I work in a large consulting firm which is marketing a mutable blockchain solution to its clients. But I personally fail to grasp any advantage of using a blockchain design if you are discarding immutability. Am I missing something here?
I agree more context is needed to provide a more directly meaningful answer.
But in general: a blockchain allows users to interact without trusting a central authority, instead they verify the system’s state on their own. As a user, I can verify that I own some amount of bitcoin by running a node, and seeing the transactions that paid me as confirmed in the blockchain record. I know that the network’s rules do not allow those coins to be moved unless a transaction is signed by the valid keys (which I control). If the system allows mutability, then how confident can I be that the transaction which paid me will not be changed in the future? With an immutable blockchain, the energy spent on POW mining helps guarantee this, but if mutability is allowed... ??? How can I be sure?
If some form of mutability can be added while preserving this factor of trustlessness, then maybe a ‘mutable blockchain’ is reasonable. But I am yet to see such a proposal. If the user must trust a central authority to not change the system, then using blockchain architecture is superfluous.
In reality, I suspect in most cases a ‘mutable blockchain’ would be an inefficient way to implement a database, albeit one that can be dressed up in marketing buzzwords and sold to customers more easily due to the ‘blockchain hype’.
But I personally fail to grasp any advantage of using a blockchain design if you are discarding immutability. Am I missing something here?
Advantage: customers love blockchain hype, so they’ll be more willing to buy what you’re selling.
I’m sorry to say, but I think your intuition is correct. Frankly, such projects seem either misguided, or scammy to me.
EDIT: perhaps worth mentioning as well: allowing a central authority to alter the blockchain record creates a monstrous security risk. No matter how well intentioned or benevolent the authority is, an attacker targeting their system of control represents a huge existential risk to the network. There is no such thing as a secure backdoor that only ‘the good guys’ can access.
Most block-chains are actually mutable in some way and this has some advantages. This came clear in some incidents where a bug or a flaw in a chain was exploited resulting in many people losing a large amount of assets. Afterwards people would hard fork a previous state of the chain and continue from there, after getting the consensus of the community.
People usually need immutability to protect the chain from illicit modifications of the network. So having a way to easily modify the history of a block-chain would be very beneficial if you can ensure that the such changes are only done under some kind of consensus and is protected from different attacks.
Examples where this comes in handy is if you want to have a dynamic chain where you keep the current price of stocks for example or a social-media chain where you could edit or remove a post. Of course some workarounds exist in immutable chains, such as creating a new transaction that flags an old transaction removed, but people would still be able to read it and this would add overhead to the chain make the client more complicated. However, that shows how important is the consensus mechanism to decide who has the rights to modify the chain and in which manner.
I can see a potential mutable blockchain being valuable for a number of reasons:
- If only 'trusted nodes' are allowed to change the blockchain, and there is significant reputation and financial resources at stake at making a false claim and
- The mutability is limited to 'fraud protection'/reversing transactions, so that if someone is genuinely a victim of fraud (or they lose their coins...) the blockchain can be modified to release these funds, rather than permanent loss as is the case with bitcoin/immutable blockchains.