What led you to start working with Decentralized Identity?
I studied computer science at university and wrote my dissertation on Bitcoin, as well as developing using Hyperledger Fabric, so I’ve been in the Web3 world for a while.
After uni I got involved in a research project looking into how the mortgage application process could be improved, as part of a knowledge transfer partnership between a university and a bank.
Applying for a mortgage is slow and painful, so streamlining it should make it better for customers and cheaper for the bank. The number of parties — including buyer and seller, lenders, conveyancers, estate agents, credit reference agencies and the Land Registry — made us think “This seems like a problem blockchain can help to solve”.
However when we spoke to the parties involved, it became clear the core challenge is a data problem. There’s a need to securely exchange data about the property and the participants. That’s not easy, as a lot of the information is private.
All the guidance was, “Don’t put personal information on the blockchain”. So we searched around and learned about SSI (Self Sovereign Identity) and Verifiable Credentials (VCs). We visited the Rebuilding the Web of Trust (RWOT) conference and got excited. We thought, “This seems like the right way to do it”.
At this point I realised I wanted to work in decentralized identity, so I moved to Berlin and worked at Jolocom.
Tell us about TrustBox, please
I’m not the only one who’s seen these problems with the property buying process. After I returned to the UK I got involved with the Property Data Trust Framework, which is a group of financial institutions, conveyancers and other industry participants working together to standardise data schemas.
I came in to try to find a way to standardise the data exchange using VCs. To do that, you need to know who’s allowed to issue and verify certain things, so I started looking into the DIF Trust Establishment specification, which is when I saw the publicity for the DIF Hackathon.
I thought “I can build something using Trust Establishment”. Initially I was looking at the organizational side. If each organization has a DID, they can recognize each other. I was less clear how it would work for buyers and sellers. DIDs are quite an abstract concept for people to get their heads around.
People do most of their research on the web, and I liked the idea of mimicking the SSL padlock (the icon displayed by browsers when visiting a secure website) so it verifies a site is part of the Property Data Trust Framework. So we developed a browser extension called TrustSight, building on existing work around DID configurations (used to cryptographically link a DID to a domain).
We also built a tool for deploying trust frameworks, and a tool to visualize trust relationships.
Why did you choose to develop using Veramo?
It’s tricky maintaining all these DIDs and DID configurations. There's a package by Sphereon around DID configurations, but it doesn’t provide any tooling around issuing or verifying credentials.
I was struggling to get the browser extension to work with JSON LD libraries, so I decided to use Veramo for DID and VC related operations and connect to the Sphereon resolver.
It’s made my life much easier, particularly if I want to use a new DID method or different VC formats.
What user benefits are you targeting?
As a buyer, ideally you’d apply to a hundred different lenders in order to get the best deal. However, it’s currently too expensive for lenders to do the checks. As a result, each buyer can really only apply for one mortgage, so people are very conservative about what they are looking for, to ensure they secure a mortgage.
We’re aiming to make the mortgage application process quicker and easier, and to ensure people get the right product for them.
How was your experience of participating in the Hackathon?
It was very well-organised. The criteria and timelines were all clear. I really enjoyed the Discord server. Some people were very active there, which made it feel like a community. I also liked the talks that were put on.
It was great to get to the end and see everyone’s submissions, and to feel connected and part of the community.
What next for TrustBox?
There is some tooling on the Trust Framework side that doesn’t yet exist, so I’d like to work with DIF on that.
Downloading a browser extension is still quite a big piece of user friction, ultimately you want to get into the browser itself. That’s one to discuss with the browser companies!