A Common Sense Statement on Blockchain and the Business Mainnet
“We permit our employees to consider uses of the Mainnet and public blockchain networks. We contribute and adhere to strict standards of appropriate use and security, so that we can confidently employ the Mainnet in production — when it is the right tool for the job.”
We are the people who make companies work. We keep them safe. We are practical. We build on what we have. We do not waste our time chasing every new thing. The technologies we adopt stand up to years of testing and skepticism.
If we say something works, it works. And when we say something works, we mean a specific set of functions work, under a specific set of conditions, always within specific tolerances. And when we are wrong, our careers are at stake.
When it comes to blockchain and distributed ledger technology (DLT), the path to proper use starts with knowing where not to use it.
We do not say that blockchain or DLTs always work as an acceptable way to store data, particularly not when sensitive information is involved and access control is important. We say instead, “Use traditional systems of record.”
We do not say that any public blockchain has the performance, control, or security attributes required to store all the data or run all the business logic of our mission-critical systems. We say instead, “Deploy mature platforms under well-tested procedures, with security at the top of the priority list…and performance next.”
But having spent a decade looking at public blockchain technology, we can now say with confidence that there are practical ways to use it in business that deliver unique value, support our best security practices, and reduce costs.
We define the Mainnet as: A state machine maintained as a public good that is always-on, pay-as-you-go, and which strongly resists anyone gaining control over the system, locking others out of valid operations, or tampering with the established record.
We do not expect the Mainnet to be particularly fast. We do not expect it to run all of our operations. We do not expect any single distributed system — private, public, permissioned, or otherwise — to handle the reads and writes of even a fraction of our applications, let alone everyone’s.
But we observe that today a public Mainnet like Ethereum can scale to handle certain kinds of key, low-level operations for enterprise use cases, particularly those that involve ordering events and managing hashes. That might sound boring, but in our world, important trumps boring. In our world, non-repudiation is boring, but it is also a billion-dollar problem that the proper use of a Mainnet can solve.
A Mainnet can provide a low-level state service for distributed systems that must agree on event order and enforce consistency. And it can perform these functions within acceptable tolerances for some, perhaps many, business applications — so long as it is not used to store sensitive data or run functions that could be used without permission to discover one’s business activities or relationships.
We do not say which particular platform or network, when all is said and done, will get the job of being the Mainnet. These things take time. We do say that there can be many blockchains, many DLTs, for many purposes, but there is utility in ultimately having one Mainnet perform the job of the Internet’s final check when different systems must agree on their state.
And so, we permit our employees to consider uses of the Mainnet, and we adhere to strict standards of appropriate use and security, so that we can confidently employ the Mainnet in production — when it is the right tool for the job.
Celia the CISO
These words were written with many people, many “Celias and Cecils” in mind. My personal belief is this: the Ethereum public network is the most likely candidate for the job of “Mainnet,” and techniques like the Baseline Protocol open up industry to its appropriate use. You may have other networks and protocols in mind, and that’s a conversation worth having.