Governance frameworks in the IAM stack
Governance provides the rules and procedures to establish behavior, expectations and trust within an environment. While governance is a critical component of any multi-party network, it is especially critical in decentralized environments, where there is no central authority to define and orchestrate governance mechanisms over every component of the ecosystem.
As an example, consider an application built on top of the Energy Web Chain. Each application must ensure that:
Components are in compliance with existing digital frameworks that their application depends on (e.g. , peer-to-peer protocols or )
The application has a governance framework that is robust enough to garner stakeholder trust and compliant participation within the application itself (i.e. defining and enforcing who is allowed to do what within the application)
Governance in a network is established through a governance framework (also referred to as a trust framework). The framework provides concrete policies, rules and expectations for the stakeholders within the network.
Energy Web’s IAM governance relies on two systems: and . Used together, these components provide a governance framework for users to interact with the digital infrastructure, and with other users in a secure and self-sovereign manner.
Role credentials are associated with a user’s , which is anchored on the Energy Web Chain in the . This means that a user’s roles and credentials are not siloed within any one application; because a user can use their DID to register with any application built on top of the Energy Web Chain, their roles and credentials are portable.
In the Energy Web IAM ecoystem, role-based hierarchies are defined by organizations, applications, and designated roles within them. The tech stack leverages to define and namespace relational hierarchies within a system. We decided to deploy our own copy of ENS on the Energy Web Chain as it provides a standard set of widely-used, well-tested smart contracts. Read more about the ENS smart contracts deployed on the Energy Web Chain .
The namespace hierarchy is built on four levels:
Organization: a top-level organizing body
Sub-organization(s)
Application: a distinct service or functionality provided by an organization or sub-organization
Role: a distinct functionality within an application or within an organization
When roles are created within an organization or an application, the creator can define conditions or criteria that restrict who is qualified to take on the role. The role creator can also determine which users (by DID or role) are authorized to issue or revoke a role.
Below is a resolved role definition for a role of "install lead". Note that it contains an enrollment precondition that the subject already has the role (credential) of 'project installer'. The role definition also specifies an expiration date, and asserts that only users that have the role of 'install manager' can issue or revoke this role.
Verifiable Credentials enable users and their assets to take on (that is, within an organization, a sub-organization or an application within a hierarchy, as discussed ).
See more extensive documentation on credentials in the IAM stack .
provides the user interface for creating and defining these hierarchies. See the Switchboard guide on Governance and role creation .
The supporting provide the functionality for persisting and resolving namespaced domains Namespace domains that are registered and managed in the . Read more about the role of Ethereum Name Space in Energy Web Digital Infrastructure .
These libraries also support governance by providing credential verification mechanisms. This is discussed further in the .
{
roleName: "installlead",
defaultValidityPeriod: 31536000000,
enrolmentPreconditions: [{
conditions: ['projectinstaller.roles.testdidproject.apps.suborgs.whitney.iam.ewc'],
type: role
}],
issuer: {
issuerType: "ROLE",
rolename: "installmanager.roles.suborgs.whitney.iam.ewc"
},
revoker: {
issuerType: "ROLE",
rolename: "installmanager.roles.suborgs.whitney.iam.ewc"
},
roleType: "org",
version: 1,
issuerFields: [],
requestorFields: []
}In the context of EW-DOS, an Asset is a digital representation of a physical or virtual device on the Energy Web Chain. An Asset could represent, for example, a solar photovoltaic panel, a battery, an electric vehicle, or an IOT device.
Assets must have a Decentralized Identifier - DID in the Energy Web Chain's DID Registry in order to participate in applications and marketplace activities. Once an Asset has a DID, it can take on roles within an organization or application. This is discussed further below.
Assets and their chain of custody are managed by that are deployed on the Energy Web Chain.
When an Asset is first created, it is registered in the as an owned identity (the owner address being the Asset owner, discussed ).
The main purpose of the IdentityManager smart contract is to have a on-chain location which aggregates known assets. In other words, it provides a kind of "asset-registry" functionality. This can allow one for instance, to more easily answer questions such as "how many assets have been registered in total?".
While there are many OfferableIdentity smart contracts, there is intended to be fewer IdentityManager smart contracts as the IdentityManager's main purpose is the aggregation of OfferableIdentity information. For instance, an enterprise could have a single IdentityManager to all assets which are registered by their employees.
The Asset owner can offer ownership to another DID. The provides methods to verify, offer and transfer (these concepts are discussed in further detail ).
Every Asset must have an owner. Asset owners initiate the registration, transference and enrolment activity of their Assets. This requires them to make transactions on behalf of their Asset, so the owner must have an address on the Energy Web Chain that is connected to an . The owner of an Asset is recorded in the Asset's identity on-chain.
The provides high-level methods to facilitate the chain of custody for an Asset.
Chain-of-custody events (, and ) for an Asset are emitted from the . listens for and persists the details of these events. All historical owners of an Asset and the dates of offer, transference and acceptance are accessible through SSI Hub's API.
Registering as Asset involves creating an OfferableIdentity smart contract for the Asset on the Energy Web Chain, and registering its identity in the . This is initiated by the Asset owner. Because each Energy Web Chain address is a valid DID under the did:ethr DID method, each asset inherently has a DID.
The owner of a registered Asset can transfer ownership to another address on the Energy Web Chain. (The new owner's address must have signing capabilities in order to associated with asset management).
The provides methods to facilitate Asset transferance. This contract makes calls to the so that the state of the Asset is updated at each phase of the transfer.
When a transfer is initialized, an 'offer' of transfer is made to the recipient DID.
The state of the Asset identity is marked as 'offered' in the .
The DID that the Asset was offered to must accept the Offer before it is transferred to them. provides a method to accept the transfer. The Asset's owner is updated in the to reflect the new ownership.
The DID that the Asset was offered to has the option to reject the transfer. The Asset's 'offered' status in the is set to 'false'.
An Asset can take on . If their enrolment request is approved by the role issuer, the Asset is issued a role-based .
If the Asset's owner is an authorized issuer of the desired role, the Asset owner can directly issue a role-based verifiable credential to their Asset.
If the Asset's owner is not an authorized issuer of the desired role, the owner can submit an enrolment request on behalf of their Asset to the issuer.
provides the high-level methods to request and issue enrolments. See the API documentation .
provides an interface for users to register, transfer and enroll Assets. If you are logged into Switchboard, you can view the Asset management interface .
contains the high-level functions for managing (registering, fetching, transferring, etc.) assets and their corresponding data. You can view the service API documentation for Assets in the library .
Energy Web and the (DSO) jointly developed a decentralized energy asset management system leveraging the EW-DOS components and architecture .
The goal of this collaboration was to:
Facilitate secure, encrypted communication between (i.e. solar panels, batteries, etc) and the grid
Enable to provide grid services (e.g. selling excess energy back to the grid)
Grid assets (e.g, smart meters, distribution automation devices), and were assigned DIDs. The DID is anchored on the asset's pre-existing SIM cards. Each asset exists as an identity in the on the Energy Web Chain. Cryptographically signed information (such as control signals and commands) from the DSO (Stedin) can then be sent to targeted assets. This allows for an awareness and exchange of grid services between the DSO and DERs.
You can read more about this use case in the official press release .
iam-client-libraryiam-client-libraryiam-client-libraryPossible credential metadata
In a verifiable credential ecosystem, there are various metadata that are useful for effectively using the data in the credentials. This page aims to describe possible metadata specifications and technologies that may be used with credential data.
Note: Not all of the following metadata is available for all credentials within the Energy Web credentials ecosystem.
Data semantics refers to the meaning of the data. This includes semantic disambugation (know precisely what a term is refering to) as well as data descriptions and relationships.
Linked Data provides semantic disbiguation by requiring that terms be IRIs.
maps terms to IRIs. This allows JSON-LD documents to be written in more concise, readable manner, without sacrificing accuracy.
The VC Implementation Guide also has for verifiable credentials.
Linked Data can also be used to provide further information about the semantics of a term. This is provided in a data vocabulary or ontology.
For example, the property can be used to provide further description of a resource.
Note that vocabularies can be used to infer data validation but this is non-standard and violates the open-world assumption of W3C ontology languages.
This point is made in
Although data validation is an important practical use case for the RDF stack, until SHACL came around, there was no W3C standard mechanism for defining data constraints. Over time, people became creative in working around this limitation. Many tools simply decided that for all practical purposes, the open-world and non-unique-name assumptions should simply be ignored. OWL-aware tools including TopBraid and Protégé, for example, provide data entry forms that restrict users from entering more than one value if there is a corresponding owl:maxCardinality 1 restriction, or require the selection of a specific instance of ex:Person if that class is the rdfs:range of the property. The makes a similar point SHACL shapes are not an ontological models. They serve a different purpose. Ontologies describe concepts and help in inferring additional knowledge. Firstly, the W3C ontology languages follow the Open World Assumption, secondly SHACL follows the Closed World Assumption. If a self-description is missing an attribute, it is an error in the shape validation (and that’s how it should be!). From the ontology’s point of view, the attribute could be defined somewhere else in the WWW, if not in the JSON-LD file at hand (but this extremely decentralised view is not compatible with Gaia-X’s trust-building approach).
Data structure validation is used to validate that data, for example, contains specific properties or that properties have specific values.
There does not seem to be a single way of describing data structures in the Verifiable Credentials ecosystem as different methods have different tradeoffs.
For verifiable credentials, the schema of a credential can optionally be linked using the .
JSON Schema can be used to describe the precise shape required by the a credential.
The JsonSchemaValidator2018 credentialSchema type is defined in the
SHACL is the W3C standard for validating RDF graphs.
SHACL is being .
To check whether the claims in a Self-Description follow all constraints, such as including all mandatory attributes, the claims are validated against a shape. Technically, these shapes follow the W3C Shapes Constraint Language (SHACL). The claims themselves are represented as an RDF graph, serialised in the W3C JSON-LD format, where JSON is a data interchange format widely supported by programming languages, and JSON-LD (LD = “linked data”) makes it compatible with RDF.
is a pre-alpha project that attempts to reconcile the trade-offs of JSON Schema and SHACL.
It notes:
JSON-LD documents can be seen from two points of view: as regular JSON documents following certain conventions or as RDF graphs encoded using JSON syntax. Validation mechanisms indeed exist for both ways of looking at the information in a JSON-LD document, but each of them has important drawbacks when working with JSON-LD:
JSON-Schema can be used to validate JSON-LD as plain JSON documents. However, the information in a JSON-LD document can be encoded through multiple, syntactically different, documents, so it is hard to write a generic JSON-Schema validation without going through a normalisation step, and even if the validation is written for a normalised JSON-LD document, the description of the validation becomes verbose and complex to write, relying, for example, on the usage fo fully expanded URIs. There is also the issue of the implications for the validity of the RDF information encoded in the document when we are just validating the JSON syntax.
SHACL can be used to validate the RDF graph encoded in the JSON-LD document. SHACL is powerful and expressive but difficult to write and learn, especially for users that do not want to work with JSON-LD as an RDF format. Additionally, the performance of SHACL validators is far from the performance of syntactical JSON-Schema validators
The specification describes how a credential can be displayed.
The credential lifecycle in the IAM stack
This page documents the role and lifecycle of credentials in the IAM stack.
Credentials are documents that allow individuals to show that they possess certain accreditations (this is discussed further in depth ). Credentials are traditionally document or paper-based, such as a driver's license or a passport, and are typically registered in a centralized repository under the stewardship of the issuing body.
Verifiable credentials are purely digital components that can be verified cryptographically. Verifiable credentials are a secure and self-sovereign alternative to traditional paper-based credentials or documents, which typically must be physically or electronically transmitted for verification, and have the potential to be intercepted, altered or tampered with.
In general, verifiable credentials do not rely on Decentralized Identifiers (DIDs) but they are often used together. If DIDs are used, decentralized ledger technology (i.e. a blockchain) can provide the public key infrastructure for the cryptographic verification.
The specification that defines verifiable credentials was established and is maintained by the . This protocol continues to evolve. Verifiable credentials are one of the three core pillars of , along with and distributed ledger technology.
See the W3Cs use cases and requirements for verifiable credentials here
Verifiable credentials are a key component of Energy Web's Identity, authorization and access management. In current use cases, credentials are used to authorize users' or assets' enrolment into applications and organizations. Each credential is associated with the subject's DID that is anchored on the Energy Web Chain. Energy Web's Switchboard application provides a user interface for creating, requesting, issuing and revoking role-based credentials. Energy Web's IAM libraries and APIs facilitate the full lifecycle of verifiable credentials. The Energy Web blockchain serves as the trust or verification layer for digital proof. This documentation provides references to these libraries, and to supporting W3C documentation that is used to guide and inform the development of Energy Web's IAM solution.
Click below to see a current diagram of Energy Web IAM architecture:
A claim is an assertion that is made about a subject. A claim could assert, for example, that a battery meets specific manufacturing standards.
A credential is a claim(s) made by an issuer about a subject. When the credential issuer makes a claim about a subject, they issue a verifiable credential. For example_,_ an issuer can issue a credential for a battery that asserts that it was manufactured on a specific date in a specific location.
In order for a credential to be verifiable, it must contain a proof mechanism and supporting information to evaluate the proof. The proof must be able to be verified through cryptographic (i.e. algorithmic) means, typically through a digital signature. Energy Web's IAM libraries support digital signatures of Ethereum-compatible wallets, such as MetaMask.
Proofs provide two primary functions:
To verify the authorship of a credential
To detect tampering or alteration to the credential/presentation
In the example of the battery mentioned directly above, the proof will verify that
The issuer is authorized to assert that the battery was manufactured on a specific date and in a specific location
That the credential data and digital signature has not been changed or compromised
Note that proof mechanisms do not validate the facts that the claim asserts (in the example above, the providence of the battery). It only verifies the issuer of the claim and the integrity of the claim's data over time (i.e. has the data been altered or tampered with).
There are two main proof mechanisms for verifying credentials, both of which are used in the IAM stack. These will be discussed further below.
External proofs, which are commonly expressed by JSON Web Tokens. An external proof is one that wraps an expression of the credential data model, such as a JSON Web Token. See the W3C JSON payload encoding standards for verifiable credentials here. Verifiable Credentials expressed as JSON Web Tokens in the IAM stack is discussed below.
Embedded proofs, in which the proof is included directly in the credential's JSON data. To be compatible with W3C Verifiable Credentials standard, the credential JSON must have a property of 'proof'. Verifiable credentials expressed as JSON/JSON Linked Data in the IAM stack are discussed below.
The Energy Web IAM stack provides API methods to request, issue, publish, verify and revoke credentials.
Our ecosystem offers persistence for credentials on the Energy Web blockchain, and off the blockchain using the decentralized file system IPFS.
Credentials that are persisted on the Energy Web Chain are referred to as 'on-chain' credentials
Credentials that are persisted off-chain are referred to as 'off-chain' credentials
The distinctions are discussed in the following section.
Note that whether a user chooses to persist credentials on-chain and/or off-chain, credential data is also persisted in the SSI Hub. SSI Hub facilitates credential exchange by persisting credential request and issuance messages
On-chain credentials are registered on the Energy Web blockchain in the Claim Manager smart contract. The contract source code can be found here. A credential is registered in the contract when the credential is published by the holder.
The Energy Web Chain is a public blockchain. This means that smart contracts and their data are public, and their public methods can be called by anyone. This is an important point of consideration when deciding whether to publish credentials to the blockchain.
There are currently two smart contracts deployed on the Energy Web Chain that are used for verifying and persisting verifiable credential data (for 'on-chain credentials' only). The source code for these contracts can be found in the @energyweb/onchain-claims package.
Claim Manager
Holds mapping of issued verifiable credentials; Provides methods for credential verification
Claims Revocation Registry
Holds mapping of
Off-Chain credentials are not referenced on the blockchain. Off-chain credential data is persisted as a JSON web token on IPFS, a decentralized public filesystem. IPFS, like a blockchain, is a decentralized system and relies on a network of peer nodes to create a distributed system.
When a token is stored on IPFS, it has a content identifier (CID) that points to its location on the file system. This CID is linked to the credential's corresponding DID Document through a service endpoint. A service endpoint points to any service that supports or acts on behalf of a DID. The CID is used to fetch and resolve the full credential from IPFS when necessary.
DID Document data, including service endpoints, are publicly available on the Energy Web Chain. IPFS data is accessible to anyone who has the service endpoint that contains its location on IPFS. This is an important consideration when deciding to publish credentials to IPFS.
For each off-chain credential request, iam-client-library issues two credentials of different format and proof type.
Both signature mechanisms can be performed by Ethereum-based wallets such as MetaMask.
A JSON web token is a common standard for transferring encoded, verifiable data between parties. The JSON web token payload is generated from the claim data.
Proof Mechanism: EIP-191 signature standard
Data Model: The JWT payload partially conforms to the properties established by W3C for verifiable credentials as a JWT. You can view these properties here.
Persistence: The issued token is included in an object that contains other data about the issued credential request. (Read more about claim issuance below). The token is persisted in the SSI Hub's database when the credential is issued, and persisted in IPFS when a credential is published by the subject. The credential's location in IPFS is referenced in the user's DID Document's service endpoints.
Proof Mechanism: EIP-712 signature specification. This allows the credential data to be displayed in a human-readable format when digitally signing the message:
Data Model/Interface: The Verifiable Presentation interface conforms to W3C credential schema, which includes JSON-LD (JSON Linked Data) data format. JSON linked data ensure that credentials and presentations are universally machine readable and compatible. Read more about JSON-LD formatting in verifiable presentations and credentials in the W3C documentation here.
Persistence: The verifiable presentation is included in an object that contains other data about the issued credential. This object is persisted in SSI Hub database.
In addition to the above, there are a few distinctions to consider when deciding to register credentials off-chain or on-chain.
On-chain credentials can be read by any smart contract deployed on the Energy Web Chain, and thus used in many decentralized applications. This allows for greater application interoperability.
Off-chain credentials in IPFS do not require any transaction costs that are associated with a blockchain, and can be resolved by an IPFS client in any application.
A credential request occurs when a user requests to enrol to an organization or application’s pre-defined role. The role may have specified criteria that the subject must meet in order to take on the role. The user who is requesting to take on the role is the claim ‘subject’.
iam-client-library contains high-level methods to initiate credential requests. These methods indicate whether a credential should be registered 'on-chain' or 'off-chain'.
Claim requests are persisted in the SSI Hub.
A user makes a request to the credential issuer to enrol into Organization A as a 'Battery Installer'. This role is pre-defined by Organization A, and has a designated set of issuers (each with a DID) who are authorized to approve and issue a credential for the 'Battery Installer' role. At the time of request, the subject must specify if they want this credential to be persisted on-chain or off-chain.
The credential's issuer approves and issues (or rejects) a subject’s credential request. By issuing the credential, they are verifying that the subjects meets the requirements needed to obtain the credential.
Using the example **** above, the issuer is issuing a credential of 'Battery Installer' to the subject. In doing so they are asserting that the subject meets the requirements to hold this credential. They are signing the credential (providing proof) with their digital signature using an Ethereum- compatible wallet such as MetaMask.
If a credential has an **'**on-chain' registration type, the verified credential is registered in the Claim Manager smart contract.
As discussed above, off-chain credentials are issued in two formats:
A verifiable presentation ****
A signed JSON Web token****
The verifiable presentation and issued token are appended to the claim data, which is updated in the SSI Hub.
Once an issuer has issued a credential, the subject has the option to publish (persist) the credential. The persistence location depend on if the credential is registered on-chain and/or off-chain. Persistence for on-chain credentials is discussed above here. Persistence for off-chain credentials is discussed above here.
The W3C Verifiable Credential protocol includes revocation as a standard operation in the credential lifecycle. Credentials can be revoked due to a breach in proof integrity (digital signature), or because the subject no longer meets the requirements to hold the credential.
The IAM stack provides methods for a revoker to revoke a credential after it has been issued.
Revocation for Energy Web credentials is executed based on the Role Governance. Only an authorised revoker (a DID or any DID with specific role credential) can revoke a credential.
When an on-chain credential is revoked, the revocation is registered in the ClaimsRevocationRegistry smart contract. This smart contract holds a reference to every revoked credential. The ClaimsRevocationRegistry contract references the ENSRegistry contracts to ensure that the revoker has the right authority to execute revocation and ClaimManager contract to validate if the subject has the role credential (to be revoked) and if revoker's authoritative credential has either expired or revoked (for the case where revoker's authority is based on a specific role credential).
For a revoker to be able to revoke an on-chain credential, their authoritative credential should also be published on-chain.
For a credential to be verifiable on-chain, a verifier should be able to check the revocation status of the issued credential along with the issuance.
It should be possible to get a single source of truth for a revocation status i.e. a verifier should be able to get a valid revocation from a registry without the need to verify it again.
For a revocation to be registered on-chain it is necessary that the credential first be registered on-chain with ClaimManager, where holder has provided their consent to make the credential publicly available.
The publishing of a holder's credential reduces the privacy and allows someone to determine if a holder has been issued a role credential or not.
In other words, revocation of a credential which has not been published yet would imply that the holder holds the role credential reducing the privacy of the holder's credential.
Therefore, the revocation registry requires that the holder consent to their credential being on-chain and publishing it to the ClaimManager registry.
Having the credential in the ClaimManager allows the ClaimsRevocationRegistry to provide verifiability of:
Only verified and valid credential being revoked, ClaimManager verifies credential's integrity and its issuer's authority before registration.
Credential's validity with regards to expiration.
Revoker's authority who revoked Holder's credential.
Thus establishing a design which provides Verifier with a trusted mechanism for revocation status check which ensures that a valid credential can only be revoked by an authoritative revoker.
When an off-chain credential is revoked, the credential JSON is updated with a "credentialStatus" property that is conformant to W3C's StatusList2021 data model. This data model provides a link for verifiers to use to see the status of a credential (i.e. if a credential has been revoked by the revoker). The verifier can verify this credentialStatus property and derive the revocation status of the credential. You can view the properties of the StatusList2021 data model here.
When a credential is revoked using iam-client-library, which utilises status-list module to append this credentialStatus property to credentials, the credentialStatus will have a statusPurpose that reflects it has been revoked:
Status list credentials are persisted in SSI Hub. The credentialStatus object has an attribute statusListCredential that provides a URL thorugh to fetch the credential's status for verification purposes.
More detailed documentation around StatusList2021 can be found here
A verifier is responsible for verifying the authenticity of a credential. The criteria for this evaluation is specified by the W3C Verification standards here. iam-client-library provides verification methods that evaluates these criteria, namely:
The credential proof (typically the digital signature) is valid.
The credential is not revoked.
The credential is not expired.
Beyond the scope of basic verification, iam-client-library's verification methods also validate the issuer's authorization to issue the credential and does this for all the issuers in the hierarchy.
This verification is executed for off-chain credentials and relies on the subject's (holder or issuer) DID Document resolution.
credentialStatus: {
id: 'https://credential-status/admin',
type: StatusListEntryType.Entry2021,
statusPurpose: CredentialStatusPurpose.REVOCATION,
statusListCredential:
'https://isc.energyweb.org/api/v1/status-list/700xxxxx-5xxx-4xxx-bxxx-43acfaxxxxxx',
statusListIndex: '0',
}
