Why Blockchains are Here to Stay
Blockchains can’t do everything, but they do two things really well: return ownership of personal data to individuals, and verify claims about identity.
As a company that sells software for issuing official records using a blockchain, Learning Machine sometimes encounters candid worries and concerns about the value and viability of blockchain technology. I wanted to take a moment to acknowledge some of those concerns and address them directly. If you have reservations about blockchain technology, hopefully by the end of this piece you will feel reassured and at least a little more open to the value blockchain technology can provide for your institution.
Now is a good time to write this piece, because over the past few weeks, Nouriel Roubini has emerged as as a prominent public critic of blockchain technology. The other day the Learning Machine team got an email from someone asking us how much of Roubini’s recent article, “Cryptocurrencies and the Big Blockchain Lie,” they should be concerned about. Our response was perhaps not what some blockchain boosters would expect: we actually agree with quite a bit of what Roubini says. Nevertheless, we disagree with him on a few main points, which in our view undermine his conclusion that blockchain technology “has not even improved upon the standard electronic spreadsheet.”
What Roubini Gets Right
Let’s start with where we agree with Roubini’s analysis:
1. The vast majority of cryptocurrencies (tokens) will end up with a lower value than they started with, as virtually no one will use them.
2. Many of them were outright scams to make money in the short term.
3. There is a lot that blockchain technology cannot do by itself—for example, ending poverty, famine, or cancer.
4. Putting “the entirety of social and political life” on blockchains is a terrible idea.
5. Although blockchains themselves are highly secure, cryptocurrency exchanges (websites that allow people to trade tokens) have traditionally been highly insecure, with many notable hacks.
6. Some blockchains are more centralized than others. Ethereum, as Roubini points out, does have a “benevolent dictator” — Vitalik Buterin.
7. “Distributed Ledger Technology” — DLT — is in many ways an improved type of shared database because its governance and administration are centralized under the control of one or more institutions. For this reason, one could argue that it is not truly a “blockchain,” which has more decentralized governance.
What Roubini Gets Wrong
While much of what Roubini observes is true, his conclusions veer in the wrong direction. This is because:
- He misunderstands how public blockchains work.
Roubini’s belief that public blockchains are insecure because they are public is a common (and understandable) misconception. He says, “there is no institution under the sun … that would put its balance sheet or register of transactions, trades, and interactions with clients and suppliers on public decentralized peer-to-peer permission-less ledgers.” This is trivially true. Pointing out use cases that make no sense does not argue against those that do, and for precisely the reasons Roubini lists, public blockchains simply are not used in the ways he suggests.
Public blockchains are called “public” because anyone can use them. Similarly, the internet is public because anyone can use it. In neither case would misguided uses of these public protocol stacks discredit the infrastructure itself.
When it comes to the question of data security, in many ways public blockchains are much more secure than the internet. This is because the data that needs to be verified is not recorded on the chain directly, but rather linked to the chain through various one-way methods. Some of those include hashing, pairwise decentralized identifiers (DIDs), and zero-knowledge proofs (ZKPs). These technologies are all specifically designed to make off-chain data verifiable using public blockchains without putting the actual data on the chain. This means that data can also be “de-linked” when necessary. These methods make using public blockchains far more secure than a traditional database, because a public blockchain cannot be “hacked” to reveal sensitive information.
2. Roubini doesn’t see the value of blockchains for verifying identity and returning ownership of data back to individuals.
This is what Learning Machine uses blockchains for. There is a lot that blockchains can’t do, but there is also a lot they can — and the two main things blockchains are already being used for today are verifying official records (claims about identity) and returning ownership of data back to individuals. It’s quite simple: without blockchains, individuals will continue to be dependent on platforms owned by large software providers to access and use their data.
It’s quite simple: without blockchains, individuals will continue to be dependent on platforms owned by large software providers to access and use their data.
Data ownership is also one of the big advantages of using public blockchains specifically. If you’re using a private blockchain, the company or government running that chain still owns the network used to verify your data and prove your ownership. But if you’re using a blockchain that no one owns, you can take your data with you wherever and verify it wherever without worrying about losing access or the ability to verify.
You can read some real-world case studies of using the blockchain for data ownership and identity verification here: they include institutions like MIT, Central New Mexico Community College, and countries like the Republic of Malta. The Federation of State Medical Boards, one of the most trusted institutions providing credentialing for medical professionals, is vocal about the potential blockchain technology has demonstrated for creating a more secure, independent credentialing ecosystem for everyone. Global adoption of blockchain for secure credentialing is gaining ground every month.
3. Roubini says that because a Distributed Ledger is basically a shared database, it’s no better than a spreadsheet. That’s not true.
Of course, as Roubini notes, not all blockchains are public. Private blockchains, or DLTs, that require permissions to use can also have value. These are like traditional databases in that they generally record much more data on-chain and have an access control system that authorizes interactions with the DLT. The more data you put on-chain, obviously the more you have to restrict who has permission to use that chain. This is why public blockchains record very minimal data, while private blockchains often record lots of data. Either may have value for different types of use cases, and each has its own security characteristics.
Distributed ledgers, even though they are permissioned and more centralized than public blockchains, can still do things spreadsheets can’t — for example, cut down the time it takes to verify the provenance of a mango or a diamond from 7 days to 2.2 seconds. By creating a shared record of movements through a supply chain, DLT can also dramatically reduce the number of disputes and the amount of time and money companies spend arbitrating between different suppliers and shipment endpoints.
These efficiencies can both save lives and save suppliers significant money. This is why Walmart is now mandating that its perishable food suppliers use the blockchain to record the movement of each food SKU through the supply chain. That way, if, say, an E. coli outbreak occurs, it can be isolated to a specific batch of spinach rather than forcing the grocer to destroy all of its produce of the affected type. The issue can then be addressed much more directly with targeted process improvements up the supply chain.
Blockchain for Identity Verification Is Inevitable
I recently heard Sandia Labs’ Nicholas Pattengale say that the impact of new technologies is overestimated in the short term, but underestimated in the long term. Now that the blockchain hype cycle is dying down, a lot of the “overestimation” is dying down as well: people are realizing that blockchains are not a cure-all for every problem. But what is still underestimated is the impact blockchains will have for enabling individual ownership of data and verification of claims about identity. Over the next 5 years, most large institutions will quietly start using blockchains for identity verification.
It’s not a question of if; it’s a question of when.
How does Learning Machine approach using the blockchain for recipient identity verification? We do so by using the Blockcerts open standard, which we incubated with the MIT Media Lab. By using Blockcerts, you ensure that your institution implements blockchain for records verification in a manner that is self-sovereign — empowering your organization and the citizens, employees, learners, and partners to whom you issue official records to own and verify their data for life. What that means is that even if Learning Machine, or any other software provider, used to issue Blockcerts goes away, you still own those certificates and can verify them using the open standard.
If you are an issuer of official records in any capacity, the blockchain stands to provide real value for your institution and your recipient stakeholders. Please reach out to us. We would love to continue the conversation.
