When you build a piece of heavy machinery like, say, a bulldozer, standards are of immediate and obvious importance. Without standards, the same nuts and bolts could not be used to build different machines, which leads to cost overruns and compliance issues.

The matter is a lot less clear, though, when it comes to emerging technologies such as distributed-ledger technologies (DLT). Hell, the industry can’t even standardize around the technology’s name—is it DLT or blockchain, which is technically a type of DLT, but is now used interchangeably with the acronym.

James Carlyle knows a lot about standards. He not only actually built bulldozers—in Japan, at the beginning of his career—but also worked inside banks, where standards like the ISO 20022 for payments are the rails on which crucial functions run.

Today as chief engineer at enterprise blockchain technology company R3, Carlyle’s passion for standards hasn’t diminished, especially as companies such as his look for ways to make blockchain systems interoperable.

If we are not coming up with standards that make the exchange of data some orders of magnitude easier, we will only gain a fraction of the digitization dividends that DLT promises to bring.

Hans Huber, Main Incubator

“The industry is at a critical juncture in its efforts to define and adopt improved data and process standards,” Carlyle says. “There is no commercial advantage to organizations developing and maintaining standards separately.”

Without industry data standards, two blockchain systems, even if connected, could not exchange information in a meaningful way. And many different areas of blockchain technology still need standardizing, such as data taxonomies and smart contracts. But standards in blockchain are in something of a chicken-and-egg situation: Which comes first, the standards or the technology? How complicated is it to make chains interoperable?

Why Standards are Needed

Standards for blockchain have been necessary for the past four or five years, ever since it became clear that chains would need to operate between firms doing, for example, swaps transactions, says Hans Huber, project manager at Main Incubator, a subsidiary of Commerzbank.

Taking blockchain out of the equation, Huber notes that banks already run miles of cable and spend a fair amount of money and man-hours connecting systems so that the data can flow seamlessly across the organization. While information can flow across a single distributed ledger easily, problems occur when trying to connect a blockchain with a legacy system or another distributed ledger.

“This has been, and is still, a tremendous effort—it is costly, it is time-consuming for all participants in business, it reduces time-to-market and excludes especially smaller companies,” he says. “If we are not coming up with standards that make the exchange of data some orders of magnitude easier, we will only gain a fraction of the digitization dividends that DLT promises to bring.”

Michael Spitz, CEO at Main Incubator, lays out the problem of blockchains that cannot interoperate, using a trade financing project as an example. If you wanted to deploy the project from one framework to the other, you would be out of luck—it cannot just be copied over.

“You cannot use [the open-source Hyperledger] Fabric and then right click, see how it works, and then deploy it to [a different framework],” he says. “There is a different framework, a different language, a different consensus mechanism. So there needs to be adjustments.”

Interoperability means one transaction can run on two different frameworks, Spitz says, which is very complicated. It took his team six weeks to move from the Hyperledger Fabric into Corda, another open-source blockchain-specific tool. But that still doesn’t mean they can get Fabric and Corda to interoperate.

It’s important that they figure out how to do this, in order to avoid the trap of vendor lock-in, Spitz says. “But we need to understand how we can actually build on different frameworks.” 

If two blockchain systems have standardized the control processes they work on, they can talk to each other, says R3’s Carlyle. The nodes in a blockchain system “talk” using semantics, communicating constantly in an ongoing discussion.

Anyone smart is designing applications that will be going forward blockchain agnostic and be able to operate over a number of different blockchains.

Hirander Misra, GMEX

“If a different system has a completely different model about how finality is achieved, or how data gets propagated, then there has to be a mapping between those conversation semantics. They can’t understand each other at a native level,” he says. 

For example, a system that has a concept of strong finality, like Fabric or Corda, finds it hard to work with a system like crypto networks Ethereum or Bitcoin, which have a probabilistic finality. The first system doesn’t know when the second system is really reaching execution, Carlyle says. “It doesn’t give clarity as to whether the data on the other system will never change in the future. So, standardization around the way these blockchain systems communicate must be modelled on trust; it becomes very important.”

To ensure two different blockchain systems are able to effectively interoperate, trust between the platforms must be determined. Trust must also be able to be verified through data provenance and the ability to trace all the way back to the transaction that created the data in the first place, Carlyle says. 

Be Specific

What does interoperability actually mean in practice? It turns out, that might be a difficult question to answer.

Caroline Malcolm, who heads up the Organization for Economic Co-operation and Development’s Blockchain Policy Center, says standards should create an environment in which different blockchains can talk to each other. The OECD is among the organizations liaising with ISO/TC 307.

“We don’t want to get locked into one particular blockchain protocol or another, it is quite important from that perspective of competition,” Malcolm says.  

It might help even to come to an agreement of what blockchain is in the first place, she says. “People use the word blockchain like they use the word Kleenex for tissues, but it does in fact have a specific meaning. Many things that get termed blockchain might actually be part of a broader group of what we would call distributed-ledger technology.”

There isn’t even a single definition of what interoperability is, says Marta Piekarska, director of ecosystem for Hyperledger, a project started by the Linux Foundation to develop open-source distributed ledgers.

“We don’t have an idea of what we mean by interoperability,” she says. “You can see it as doing atomic swaps: so basically moving assets or data from one blockchain to another.”

Atomic swaps are smart contracts that allow the exchange of one cryptocurrency with another, without an exchange.

She continues: “You can see interoperability as it is defined by some of the companies that are building a layer on top of all blockchain that allows for a kind of translation between, so not moving assets but rather translating between different blockchains. And then another interoperability would be just kind of standardizing the messaging between blockchains. And then you have side chains, which is yet another thing where you basically freeze assets on one blockchain, move it to a side chain, and then move back because the main blockchain is the core of the transactions,” says Piekarska. 

Indeed, it might even be too early to define interoperability standards at all, John Whelan, head of digital investment banking at Santander, says.

“Certain standards, however, could have immediate use,” he says. “Standards for cross-chain interoperability will come later and, from a technical standpoint, we are just beginning to understand how this level of interoperability might be implemented.”

Down Under

Another challenge facing the sector is that there are a lot of entities looking to create standards around the globe. As an example, the International Organization for Standardization (ISO) has set up a technical committee, ISO/TC 307, for this purpose. Committee chair, Craig Dunn, used to be chair of a Sydney fintech start-up hub and has a background in banking as chief executive at financial services company AMP and a board member of Australian bank Westpac.

Dunn is well aware of the dilemma about choosing the right timing for the standards.

“On the one hand there is a call for standards to help develop a technology and do it in a standardized way, in order to improve the rate of innovation, lower the cost of innovation, and those sorts of things,” he says. “On the other hand, you can’t rush the process, because if the work is only in its infancy, or there isn’t yet consensus on which particular area of the technology to develop, you are not in any position to publish a standard.” 

The working groups within the technical committee include a broad spectrum of people, including representatives from larger technology groups working on blockchain, start-ups, lawyers, research scientists and representatives from standards bodies around the world.

“Generally, we believe the more involvement we have from people that are actually developing the technology, using the technology, the more likely we are to develop standards that are useful and impactful,” Dunn says. 

ISO/TC 307 plans to first release a technical report on smart contracts and standards of terminology within the next 12 months. The committee is also looking at themes like reference architecture and system interoperability.

Setting the Pace

Some believe, however, that it will be the market and not international bodies that will determine which standards come to define blockchain operations.

Tom Grogan is a lawyer at law firm Mishcon de Reya, where he co-leads the firm’s blockchain group. He says the key blockchain platforms are competing in the area of standards. “To a certain extent, I think it would likely be a market decision as to which of those wins the day,” he says.

“Obviously, R3 Corda has probably got the biggest foothold in the financial sector, and benefits from a number of consortium members being big financial players themselves. So it may well be that ends up being the entity that wins the day, but it depends on whether or not we are saying those standards have got to be set by the platforms themselves or some sort of external body.”  

Conor Svensson, chair of the Technical Specification Standards Working Group of the Ethereum Enterprise Alliance (EEA), which is working with ISO/TC 307, believes there will be more flexibility around finalizing standards. “We are collaborating with everyone we see as relevant. We are not trying to lock anyone out of the ecosystem,” says Svensson, who is also founder and CEO of Web3 Labs, which provides analytics around smart contract applications and blockchains.

Certainly, firms just want to engage with a technology that works across as many platforms as possible. Hirander Misra, chairman and CEO at GMEX Group, which is looking to launch a derivatives exchange, says when his firm developed applications on blockchain, they looked at a range of blockchains, including Corda, IBM, Hyperledger, MultiChain and Ethereum. Given there is a lot of development overhead, resources have to be devoted to one over the other in the near term. 

“Anyone smart is designing applications that will be, going forward, blockchain agnostic and be able to operate over a number of different blockchains,” he says. “And you can use them as a form of message bus as well.” 

John Mizzi, chief strategy officer at Bond.One, a platform for tokenized debt securities, says six months ago he would [have] predicted that one blockchain protocol would win out among financial services organizations and everyone would transport their data over to work on that one consistent protocol.

We don’t believe the world is ready for a standard yet. The technology just hasn’t developed sufficiently to warrant a standard at this point.

Craig Dunn, ISO

His views have changed a bit since, he says: now, instead of requiring all of these applications to be transitioned to one single protocol over time, there are organizations coming up with interoperability solutions. 

“Microsoft is focused on this as well,” he says. “We can develop our application on [open-source blockchain platform] Quorum, we can work with a counterparty or a partner who can provide a complementary service, who has built their solution on Hyperledger or R3’s Corda, and Microsoft is providing technical tools that allow us to process transactions across different protocols somewhat seamlessly. I think what ultimately is going to be the case is there will be tools that enable these seamless cross-protocol processing of transactions, and that allows all of the development that has taken place on one protocol or another protocol to remain the way they are.” 

Limited Success

Perhaps the greatest hurdle facing this global standards push is that only a handful of truly successful, live blockchain-based platforms are currently being used on a wide scale in finance. Greg Schvey, CEO at Axoni, a capital markets technology firm that specializes in distributed-ledger infrastructures, says that until there are more real deployments, the standards push will be slow to unfold.

“There is so much work to get one of these networks or systems setup, that trying to accommodate some other project, some other network that has all of its own risks of execution is generally not the top priority, at least at this point,” he says. “Five years from now it might be different when there are some more established networks out there, but for the time being, getting these things over the line for their own business case tends to be the top priority.”      

Consensus around standards is hard to get to even in established areas, says Sam Chadwick, head of blockchain research at UBS. For example, there has been a lot of difficulty in getting to a standard for know-your-customer information, as each regulator in each country has different requirements.  

But there is value in the exercise of trying to get to standards, Chadwick says. Even if financial firms don’t end up using a distributed ledger, through the discussions on standards it will have created a common understanding of how the data might be represented, and that is valuable. He gives the example of messaging standards for Internet of Things (IoT) devices, which, like mobile phones, each have their own protocols.

“As we think about platforms for IoT interfacing with the financial services industry, those very same discussions around the standards carry over, even without a blockchain underneath it,” he says.

ISO’s Dunn concedes that it’s still early days for interoperability among distributed ledgers, and says the technical committee needs to study the issue more. 

“We don’t believe the world is ready for a standard yet. The technology just hasn’t developed sufficiently to warrant a standard at this point. That is one of the challenges of working with a very new technology,” he says. “Depending on how the technology matures and develops, it will influence how quickly a standard [for interoperability] becomes available.”