Economists have made lists of requirements that money in general needs to have, like durability, scarcity, portability etc. Since blockchains are data and not physical objects, some of these qualities, like durability and portability are always guaranteed. And on other qualities blockchain money can give even higher assurances than traditional money like gold: Future gold discoveries here on Earth or in space might vastly increase the available amount, and rising gold prices can make previously uneconomic mines economical to mine again. With programmable emission schedules, blockchains are not subject to such nasty surprises.
What then makes for good blockchain money? This below list represent my personal view on a blockchain’s hierarchy of needs.
-
soundness of cryptography and consensus logic
-
consensus stability
-
predictable and immutable emission schedule
-
good “properties”: privacy, transaction throughput, post-quantum security, scalable etc.
-
backwards compatibility for nodes and miners (no hardforks)
-
Forwards compatibility (no softforks)
-
Implies that ownership is enforceable (my money cannot be stolen) and that no one can print more money than was intended when the protocol was designed. Bitcoin’s inflation bug from 2010 or our own from July of this year are examples of money not having this 1st quality. All other concerns are irrelevant when such a problem is observed, and the protocol must be rolled back to before the incident occurred.
-
Consensus stability means that the blockchain should, for the large part, work in an append-only manner. If you’ve received a transaction, it must be impractical for the sender to present an alternative view of history where they did not make this payment. Shallow reorganizations cannot be completely avoided though, so it is the receiver’s responsibility to wait “long enough” before considering the payment final.
-
No one may change the emission schedule to benefit themselves. Once the chain has been launched, the emission schedule should not be up for debate. The reason for this is that it breaks the implicit contract with all holders: Everyone buying or using the asset do so under the assumption of the current rule set. If emissions were to increase, the value of all holdings would be diluted to a lower part than what they expected when they bought in. Notice though that this point has lower priority than consensus stability. If the proof-of-work mechanism is unable to provide consensus stability, it would be justified to change the emission schedule to reward miners more in order to ensure stability.
-
All the stuff the Neptune protocol excels at: privacy, scalability, programmability, and for the moment post-quantum security. But in a world with sufficiently powerful quantum computers, post-quantum security moves from priority 4 to priority 1.
-
Hardforks are discouraged for two reasons: a) Existing node and wallet installations should continue to work and b) the developer’s power to change the rules must be constrained. If a change improves any property of the blockchain (item 4 above), a hardfork can be justified given a high degree of consensus among the stakeholders and developers.
-
Softforks are discouraged because they, like hardforks, introduce a degree of arbitrary power into what should be a neutral and decentralized protocol. Softforks may also require changes to the setups of miners in a way that reduces the value of their investments, thus potentially making the network more insecure against reorganizations.