We’ve previously written how interoperability will hold back blockchain adoption, at least until we can find ways around the problem. The cost and friction of joining multiple blockchains may hinder widespread adoption until we can figure out how to get them to talk to each other and reduce the cost of joining a blockchain implementation. However, recent thinking suggests there are some shortcuts we can take to make better use of blockchains in the short term, as their development and adoption matures.
For example, recently I met with the Deloitte blockchain team, and Principal Eric Piscini disagreed with my premise. He believes that interoperability really isn’t that big of an issue. First, he points out that, today, we have multiple environments that don’t connect to each other and the work still happens effectively. For example, different credit card payment vendors each have unique systems but everyone can still use any of them without an issue.
He also notes that interoperability seems like a bigger issue if you look at the blockchain implementation as needing to do every part of a transaction. However, he thinks of blockchain as having three layers:
- Recording (actual transcribing of data into a block)
- Transacting (an activity or transfer, such as moving money from one participant to another)
- Business logic (the rules and controls of a process coded into the system)
You don’t have to do all three things in blockchain. You can use it for any of the three, or some combination. And as a result, you start to see how it’s possible to use blockchain technology and not necessarily have to worry about interoperability. It’s not dissimilar to evaluating automation technology, where you will, simply, fail if you try to automate everywhere possible – you’d run out of time, money and patience trying! Most experts will tell you to first focus on what not to automate, which is similar with blockchain: first figure out where you can carry on just fine without all the expense and disruption of a blockchain implementation.
Piscini also believes, in some instances, that firms do not need interoperability, but more a single blockchain per asset class, as it will be near impossible to transfer the same value across multiple blockchains.
So, where does this leave us with our interoperability decisions?
1) Blockchain interoperability needs both a technology choice and business reason to exist. We need to separate the technology of blockchain from the business application of blockchain and from the business model of blockchain-based systems. From a technology perspective, for example, multiple blockchain implementations can exist and drive value even if not connected to other blockchains.
2) Network ownership may be more important than technical interoperability. For networks that are, essentially, owned and controlled by one party (the credit card examples above) and other parties just access those networks but don’t need to integrate per se, then Piscini’s view makes total sense. It also works in situations like Ariba’s, which we’ve written about before, where participants on don’t need blockchain implementations themselves to use Ariba’s blockchain. (Ariba also notes that clients can choose to do just recording on the blockchain, further supporting Piscini’s point of separating blockchain into layers.) However, in networks where the peer-to-peer aspect is more important, and no one participant has strong power, we believe interoperability will continue to be a barrier to widespread adoption.
Bottom Line: Clarity around when/if/how interoperability is really needed for the blockchain market to mature.
We expect that, by the end of this year, as companies continue to tackle implementation challenges like interoperability and the development of common industry standards continues, will the market will begin to pick winning platforms and technologies.
 Many consortia are dealing with this issue as we speak, and government agencies are beginning to weigh in. Expect a lot of activity in standards development this year.
Posted in : Blockchain