IIW32: BBS+ and beyond

Nader Helmy • May 6, 2021 • 4 min read

The Internet Identity Workshop continues to be a central nucleus for thoughtful discussion and development of all things related to digital identity. The most recent workshop, which was held in mid-April, was no exception. Despite the lack of in-person interaction due to the ongoing global pandemic, this IIW was as lively as ever, bringing together a diverse set of stakeholders from across the globe to share experiences, swap perspectives, and engage in healthy debates.

One common theme this year was the continued development and adoption of BBS+ signatures, a type of multi-message cryptographic digital signature that enables selective disclosure of verifiable credentials. We first introduced this technology at IIW30 in April 2020, and have been inspired and delighted by the community’s embrace and contribution to this effort across the board. In the year since, progress has been made in a variety of areas, from standards-level support to independent implementations and advanced feature support.

We thought we’d take a moment to round up some of the significant developments surrounding BBS+ signatures and highlight a few of the top items to pay attention to going forward.

Over the past few months, the linked data proofs reference implementation of BBS+ published a new release that introduces a variety of improvements in efficiency and security, including formal alignment to the W3C CCG Security Vocab v3 definitions. In addition, support for JSON-LD BBS+ signatures was added to the VC HTTP API, making it possible to test this functionality in an interoperable way with other vendors participating in an open environment.

An important element in enabling BBS+ signatures is using what’s known as a pairing-friendly curve; for our purposes we use BLS12–381. We have seen some promising signs of adoption for this key pair, with multiple Decentralized Identifier (DID) methods — both did:indy from Hyperledger and did:ion from DIF — indicating they intend to add or already have support for these keys, allowing BBS+ signatures to be issued across a variety of decentralised networks and ecosystems. This development is possible due to the fact that BBS+ signatures is a ledger-independent approach to selective disclosure, effectively no custom logic or bespoke infrastructure is needed for these digital signatures to be created, used and understood.

In addition, the Hyperledger Aries project has been hard at work developing interoperable and ledger-agnostic capabilities in open source. The method used to track interop targets within the cohort and ultimately measure conformance against Aries standards is what’s known as an Aries Interop Profile (AIP). A major upcoming update to AIP will add support for additional DID methods, key types and credential formats, as well as introducing Aries support for JSON-LD BBS+ signatures as part of AIP 2.0. This will allow Aries-driven credential issuance and presentation protocols to work natively with BBS+ credentials, making that functionality broadly available for those in the Aries community and beyond.

There have also been a number of exciting developments when it comes to independent implementations of BBS+ signatures. Animo Solutions has recently implemented JSON-LD BBS+ signatures support into the popular open-source codebase Hyperledger Aries Cloud Agent Python (ACA-Py). In another independent effort, Trinsic has contributed an implementation of JSON-LD BBS+ credentials which they have demonstrated to be working in tandem with DIDComm v2, a secure messaging protocol based on DIDs. Implementations such as these help to demonstrate that open standards are transparent, can be understood and verified independently, and can be implemented with separate languages and tech stacks. They also set the groundwork for demonstrating real testing-driven interoperability via mechanisms such as the VC HTTP API and AIP 2.0. We are continuously looking to improve the documentation of these specs and standards so that their implications and nuances can be more broadly understood by builders and developers looking to engage with the technology.

On the cryptographic side of things, progress is also being made in hardening the existing BBS+ specification as well as expanding BBS+ to support more advanced privacy-preserving features. A significant development in this area is the work of cryptographer Michael Lodder who has been actively conducting research on an enhanced credential revocation mechanism using cryptographic accumulators with BBS+. This approach presents a promising alternative to existing solutions that allow authoritative issuers to update the status of issued credentials without compromising the privacy of the credential holder or subject who may be presenting the credential. We see this as another application of BBS+ signatures in the context of verifiable credentials that carries a lot of potential in pushing this technology to an even more robust state.

There was also initial discussion and tacit agreement to create a new cryptography-focused working group at Decentralized Identity Foundation. As the new WG drafts its charter, the first work item of this group will be the BBS+ Signatures spec which defines the cryptographic scheme known as BBS+ agnostic of its application in areas such as linked data signatures or verifiable credentials. In the future, this WG will likely evolve to include other crypto-related work items from the community.

This is just the tip of the iceberg when it comes to the momentum and development building around this technology in the community. We couldn’t be more excited about the future of BBS+ signatures, especially as we gear up to tackle the next set of hard problems in this area including privacy-preserving subject authentication and revocation using cryptographic accumulators. If you’re interested we encourage you to get involved, either by contributing to the Linked Data Proofs specification, checking out our reference implementations, or participating in the new WG at DIF, to name but a few of the many ways to engage with this work. We look forward to doing this retrospective at many IIWs to come, documenting the ever-growing community that continues to champion this technology in dynamic and interesting ways.

This blog was originally posted on Medium.

Get in touch with us

Our technologies give you powerful ways to build trust and prove things about people. If you’ve got a use case and want to see what it might look like, we’d love to talk to you about it.