Gartner on blockchain: Questions to ask and lots of caveats
It's immature, misunderstood, untested in the enterprise, risk-laden -- and needs to be on the CIO agenda. Gartner's Nick Heudecker frames the blockchain questions you should be asking.
By 2030, Gartner estimates the impact of blockchain on the world economy will be on the order of $3.1 trillion....
Continue Reading This Article
Enjoy this article as well as all of our content, including E-Guides, news, tips and more.
Twelve years hence is a long time, especially in technology years, but $3.1 trillion is a big number. And according to Gartner, blockchain in 2018 is still very much a technology in search of a use case.
The technology's underlying concepts are misunderstood, current blockchain tools are immature and their application in mission-critical business operations is unproven, Gartner warned CIOs in a 2017 report, underscoring that bitcoin is blockchain's only production use case to date. At the consultancy's data and analytics conference in March, analyst Nick Heudecker told an audience of chief data officers and other analytics professionals that he had "yet to see a production use case of blockchain that wouldn't be better off with a centralized database."
That doesn't mean CIOs and other digital business professionals shouldn't study blockchain and explore its use in the enterprise. Gartner believes that blockchain's potential to make digital transactions more secure, transparent and cost-efficient -- and do so in a way that bypasses traditional institutional frameworks -- has profound implications for any industry that shares data and uses data from outside its own walls. That makes it a technology of interest to any CIO involved in shaping a company's digital business model.
In a dense, information packed session at the conference, Heudecker dived into the blockchain questions he's been getting from clients and outlined the blockchain questions IT professionals should be asking as they begin to explore this latest digital disruptor to business.
This conference notebook summarizes some of the key information from the presentation, including important concepts, primary use cases and major blockchain challenges.
Basic blockchain concepts
Heudecker began his presentation by going over some of the basic concepts that underpin blockchain technology: that it is an electronic ledger which provides an authoritative record of data or events, such as monetary transactions, property records, and other valued assets; that it does so in a decentralized and distributed manner, creating linear lists of transactions cryptographically joined in sequential chains of blocks that are replicated across peer-to-peer computer networks. Its authority is derived from consensus. The record cannot be changed after the fact without changing all subsequent blocks and without the consensus of the network.
Here's the drill, as laid out in a Gartner chart: A network member creates a transaction and proposes it to the network of computers or "nodes." The proposed transaction is combined with other proposed transactions in a "block." Individual nodes compete to solve the cryptographic puzzle, and the first correct answer is distributed to the network. The confirmed block is appended to the blockchain.
"Every time a new block is added to the blockchain, it refers to the block that came before," Heudecker said. "The idea is that if everyone has access to all of the data in the chain, you can validate every single transaction that's occurred and track ownership."
Stock trading, land property rights, IP protection, supply chain auditing, ride-sharing minus the ride-hailing companies, are among the blockchain applications being explored. Not just a passive data record, blockchains can add dynamic, programmed behavior to transactions with self-executing "smart contracts." Replenishing your college student's cash card automatically when it hits a certain minimum balance is an example of a "very simple smart contract" one could put on a blockchain. "Or how about automatically enforcing a prenuptial agreement on receipt of a divorce notice, with no lawyers involved," Heudecker said to some groans from the audience.
Blockchain questions around consensus
The concept of consensus, however brilliant and groundbreaking, turns out to be one of blockchain's many challenges. Validating every block in the network is "incredibly expensive," Heudecker said, at least in the bitcoin blockchain, which uses what's called a "proof of work" consensus model.
"The electricity consumed by computers mining bitcoin or to confirm blocks -- these two things are the same -- is massive. I've read estimates of anywhere from 1.5% to 3% of global electricity production goes into bitcoin mining alone," Heudecker said.
Among the blockchain questions CIOs need to ask: What other consensus methods are companies exploring that lower the electricity consumption? Heudecker cited proof of stake, where participants who don't have a stake in a transaction don't need to confirm it, and leader-based consensus as two methods which lower the overhead.
"The challenge in all consensus methods is that you have to ensure they are going to scale," he said, adding that in bitcoin, the average number of transactions that can be processed is about seven to eight per minute.
"This is going to be the main reason that blockchain efforts today will be throttled in most enterprises -- not getting the throughput," he said. Also, the blocks are not that big. "You can only put so much stuff in them," he said, the average being 1 MB.
Four challenges in building a blockchain application
Heudecker drilled into four challenges IT experts will face in building out a blockchain application.
The first is that you have to pick the platform you're going to participate in and current technology is neither scalable nor complete in Heudecker's view. "Today, there are really two, maybe two-and-a-half," he said, calling out Ethereum and Hyperledger.
Ethereum is a public, open source blockchain with smart contracts functionality. "You can build business applications on it, but what it looks like from a production standpoint and from a quality and security standpoint is still undecided," he said.
Hyperledger is the open source blockchain platform that runs in the cloud, is operated by the Linux Foundation and is being commercialized by IBM, Oracle, Microsoft and others. "You can replace different elements of it, if it does not meet your needs," Heudecker said. "You can change out the consensus model and the databases, customizing it to your needs."
"Figuring out the maturity of where they are going and how they meet your needs is a challenge," he said.
Heudecker's second challenge was a business one: getting competitors to cooperate. Businesses will need a network to make blockchain applications viable, which might mean working with business competitors or getting supply chain partners (who compete with other) to integrate on a single platform.
"Not many companies have the clout and market to make that happen -- only some of the largest retailers and manufacturers can force these competing interests to cooperate and collaborate," he said.
Data interoperability and the need to agree on the structure and format of the data is another challenge of a distributed ledger technology.
"In a centralized mode, you can rely on a single entity to create a data standard -- SWIFT [a system for financial transactions] is a good example," he said. "But if I don't trust a centralized authority, I now inherit all the overhead of creating these interoperability standards. And you have to get all of your cohorts in the network to agree on these standards -- a challenge in its own right."
The "solution challenge" rounds out the list. Building a powerful, flexible application using blockchain that is integrated with your company's business processes and meets regulatory mandates requires enormous resources. And, according to Heudecker, what this solution should look like has yet to be defined.
"That doesn't mean you shouldn't experiment. You have to start building some competencies and skills around what is possible, knowing that these things will change over time," he said. "We're just getting started."
Three Gartner lists to help you along the way
Before you start, answer these five blockchain questions:
- Where will it be implemented?
- Who all have skin in the game?
- What is the scope of the work?
- Why is blockchain the best option?
- Where this project is needed at all?
Use these guidelines to keep you on the right track
- Think strategically not tactically.
- Assume whatever blockchain technology you choose will be obsolete in 18 to 24 months.
- Undertake proofs of concept to learn about the major platforms.
- Select a limited, narrow scope use case for real deployment on a chosen platform.
- Prepare to migrate off that platform in 24 months.
- Use the experience to reimagine business processes, business models, markets, products for the era of the programmable economy.
Additional capabilities blockchain needs to become enterprise-ready
- Bug-resistant smart contract capability
- Multiple independent implementations of protocol specs
- Ecosystem of tools and frameworks
- Transparent governance with agile effective response to threats
- Scalable to global economy
- True open source from start
- Modular architecture and open APIs enable ecosystem
- Hardened via public blockchain deployment
- Configurable for private or hybrid blockchain
Gartner, "Building Blockchain Into Your Data and Analytics Program," Nick Heudecker, 2018.