Skip to content

Latest commit

 

History

History
366 lines (301 loc) · 17.3 KB

explainer.md

File metadata and controls

366 lines (301 loc) · 17.3 KB

Verifiable Credentials Data-Model Explainer

by Tzviya Siegman, Wiley; Manu Sporny, Digital Bazaar; Ken Ebert, Sovrin; Brent Zundel, Evernym

NOTE: "Verifiable claims" are now known as "verifiable credentials". The W3C Verifiable Claims Working Group's experience with using the term "verifiable claims" demonstrated that it led to confusion. The group has since found consensus in shifting to use the term "verifiable credentials", which contain "claims".

Introduction

It is currently difficult to transmit credentials such as driver's licenses, proofs of age, education qualifications, and healthcare data, on the Internet in a way that is verifiable yet protects individual privacy.

Starting in 2013, the W3C Credentials Community Group started work in earnest on solutions in this space followed shortly thereafter by the Rebooting Web of Trust Community and the Verifiable Claims Working Group. These groups, composed of 150+ individuals and organizations, are currently focused on the creation, storage, transmission, and verification of digital credentials.

Goals

The mission of the Verifiable Claims Working Group (VCWG) is to make expressing, exchanging, and verifying claims easier and more secure. The data model outlines core concepts, such as claims, credentials, and presentations, that form the foundation of the Specification.

Non-goals

The VCWG is not defining protocols or APIs. The model is identifier agnostic.

Getting Started

The verifiable credentials ecosystem is composed of five primary roles:

  • issuer - A role an entity might perform by creating a verifiable credential, associating it with a specific subject, and transmitting it to a holder. Example issuers include corporations, non-profit organizations, trade associations, governments, and individuals.

  • subject - A role an entity might perform by having one or more verifiable credentials asserted about it. Example subjects include human beings, animals, and things.

  • holder - A role an entity might perform by possessing one or more verifiable credentials about a subject and generating presentations from them. Example holders include students, employees, and customers.

  • verifier - A role an entity might perform by requesting and receiving a verifiable presentation that proves the holder possesses the required verifiable credentials. Example verifiers include employers, security personnel, and websites.

  • verifiable data registry - A role a system might perform by mediating the creation and verification of identifiers, keys, and other relevant data, such as verifiable credential schemas and revocation registries, which might be required to use verifiable credentials. Some configurations might require correlatable identifiers for subjects. Example verifiable data registries include trusted databases, decentralized databases, government ID databases, and distributed ledgers.

Note: In many cases the holder and subject of a verifiable credential will be the same, but this is not always the case. For example, a parent (the holder) may hold a verifiable credential for a child (the subject), or a pet lover (the holder) may hold a verifiable credential for their pet (the subject).

A visual depiction of the verifiable credentials ecosystem is shown below.

Claims, Credentials, and Presentations

The ecosystem roles exchange data enabling the realization of the use cases outlined below. The data exchanged differs depending on the roles participating, but is fundamentally composed of claims, credentials, and presentations.

A claim is a statement about a subject, expressed as a subject-property-value relationship.

Proofs

The cryptographic mechanism used to prove that the information in a verifiable credential or a verifiable presentation was not tampered with is called a proof. There are many types of cryptographic proofs including, but not limited to, digital signatures, zero-knowledge proofs, proofs of work, and proofs of stake.

The data model does not detail proof mechanisms.

Zero-Knowledge Proofs

The verifiable credentials data model supports the use of zero-knowledge proof (ZKP) technology. This allows credentials with ZKP-compatible elements to support presentations that enable:

  • Selective disclosure of each credential attribute.
  • Predicate proofs of numeric values (for example, integers, dates, and enumerations) that are:
    • Greater-than
    • Less-than
    • In a range (for example, 5 < x < 100).
  • Set-membership proofs.

Use Cases

The VCWG has created a Use Case document, demonstrating complicated scenarios and full encoding.

Some simple use cases include:

  • A student presents a government-issued identity to verify who she is when presenting herself for a standardized test.
  • An airline offers loyal customers upgrades to first-class using digital coupons. The coupons are issued as verifiable credentials.
  • A Board of Physicians maintains its list of board-certified physicians in a credentials repository, enabling the board to assert that a physician is certified, or to revoke certification as needed. The information in the repository can be verified as physicians apply for positions, or by patients as they seek information about the doctors they are considering. This can be done as a zero-knowledge proof of set membership.
  • A loan applicant presents proof of sufficient income, derived from a credential issued by her employer. This can be done without revealing her exact income by using a zero-knowledge predicate proof.

Code Samples

What does a Verifiable Credential look like?

There are many pieces of information that MAY be included in a verifiable credential, of which there are numerous examples in the Specification.

The following is an example of the ID property in a credential, using the Decentralized Identifier (DID) scheme.

{
  "@context": [
    "https://www.w3.org/2018/credentials/v1",
    "https://example.com/examples/v1"
  ],
  "id": "http://example.edu/credentials/1872",
  "type": ["VerifiableCredential", "AlumniCredential"],
  "issuer": "https://example.edu/issuers/565049",,
  "issuanceDate": "2010-01-01T19:73:24Z",
  "credentialSubject": {
    "id": "did:example:ebfeb1f712ebc6f1c276e12ec21",
    "alumniOf": "Example University"
  },
  "proof": {
    "type": "RsaSignature2018",
    "created": "2017-06-18T21:19:10Z",
    "verificationMethod": "https://example.com/jdoe/keys/1",
    "jws": "BavEll0/I1zpYw8XNi1bgVg/sCneO4Jugez8RwDg/+
      MCRVpjOboDoe4SxxKjkCOvKiCHGDvc4krqi6Z1n0UfqzxGfmatCuFibcC1wps
      PRdW+gGsutPTLzvueMWmFhwYmfIFpbBu95t501+rSLHIEuujM/+PXr9Cky6Ed
      +W3JT24="
  }
}

The following is an example verifiable credential that supports ZKPs.

{
  "@context": [
    "https://www.w3.org/2018/credentials/v1",
    "ctx:sov:anoncred:v1",
    "ctx:sov:GppHbMLLeKNYRhcQiXh3GjP2Yh",
  ],
  "type": [
    "VerifiableCredential",
    "AnonCred",
    "ExampleNameDOB"
  ],
  "issuer": "did:sov:4t1FPo72LzDMwpqtTVGVjysD6GUqS",
  "issuanceDate": "2018-11-27T12:37:15Z",
  "credentialSubject": {
    "name": "John Doe",
    "birthDate": "1969-02-14",
  },
  "proof": {
    "m_2": "0x375A05242CF33E9AE2E527DB6D6D5A2FA78A3042EF25D21013F82D5C642E98FA",
    "attributes": [
      "0x6998340478030A68F6DB6A3D4CB94304C4C60576E0992CBE81C4D32764876AD4",
      "0x5BFD3A7B",
      "0x3C414697B170EB2691A2AE126DD07F936005C478FFAEDB3B67255DA017C64A4B",
      "0x629E",
    ],
    "cred_def": "cdf:sov:Q6kuSqnxE57waPFs2xAs7q:3:CL:12:CDL1",
    "signature": {
      "A": "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
      "e": "0x1000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000006887BBE4F8F22F6FF0EE2863E42615"
      "v": "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
    },
    "signature_correctness_proof": {
      "se": "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
      "c": "0xD074B91DC755D75F56BA00F0B650190CF7035BFBE65678A290A511DFEF554521"
    }
  }
}

Here is an example of a verifiable credential encoded as a JWT that supports JWS proofs. The example shows the header and the payload as well as the final JWS compact serialization (base64 encoded).

{
    "alg": "RS256",
    "typ": "JWT",
    "kid": "did:example:abfe13f712120431c276e12ecab#keys-1"
}

{
  "sub": "did:example:ebfeb1f712ebc6f1c276e12ec21",
  "jti": "http://example.edu/credentials/3732",
  "iss": "did:example:abfe13f712120431c276e12ecab",
  "iat": "1541493724",
  "exp": "1573029723",
  "nonce": "660!6345FSer",
  "vc": {
    "@context": [
      "https://www.w3.org/2018/credentials/v1",
      "https://example.com/examples/v1"
    ],
    "type": ["VerifiableCredential", "UniversityDegreeCredential"],
    "credentialSubject": {
      "degree": {
        "type": "BachelorDegree",
        "name": "Bachelor of Science in Mechanical Engineering"
      }
    }
  }
}

eyJhbGciOiJSUzI1NiIsInR5cCI6IkpXVCIsImtpZCI6ImRpZDpleGFtcGxlOmFiZmUxM2Y3MTIxMjA0
MzFjMjc2ZTEyZWNhYiNrZXlzLTEifQ.eyJzdWIiOiJkaWQ6ZXhhbXBsZTplYmZlYjFmNzEyZWJjNmYxY
zI3NmUxMmVjMjEiLCJqdGkiOiJodHRwOi8vZXhhbXBsZS5lZHUvY3JlZGVudGlhbHMvMzczMiIsImlzc
yI6ImRpZDpleGFtcGxlOmFiZmUxM2Y3MTIxMjA0MzFjMjc2ZTEyZWNhYiIsImlhdCI6IjE1NDE0OTM3M
jQiLCJleHAiOiIxNTczMDI5NzIzIiwibm9uY2UiOiI2NjAhNjM0NUZTZXIiLCJ2YyI6eyJAY29udGV4d
CI6WyJodHRwczovL3czLm9yZy8yMDE4L2NyZWRlbnRpYWxzL3YxIiwiaHR0cHM6Ly9leGFtcGxlLmNvb
S9leGFtcGxlcy92MSJdLCJ0eXBlIjpbIlZlcmlmaWFibGVDcmVkZW50aWFsIiwiVW5pdmVyc2l0eURlZ
3JlZUNyZWRlbnRpYWwiXSwiY3JlZGVudGlhbFN1YmplY3QiOnsiZGVncmVlIjp7InR5cGUiOiJCYWNoZ
WxvckRlZ3JlZSIsIm5hbWUiOiJCYWNoZWxvciBvZiBTY2llbmNlIGluIE1lY2hhbmljYWwgRW5naW5lZ
XJpbmcifX19fQ.VTotLRYblDOtJBTlOYbvibqC_uu8RXdvv6m_lR6cdEdFcGf4oNKiFZ_WJr07n1A-_E
jTLzjD5XwmDPzb8lxDlEkCLJ5WQS4_jwzCZAeetNG7YO2slTCFRiE_2xBM2R01ssI8bEPnqLBnc-Lu88
DmO21wL8-Yud1eL45N_pEE5DqTF5DJ-IesmVkWvC8159GUHKShFIpgHiE1EDDEjUzjYA5ZyzS2_ZmSTj
5NDLOt8muVlORpO7xJ6aWRdtibkmSTKykUzh75-4Rklz2H9-AWBXEh5ajkyB8yINh_y4jK3g7ypVACRI
Z9DZhdw2K39KCilAPVJsmejiKlxNhQAOlgcYUlhCzphLsqo-FA90fFGrsg-3JuQihnNw6RSPImjVt_yV
appfjilEzhyfWT-Smm_KN8LRbFdNtU-awwhKbjDNW-7fNVrnsWHKvLsd_zlch8YlHZ6g0tHJnxo_yOTM
BSSpt0jzyl1ByqjumgBFNpTR-NTVog4B7vLEvq58RuShraL5VNr7bjNzq2gisp3jq3LpfUmiwc7rQXw6
AlQuattLRolXx3EtPysrZe-wU7yrEtNPvpGs-OyJAczfJPzza9lGTbx6IWS-0pTmNq6hwNd0ODMiB3uL
3TeBN1xLoue9Hdc3toUvmdyXecSvltPcaiRoN-uQo8RRAvfK7GALAzaHw

Important Design Choices

This section summarizes a number of design choices that the VCWG spent a considerable amount of time debating.

Privacy-Enhancing Architecture Agility

The VCWG spent a considerable amount of time exploring various privacy concerns including, but not limited to, multiple ZKP systems, selective disclosure schemes, avoiding cryptographic fingerprinting, and the intersection of tracking technologies and verifiable credentials. The outcome is a specification that attempts to strike the right balance between what is currently achievable while ensuring that future privacy-enhancing systems are able to use the same architecture to protect individual and organizational privacy.

Syntax Agility

While the document outlines a data model that is expressible in a number of different syntaxes, there was considerable thought put into the question of which representation syntaxes the specification would highlight. JSON-LD, which is a fully compatible subset of JSON, was suggested as the primary format. There were concerns that a JSON-LD processor would be required to process verifiable credentials. After more than a year of technical work and collaboration with the JSON-LD 1.1 Working Group, the VCWG has demonstrated that a JSON-LD processor is not necessary to conform to the Specification. The vast majority of developers will be able to work with a verifiable credential as if it is just another JSON object.

Authorization

The VCWG debated the use of verifiable credentials as an authorization mechanism. The result of this debate was the notion that verifiable credentials could be used in authorization systems, but did not constitute an authorization system by themselves, and that great consideration should be put into any system that uses verifiable credentials in an authorization framework.

Terms of Use

There were multiple discussions related to how a verifiable credential should be used. For example, issuers might want to place restrictions on how a verifiable credential should be used. Holders might want to also place restrictions on how a verifier can use their information. While there was consensus that these restrictions should be able to be expressed in the data model, there was no consensus on what sorts of restrictions should be suggested by the Specification. The result is an open mechanism for expressing the terms of use for verifiable credentials and verifiable presentations.

Proof Format Agility

Currently there are at least four different proof formats in active use by participants in the VCWG, including JSON Web Signatures, Zero-Knowledge Proofs using CL signatures, Common Binary Object Representation Object Signing and Encryption, and Linked Data Proofs. It was challenging for the VCWG to select one format because there are a number of benefits and drawbacks provided by each approach. In addressing this challenge, the VCWG has ensured that the Verifiable Credentials Data Model is flexible enough to be compatible with each signature format with no changes needed to the core data model.

Support for JWT

JSON Web Token (JWT) is a widely used means of expressing claims between two parties. Providing a representation of the Verifiable Credentials Data Model for JWT allows existing systems and libraries to participate in the ecosystem.

The Specification defines the encoding rules of the Verifiable Credential Data Model onto JWT and JWS. It further defines processing rules on how and when to make use of specific JWT-registered claim names and specific JWS-registered header parameter names, allowing systems based on JWT to comply with the specification while avoiding duplicate representation of specific information contained in the enclosed JSON or JSON-LD object.

Selective disclosure

The ability of a holder to selectively disclose the attributes in a credential is recognized as a valuable property, in harmony with the principles of data minimization and self-sovereign identity.

It was debated whether this property should be required of all verifiable credential implementors, but it was determined that requiring such capability might be too onerous. The data model in its current form supports selective disclosure as a best practice, but does not require it.

Features at Risk

We have asked for preliminary commitments from potential implementers. We expect all features to have at least two implementations. With current commitments, some features have exactly two.
Of course we will continue to recruit for additional implementations.

Implementations and Reviews

The VCWG has begun the process of horizontal review. Feedback from APA and PING has been, or is in the process of being, incorporated.

This data model has numerous successful implementations, including governments, universities, major technology organizations, and not-for-profit organizations.

References and Acknowledgements

Most of the text came from the data model. Thanks to Manu Sporny for writing much of the source material, Tzviya Siegman for beginning this document, and Oliver Terbu and Ken Ebert for adding examples of verifiable credentials.