated timestamp. It is RECOMMENDED to include this property.
  2. The key for this property MUST be updated.
  3. The value of this key MUST follow the formatting rules (3, 4, 5) from section .


{ "updated": "2016-10-17T02:41:00Z"

A proof on a DID Document is cryptographic proof of the integrity of the DID Document according to either:

  1. The entity as defined in section , or if not present:
  2. The delegate as defined in section 4.3.

This proof is NOT proof of the binding between a DID and a DID Document. See Section . The rules for a proof are:

  1. A DID Document MAY have exactly one property representing a proof.
  2. The key for this property MUST be proof.
  3. The value of this key MUST be a valid JSON-LD proof as defined by Linked Data Proofs.


{ "proof": { "type": "LinkedDataSignature2015", "created": "2016-02-08T16:02:20Z", "creator": "did:example:8uQhQMGzWxR8vw5P3UWH1ja#keys-1", "signatureValue": "QNB13Y7Q9...1tzjn4w==" }

One of the goals of the Decentralized Identifiers Data Model is to enable permissionless innovation. This requires that the data model is extensible in a number of different ways:

  • The requirement to model complex multi-entity relationships is provided through the use of a graph-based data model.
  • The requirement to enable extending the machine-readable vocabularies used to describe information in the data model — without relying on a centralized system — is accomplished via the use of [[LINKED-DATA]].
  • The requirement to support multiple types of cryptographic proof formats is accomplished via the use of Linked Data Proofs [[LD-PROOFS]], Linked Data Signatures [[LD-SIGNATURES]], and a variety of signature suites.
  • The requirement to provide all of the extensibility mechanisms outlined above in a data format that is popular among software developers and web page authors is enabled via the use of [[!JSON-LD]].

This approach to data modeling is often called an "open world assumption", meaning that any entity can say anything about any other entity. This approach often feels in conflict with building simple and predictable software systems. Balancing extensibility with program correctness is always more challenging with an open world assumption than it is with closed software systems.

The rest of this section describes how both extensibility and program correctness are achieved through a series of examples.

Let us assume that we start with the following DID Document:

{ "@context": "", "id": "did:example:123456789abcdefghi", "publicKey": [{ ... }], "authentication": [{ ... }], "service": [{ ... }]

The contents of the publicKey, authentication, and service properties are not important for the purposes of this section. What is important is that the object above is a valid DID Document. Let's assume that a developer wanted to extend the DID Document to express an additional piece of information: the subject's public photo stream.

The first thing that a developer would do is create a JSON-LD Context containing the new term:

{ "@context": { "PhotoStreamService": "" }

Now that the JSON-LD Context has been created, the developer MUST publish it somewhere that is accessible to any DID Document processor. For this example, let us assume that the JSON-LD Context above is published in the decentralized ledger at the following URL: did:example:contexts:987654321. At this point, extending the first example in this section is a simple matter of including the context above and adding the new property to the DID Document.

{ "@context": "", "id": "did:example:123456789abcdefghi", "authentication": [{ ... }], "service": [

{ "@context": "did:example:contexts:987654321", "id": "did:example:123456789abcdefghi;photos", "type": "PhotoStreamService", "serviceEndpoint": "" }

] }

The examples so far have shown that it is easy to extend the Decentralized Identifiers Data Model in a permissionless and decentralized way. The mechanism also ensures that Decentralized Identifiers created in this way prevent namespace conflicts and semantic ambiguity.

An extensibility model that is this dynamic does increase implementation burden. Software written for such a system will have to determine if accepting DID Documents with extensions is acceptable based on the risk profile of the application. Some applications may choose to accept but ignore extensions, others may choose to only accept certain extensions, while highly secure environments may disallow extensions. These decisions are up to the application developers and are specifically not the domain of this specification.

Implementations MUST produce an error when an extension JSON-LD Context overrides the expanded URL for a term specified in this specification. To avoid the possibility of accidentally overriding terms, developers are urged to scope their extensions. For example, the following extension scopes the new PhotoStreamService term so that it may only be used within the service property:

{ "@context": { 

"service": { "@id": "", "@context": { "PhotoStreamService": "" } }

} }

Developers are urged to ensure that extension JSON-LD Contexts are highly available. Implementations that cannot fetch a context will produce an error. Strategies for ensuring that extension JSON-LD Contexts are always available include using content-addressed URLs for contexts, bundling context documents with implementations, or enabling aggressive caching of contexts.

To enable the full functionality of DIDs and DID Documents on a particular distributed ledger or network (called the target system), a DID method specification MUST specify how each of the following CRUD operations is performed by a client. Each operation MUST be specified to the level of detail necessary to build and test interoperable client implementations with the target system. Note that, due to the specified contents of DID Documents, these operations can effectively be used to perform all the operations required of a CKMS (cryptographic key management system), e.g.:

  • Key registration
  • Key replacement
  • Key rotation
  • Key recovery
  • Key expiration

The DID method specification MUST specify how a client creates a DID record—the combination of a DID and its associated DID Document—on the target system, including all cryptographic operations necessary to establish proof of ownership.

The DID method specification MUST specify how a client uses a DID to request a DID Document from the target system, including how the client can verify the authenticity of the response.

The DID method specification MUST specify how a client can update a DID record on the target system, including all cryptographic operations necessary to establish proof of control.

Although a core feature of distributed ledgers is immutability, the DID method specification MUST specify how a client can revoke a DID record on the target system, including all cryptographic operations necessary to establish proof of revocation.

A DID resolver is a software component with an API designed to accept requests for DID lookups and execute the corresponding DID method to retrieve the authoritative DID Document. To be conformant with this specification, a DID resolver:

  1. SHOULD validate that a DID is valid according to its DID method specification, otherwise it should produce an error.
  2. MUST conform to the requirements of the applicable DID method specification when performing DID resolution operations.
  3. SHOULD offer the service of verifying the integrity of the DID Document if it is signed.
  4. MAY offer the service of returning requested properties of the DID Document.

NOTE TO IMPLEMENTERS: During the Implementer’s Draft stage, this section focuses on security topics that should be important in early implementations. The editors are also seeking feedback on threats and threat mitigations that should be reflected in this section or elsewhere in the spec. As the root identifier records for decentralized identifiers, DIDs and DID Documents are a vital component of decentralized identity management. They are also the foundational building blocks of DPKI (decentralized public key infrastructure) as an augmentation to conventional X.509 certificates. As such, DIDs are designed to operate under the general Internet threat model used by many IETF standards. We assume uncompromised endpoints, but allow messages to be read or corrupted on the network. Protecting against an attack when a system is compromised requires external key-signing hardware. See also section regarding key revocation and recovery. For their part, the DLTs hosting DIDs and DID Documents have special security properties for preventing active attacks. Their design uses public/private key cryptography to allow operation on passively monitored networks without risking compromise of private keys. This is what makes DID architecture and decentralized identity possible.

  1. DID method specifications MUST include their own Security Considerations sections.
  2. This section MUST consider all the requirements mentioned in section 5 of [[RFC3552]] (page 27) for the DID operations defined in the specification. In particular:

Discussions at Rebooting the Web of Trust 5 resulted in consensus to move Authorization to DID Method specifications. It is currently expected that there will be an attempt to create a generalized authorization mechanism that is build on object capabilities.

At least the following forms of attack MUST be considered: eavesdropping, replay, message insertion, deletion, modification, and man-in-the-middle. Potential denial of service attacks MUST be identified as well. If the protocol incorporates cryptographic protection mechanisms, it should be clearly indicated which portions of the data are protected and what the protections are (i.e., integrity only, confidentiality, and/or endpoint authentication, etc.). Some indication should also be given to what sorts of attacks the cryptographic protection is susceptible. Data which should be held secret (keying material, random seeds, etc.) should be clearly labeled. If the technology involves authentication, particularly user-host authentication, the security of the authentication method MUST be clearly specified.

  1. This section MUST also discuss, per Section 5 of [[RFC3552]], residual risks (such as the risks from compromise in a related protocol, incorrect implementation, or cipher) after threat mitigation has been deployed.
  2. This section MUST provide integrity protection and update authentication for all operations required by Section 7 of this specification (DID Operations).
  3. Where DID methods make use of peer-to-peer computing resources (such as with all known DLTs), the expected burdens of those resources SHOULD be discussed in relation to denial of service.
  4. Method-specific endpoint authentication MUST be discussed. Where DID methods make use of DLTs with varying network topology, sometimes offered as "light node" or “ thin client ” implementations to reduce required computing resources, the security assumptions of the topology available to implementations of the DID method MUST be discussed.
  5. DID methods MUST discuss the policy mechanism by which DIDs are proven to be uniquely assigned. A DID fits the functional definition of a URN as defined in [[RFC2141]]—a persistent identifier that is assigned once to a resource and never reassigned. In a security context this is particularly important since a DID may be used to identify a specific party subject to a specific set of authorization rights.
  6. DID methods that introduce new authentication service endpoint types (Section ) SHOULD consider the security requirements of the supported authentication protocol.

Signatures are one method to allow DID Documents to be cryptographically verifiable.

By itself, a verified signature on a self-signed DID Document does not prove ownership of a DID. It only proves the following:

  1. The DID Document has not been tampered with since it was registered.
  2. The owner of the DID Document controlled the private key used for the signature at the time the signature was generated.

Proving ownership of a DID, i.e., the binding between the DID and the DID Document that describes it, requires a two step process:

  1. Resolving the DID to a DID Document according to its DID method specification.
  2. Verifying that the id property of the resulting DID Document matches the DID that was resolved.

It should be noted that this process proves ownership of a DID and DID Document regardless of whether the DID Document is signed.

Signatures on DID Documents are optional. DID Method Specs SHOULD explain and specify their implementation if applicable.

It is RECOMMENDED to combine timestamps with signatures.

There are two methods for proving ownership of the private key corresponding to a public key description in the DID Document: static and dynamic. The static method is to sign the DID Document with the private key. This proves ownership of the private key at a time no later than the DID Document was registered. If the DID Document is not signed, ownership of a public key described in the DID Document may still be proven dynamically as follows:

  1. Send a challenge message containing a public key description from the DID Document and a nonce to an appropriate service endpoint described in the DID Document.
  2. Verify the signature of the response message against the public key description.

A DID and DID Document do not inherently carry any PII (personally-identifiable information). The process of binding a DID to the real-world owner of an identifier using claims about the owner is out of scope for this specification. However this topic is the focus of the verifiable claims standardization work at the W3C (where the term "DID" originated).

If a DID Document publishes a service endpoint intended for authentication or authorization of an entity (section ), it is the responsibility of the service endpoint provider, entity, and/or relying party to comply with the requirements of the authentication protocol(s) supported at that service endpoint.

Non-repudiation of DIDs and DID Document updates is supported under the assumption that: (1) the entity is monitoring for unauthorized updates (see Section ) and (2) the entity has had adequate opportunity to revoke malicious updates according to the DID method's access control mechanism (section ). This capability is further supported if timestamps are included (sections and ) and the target DLT system supports timestamps.

One mitigation against unauthorized changes to a DID Document is monitoring and actively notifying the entity when there are changes. This is analogous to helping prevent account takeover on conventional username/password accounts by sending password reset notifications to the email addresses on file. In the case of a DID, where there is no intermediary registrar or account provider to generate the notification, the following approaches are RECOMMENDED:

  1. Subscriptions. If the ledger or network on which the DID is registered directly supports change notifications, this service can be offered to DID owners. Notifications may be sent directly to the relevant service endpoints listed in an existing DID.
  2. Self-monitoring. An entity may employ its own local or online agent to periodically monitor for changes to a DID Document.
  3. Third-party monitoring. An entity may rely on a third party monitoring service, however this introduces another vector of attack.

In a decentralized identifier architecture, there are no centralized authorities to enforce key or signature expiration policies. Therefore DID resolvers and other client applications SHOULD validate that keys have not expired. Since some use cases may have legitimate reasons why already-expired keys can be extended, a key expiration SHOULD NOT prevent any further use of the key, and implementations SHOULD attempt to update its status upon encountering it in a signature.

Section specifies the DID operations that must be supported by a DID method specification, including revocation of a DID Document by replacing it with an updated DID Document. In general, checking for key revocation on DLT-based methods is expected to be handled in a manner similar to checking the balance of a cryptocurrency account on a distributed ledger: if the balance is empty, the entire DID is revoked. DID method specifications SHOULD enable support for a quorum of trusted parties to enable key recovery. Some of the facilities to do so are suggested in section 6.5, Authorization. Note that not all DID method specifications will recognize control from DIDs registered using other DID methods and they MAY restrict third-party control to DIDs that use the same method. Access control and key recovery in a DID method specification MAY also include a time lock feature to protect against key compromise by maintaining a second track of control for recovery. Further specification of this type of control is a matter for future work (see section ).

It is critically important to apply the principles of Privacy by Design to all aspects of decentralized identifier architecture, because DIDs and DID Documents are—by design—administered directly by their owners. There is no registrar, hosting company, or other intermediate service provider to recommend or apply additional privacy safeguards. The authors of this specification have applied all seven Privacy by Design principles throughout its development. For example, privacy in this specification is preventative not remedial, and privacy is an embedded default. Furthermore, decentralized identifier architecture by itself embodies principle #7, "Respect for user privacy—keep it user-centric." This section lists additional privacy considerations that implementers, delegates, and entities should bear in mind.

  1. DID method specifications MUST include their own Privacy Considerations sections, if only to point to the general privacy considerations in this section.
  2. The DID method privacy section MUST discuss any subsection of section 5 of [[RFC6973]] that could apply in a method-specific manner. The subsections to consider are: surveillance, stored data compromise, unsolicited traffic, misattribution, correlation, identification, secondary use, disclosure, exclusion.

If a DID method specification is written for a public ledger or network where all DIDs and DID Documents will be publicly available, it is STRONGLY RECOMMENDED that DID Documents contain no PII. All PII should be kept off-ledger behind service endpoints under the control of the entity. With this privacy architecture, PII may be exchanged on a private, peer-to-peer basis using communications channels identified and secured by key descriptions in DID records. This also enables entities and relying parties to implement the GDPR right to be forgotten, as no PII will be written to an immutable ledger.

Like any type of globally unique identifier, DIDs may be used for correlation. Identity owners can mitigate this privacy risk by using pairwise unique DIDs, i.e., by sharing a different private DID for every relationship. In effect, each DID acts as a pseudonym. A pseudonymous DID need only be shared with more than one party when the entity explicitly authorizes correlation between those parties. If pseudonymous DIDs are the default, then the only need for a public DID—a DID published openly or shared with a large number of parties—is when the entity explicitly desires public identification.

The anti-correlation protections of pseudonymous DIDs are easily defeated if the data in the corresponding DID Documents can be correlated. For example, using same public key descriptions or bespoke service endpoints in multiple DID Documents can provide as much correlation information as using the same DID. Therefore the DID Document for a pseudonymous DID SHOULD also use pairwise-unique public keys and pairwise-unique service endpoints.

When an entity is indistinguishable from others in the herd, privacy is available. When the act of engaging privately with another party is by itself a recognizable flag, privacy is greatly diminished. DIDs and DID methods SHOULD work to improve herd privacy, particularly for those who legitimately need it most. Choose technologies and human interfaces that default to preserving anonymity and pseudonymity. In order to reduce digital fingerprints, share common settings across client implementations, keep negotiated options to a minimum on wire protocols, use encrypted transport layers, and pad messages to standard lengths.

The current specification does not take a position on maximum length of a DID. The maximum interoperable URL length is currently about 2K characters. QR codes can handle about 4K characters. Clients using DIDs will be responsible for storing many DIDs, and some methods would be able to externalize some of their costs onto clients by relying on more complicated signature schemes or by adding state into DIDs intended for temporary use. A future version of this specification should set reasonable limits on DID character length to minimize externalities.

Including an equivalence property, such as equivID, in DID Documents whose value is an array of DIDs would allow entities to assert two or more DIDs that represent the same entity. This capability has numerous uses, including supporting migration between ledgers and providing forward compatibility of existing DIDs to future DLTs. In theory, equivalent DIDs should have the same identifier rights, allowing verifiable claims made against one DID to apply to equivalent DIDs. Equivalence was not included in the current specification due to the complexity of verifying equivalence across different DLTs and different DID methods, and also of aggregating properties of equivalent DID Documents. However equivalence should be supported in a future version of this specification.

Verifiable timestamps have significant utility for identifier records. This is a good fit for DLTs, since most offer some type of timestamp mechanism. Despite some transactional cost, they are the most censorship-resistant transaction ordering systems in the world, so they are nearly ideal for DID Document timestamping. In some cases a DLT's immediate timing is approximate, however their sense of "median time past" (see Bitcoin BIP 113) can be precisely defined. A generic DID Document timestamping mechanism could would work across all DLTs and might operate via a mechanism including either individual transactions or transaction batches. The generic mechanism was deemed out of scope for this version, although it may be included in a future version of this specification.

Section mentions one possible clever use of time locks to recover control of a DID after a key compromise. The technique relies on an ability to override the most recent update to a DID Document with Authorization applied by an earlier version of the DID Document in order to defeat the attacker. This protection depends on adding a time lock (see Bitcoin BIP 65) to protect part of the transaction chain, enabling a Authorization block to be used to recover control. We plan to add support for time locks in a future version of this specification.

Not all DLTs can support the Authorization logic in section 6.5. Therefore, in this version of the specification, all Authorization logic must be delegated to DID method specifications. A potential future solution is a Smart Signature specification that specifies the code any conformant DLT may implement to process signature control logic.

Although DIDs and DID Documents form a foundation for decentralized identity, they are only the first step in describing an entity. The rest of the descriptive power comes through collecting and selectively using verifiable claims. Future versions of the specification will describe in more detail how DIDs and DID Document can be integrated with—and help enable—the verifiable claims ecosystem.

This version of the specification relies on JSON-LD and the RDF graph model for expressing a DID Document. Future versions of this specification MAY specify other semantic graph formats for a DID Document, such as JXD (JSON XDI Data), a serialization format for the XDI graph model as defined by the OASIS XDI Core 1.0 specification.

A future-facing real-world context is provided below:

{ "@context": "", "id": "did:example:123456789abcdefghi", "publicKey": [{ "id": "did:example:123456789abcdefghi#keys-1", "type": "RsaVerificationKey2018", "owner": "did:example:123456789abcdefghi", "publicKeyPem": "-----BEGIN PUBLIC KEY...END PUBLIC KEY-----\r\n" }, { "id": "did:example:123456789abcdefghi#keys-2", "type": "Ed25519VerificationKey2018", "owner": "did:example:123456789abcdefghi", "publicKeyBase58": "H3C2AVvLMv6gmMNam3uVAjZpfkcJCwDwnZn6z3wXmqPV" }, { "id": "did:example:123456789abcdefghi#keys-3", "type": "RsaPublicKeyExchangeKey2018", "owner": "did:example:123456789abcdefghi", "publicKeyPem": "-----BEGIN PUBLIC KEY...END PUBLIC KEY-----\r\n" }], "authentication": [{ // this mechanism can be used to authenticate as DID ...fghi "type": "RsaSignatureAuthentication2018", "publicKey": "did:example:123456789abcdefghi#keys-1" }, { // this mechanism can be used to biometrically authenticate as DID ...fghi "type": "ieee2410Authentication2018", "publicKey": "did:example:123456789abcdefghi#keys-2" }], "service": [{ "type": "OpenIdConnectVersion1.0Service", "serviceEndpoint": "" }, { "type": "CredentialRepositoryService", "serviceEndpoint": "" }, { "type": "XdiService", "serviceEndpoint": "" }, { "type": "HubService", "serviceEndpoint": "" }, { "type": "MessagingService", "serviceEndpoint": "" }, { "type": "SocialWebInboxService", "serviceEndpoint": "", "description": "My public social inbox", "spamCost": { "amount": "0.50", "currency": "USD" } }, { "type": "DidAuthPushModeVersion1", "serviceEndpoint": "" }, { "id": "did:example:123456789abcdefghi;bops", "type": "BopsService", "serviceEndpoint": "" }]