Robert Kugel's Analyst Perspectives

Blockchains Pose Problems of Persistence and Trust

Written by Robert Kugel | Jan 15, 2020 2:00:00 PM

I’ve written before about blockchain’s significant potential. A lot of the current discussion on the topic centers on cryptocurrencies and financial trading platforms, both of which are already in operation. However, my focus is on its applicability to business generally, especially in B2B commerce, where I believe there is significant potential for it to serve as a universal data connector. There’s also a great deal of potential for blockchain to provide individuals with greater power in managing their identity and greasing the wheels of trade. That noted, those designing and planning to implement commerce-related blockchains must address fundamental issues if blockchain technology is to achieve its potential.

One issue is the persistence and trustworthiness of underlying records that are related to a blockchain. This may come as a surprise to some — I’ve found that many people believe that the blockchain will house the data itself. That’s likely to be the case if the amount of data associated with the blockchain record is trivial. However, it’s highly unlikely that larger digital objects will be incorporated, so it will be necessary for the blockchain to reference these larger records. What’s trivial? For now, one should assume that the limit is measured in kilobytes — far too small for, say, many medical records or facsimiles of paper documents. For those objects, there will be some sort of pointer to a record that’s stored somewhere. And the persistence of that “somewhere” is no small matter.

Blockchains derive their security from having multiple nodes that are constantly synchronizing. This structure makes it almost impossible for anyone to delete or tamper with the data since any illegitimate attempt to add or alter a record on one node will be detected immediately in the others. But what of the underlying documents? Even if the pointer to the original image of a deed, certificate or body scan persists, it becomes useless if the physical server on which the underlying record is stored is removed from service. I call this the “error 404” problem since the result is the same as a broken link on the internet. For blockchain to live up to its promise, there must be some efficient mechanism that guarantees the integrity of the storage node.

Addressing these potential issues of persistence and trust will require redundancy. There must be multiple copies of source records to prevent the unintended destruction of the underlying data. There also must be some sort of protocol when a systemically significant data storage site is removed from service or even taken down.

There’s also a trust issue with respect to underlying data held outside the blockchain. Even if the data record is there, how do I know that it hasn’t been altered at some point along the way? To ensure integrity, blockchains are designed to require independent consensus before a blockchain record can be changed. Unless that applies to the underlying data, though, that integrity is a hollow promise. Maintaining multiple copies of source image documents or other large binary files is also essential to avoid what is intended to be unimpeachable evidence being tampered with or destroyed. This is especially important with respect to identity management blockchains. What’s more, if one subscribes to the Universal Declaration of Human Rights, birth- and citizen-related data is the property of the individual, not the state. And the duty of the state is to preserve these records unaltered. Ensuring this is the case is likely to be an ongoing challenge for those using identity management systems. This applies to individuals who rely on identity management blockchains to store and verify their credentials as well as those who rely on the accuracy of that information. Trustworthy data is especially a concern in countries that do not acknowledge the rights of the individual. If the state believes it alone is the rightful owner of the data, a citizen of that state cannot be secure in his or her identity.

The challenge of the persistence and trustworthiness of data held external to a blockchain is just one of the many practical issues that those who design and maintain these blockchain systems must address if blockchain technology is to achieve its full potential. In many cases, standards will be established by a community. There will be a need to establish the means of confirming that the managers of the blockchain nodes are adhering to these standards. Matters of cost (who pays for maintaining multiple nodes?) and jurisdictional issues (will there be an international body that maintains records?) won’t be easy to resolve. The substantial effort that’s been required to get TradeLens (a digital platform that serves as a single, secure source of shipping data) to its current state is testimony to how difficult the non-technology-based issues can be. The same holds for identity, traceability and property blockchains. It’s unlikely that there will be a single practical solution that’s applicable to all blockchains. What’s needed now is a broad recognition that holding external data securely and persistently is essential — so that the hard work of developing standards, frameworks and methodologies can begin immediately.

Regards,
Robert Kugel