The Three Laws of Robotics, presented by the writer Isaac Asimov, are a set of rules introduced in his 1942 short story “Runaround”:
- A robot may not injure a human being or, through inaction, allow a human being to come to harm.
- A robot must obey orders given it by human beings except where such orders would conflict with the First Law.
- A robot must protect its own existence as long as such protection does not conflict with the First or Second Law.
Why did I mention Asimov’s Laws? Because they are principles for the design of a technological system.
Blockchain technology is a network that manages the recording of transactions in a ledger in a decentralized way, based on an algorithmic consensus between the participants, without a central authority or intermediaries involved.
The design of a blockchain and human interaction during its use define the success or failure of an ecosystem, depending on the balance of its incentives.
—What is the limit between algorithmic objectivism and human subjectivism?
—How can human action and the algorithm coexist in a “fair” balance to be favorable to decentralization?
—What are the adequate incentives for human action to behave in an ideal way to develop a decentralized ecosystem?
The Principles
A principle is a broad concept that supports a base of ideals, foundations, rules and/or policies, from which ideologies, theories, doctrines, religions and sciences are born.
The principles can be moral, the behavior according to certain social values, or ethical, the appropriate behavior of those social values. They can also be scientific, like Archimedean Principle, in physics.
Blockchain design principles bring together a bit of all of these.
I will explain the main principles, and although some more could be included, I consider that they would be related, or would be a consequence of these. Let’s see.
Decentralization
It is the most cited, and is one of the principles on which blockchain technology was built to differentiate itself from the traditional financial world, which has power concentrated in a few hands that digest the global markets.
The blockchain is a network that does not have a single core that manages all the information, but each node has the same copy, which is kept updated automatically, where each node can produce new information, given certain algorithmic conditions to do so. Technically speaking, most blockchains are built as Distributed Ledger Technology (DLT).
The main difference between a distributed network and a decentralized one comes down to where the decision is made, and how the information is sent through the different nodes of the system, and this approach is based on the network topography and not on its resulting state.
In a distributed network, its topography allows data and decision-making to be distributed evenly across all servers. Essentially, and in theory, the structure of the network is that each distributed server receives exactly the same amount of control power.
On the other hand, decentralized and distributed networks are often referred to interchangeably, and although decentralized networks also use different connection points, or nodes, to make information accessible, they do not have distributed decision-making power in each node.
For this reason, when we talk about decentralization as a principle, we usually refer to DLT networks that maintain decentralization in consensus, that is, that the generation of records is not concentrated in a few nodes, since DLT networks can end up centralized, due to their dynamics and inadequate design of incentives.
I leave my article at the end so that you better understand the difference between distributed and decentralized networks (1).
Immutability
The data stored in a blockchain is immutable, that is, once registered in the ledger it cannot be modified.
New data can only be added to the chain when a node solves a type of cryptographic proof, in the Proof of Work protocol, or when it has enough staking to be designated by the draw in the Proof of Stake protocol, but then that signed block must be approved by the rest of the network members based on its algorithmic recognition.
Adding new data is relatively easier than rewriting an entire blockchain, unless one entity controls 51% of the network’s consensus power, and can fork a new ledger, pretending to be the longest chain to fool the rest of the nodes, and that they continue writing in that ledger, which is very difficult in certain more secure networks.
Security
This principle has some relation to that of immutability, when referring to prevention of the 51% attack, but it also refers to the encrypted nature of the blockchain, which provides security over the information that is stored in its ledger, the issuance of new tokens, and the wallets that interact with the ledger to execute transactions.
The security of the algorithmic design of monetary policy is found in this principle.
To operate on the network, it is necessary to have a set of keys, public and private, corresponding to asymmetric encryption, which means that if you have a public key it is not possible (due to its degree of complexity) to decrypt the private key, instead it is possible the other way around.
Not all blockchains use the same hash and asymmetric key encryption protocol, but all transactions are signed by the issuer’s private key, which means that encryption hacking tends to be impossible, and I mean technology, not to the carelessness of people who could leave their keys in full view of malicious actors.
Privacy
The most used and popular chains of blocks are public, that is to say that their data can be reviewed by any user, and no permission is needed from anyone to participate in the network. However, the identity of its users remains hidden on-chain, and the only relevant information for the network are the addresses that carry out the operations and the amounts or terms of the agreement. Identification is not required. The only way to relate transactions and users is off-chain (social networks, IP data, KYC in exchanges, etc).
There are also private blockchains, such as the case of Monero, which hides its ledger, and only shows specific operations that can be read if the hash ID is possessed, plus the sending or receiving address.
Trustless
The intermediation of a third party that provides trust or that is a guarantor of the transaction is not required.
This principle is related to security, since the success of a transaction does not depend on the will of a third party.
the orpen source is key to this principle, since the network’s programming code is published (usually on GitHub) and can be reviewed by other programmers for bugs, data collection codes, or backdoors. Don’t Trust, Verify.
Permissionless
It means that anyone can participate in the blockchain without the need for authorization, either to carry out transactions or to set up a block-producing node.
This principle is related to public blockchains, since private networks are permissioned, and their access requires the permission of an authority.
The Consequences of Violation of the Principles
When a principle is violated, the system is corrupted, and one or several design features are lost, which conclude in a new system with new features.
Let’s see some consequences of violating some of the principles.
Anti-censorship is a feature made possible by decentralization, as distributed block production avoids a single point of attack, and this feature will be enhanced by privacy, as censorship is motivated if it is known who to attack for action.
If the encryption code is violated because it is weak, or the programming code registers errors, the security principle is lost. Therefore, in this regard, it does not matter if it is a well-known and reputable programmer or an anonymous one who developed the code, it is more necessary to examine the code than to know “its face”. Programming errors exist, and corruption, too. The blockchain must be open source.
The permissionless principle can be violated, for example, if a validator chooses to reject certain transactions, or also if certain contingency tools are introduced in the protocol, for that same action, or to reject delegations in the case of PoS, and thus the user of the network is left in a situation of requiring “de facto” permissions.
Final Words
The Layer 1 (L1) protocol of a blockchain must be as objective and deterministic as possible, to prevent discretion and manipulation at the discretion of the actors. If the design allows too much human input into the transaction processing decision, this lends itself to subjective arbitrariness. This subjectification is a point of attack for censorship.
As an example, let’s take what happened with Tornado Cash, since the regulators were able to sanction its developer Alexey Pertsev, depriving him of his freedom, for alleged charges of money laundering carried out on his platform, but they were unable to eliminate the protocol.
I believe that L1 should be neutral, and the specific new features that are added, and have some interference in the ledger registry, should be developed in L2, so that in theory, the impact is not direct on its base design.
We see how the flexibility of the principles in many blockchains is changing the essence, and this leads to the end of decentralization and anti-censorship immutability.
If the basic principles are broken, the blockchain will be the worst Orwellian instrument of mass surveillance and control ever created.
. . .
(1) Decentralization Is Often An Overrated And Misunderstood Concept
