Last month, Accenture quietly open-sourced its Blockchain Integration Framework (BIF) via Hyperledger. The project is an interoperability solution that currently supports Hyperledger Fabric, Quorum and R3’s Corda. BIF aims to enable the exchange of on-chain data or assets between separate permissioned blockchains or networks.
There are hopes for the initiative to eventually become a Hyperleddger project, but given its infancy, it is currently part of the “Hyperledger Lab”.
There’s an increasing demand to enable enterprise blockchains to interoperate. Trade networks need to talk to each other, and the banks involved in trade finance networks want to tap into the data stored on the trade blockchains.
BIF is not Accenture‘s first run at the interoperability issue. More than a year ago it unveiled a very different version which required a trusted interoperability node. That involved a centralized server sitting between two networks, a not-so-desirable solution for decentralized applications.
However, unlike its predecessor, the latest incarnation, BIF, set out to achieve interoperability “without a middleman”.
How it works: the export, import process
Essentially interoperability involves exporting data. When you export physical goods, many parties verify the shipment and the export process. That might include getting an export license, a certificate of origin, a bill of lading and several other documents.
The critical point is the importer and its bank know the transaction is legitimate even before seeing the goods because other organizations have said so.
Likewise, the BIF blockchain interoperability approach assumes that several parties confirm the exported data is correct.
So each network has a set of “interoperability validators”. These interoperability validators confirm to the second network that the outgoing data is legitimate. The second blockchain network can verify the proof based on the public keys of the first chain’s validators.
What’s in store?
A critical aspect of interoperability is identity management. Every network will have a way to manage the identity of participants. Hence part of the roadmap is to integrate the Hyperledger Indy identity framework.
Other future features include atomic swaps in which one asset is exchanged for another in real-time. If one leg fails, then both fail, to avoid the situation where an asset is transferred, but the payment doesn’t complete. Plus there are plans to integrate with additional ledgers such as Hyperledger Sawtooth and Digital Asset.
In order to graduate from Hyperledger Labs to a fully-fledged project, the initiative will need to expand its community and documentation. It already appears that Fujitsu is a participant.
Other blockchain interoperability initiatives
Accenture certainly isn’t the first to tackle interoperability, and likely not the last. There are a variety of technical approaches.
In fact, Hyperledger has another interoperability project, Quilt, which is the open-source version of Interledger originally developed by Ripple and used to exchange payments or currencies.
Also targeted at permissioned networks, Clearmatics, the developer behind the Utility Settlement Coin (USC) project has its ION protocol. This enables separate private Ethereum networks to interoperate as well as with Hyperledger Fabric.
Meanwhile, R3 would argue that Corda has interoperability baked in. Its Corda Network enables separate Corda ledgers, if they so choose, to interoperate by using a shared network and identity infrastructure. But it only works with other Corda dApps.
Quant has its Overledger protocol which it positions as a messaging layer for multi-ledger applications.
Sidechains were initially invented to help public blockchains to scale be moving transactions off the main blockchain. But these are also an interoperability tool because private blockchains can run as sidechains of public blockchains. OneLedger seems to be in the early stages of working on this concept. And along these lines, ConsenSys released a paper earlier this year entitled “Atomic Crosschain Transactions for Ethereum Private Sidechains.”
So there’s plenty of work happening on the technical solutions to interoperability. Accenture, with its diverse range of enterprise blockchain projects, is in a great position to figure out what’s needed. But the technology is only one important piece of the interoperability puzzle. Issues such as standards and governance may prove to be even harder.