GoUncle WhitePaper

In the Bitcoin paper \cite{Bitcoin}, Nakamoto thought a “one-CPU-one-vote” way for the Bitcoin participating nodes to determine representation in majority decision making, and implemented a Proof-of-Work (PoW) hashing, aka mining, game for the participants to count the CPU votes. Since then Bitcoin has been working correctly for twelve plus years. However, because the PoW mining enthuses competition, the voting game has gradually and eventually turned to “one-X-one-vote” where X stands for massively parallelized CPUs and/or GPUs, specially designed hardware rigs, and even pooling of such into huge farms. What now remains being fortunately valid is that these X’s, though no longer forming a majority of all Bitcoin participants, stay in control for Bitcoin to run correctly, thanks to the following two facts: (1) a majority of these X’s are honest, and (2) the majority of all Bitcoin participants are not functional nodes for Bitcoin anymore, at least not in Nakamoto designed “one-CPU-one-vote” way. Thus, the participants in Bitcoin, who are functionally dependable for lining up blocks need to be computationally powerful, and increasingly so. Bitcoin has an increasingly low utilization of its participants.

Bitcoin’s low utilization of its participants can also be clearly seen from its algorithmic method to diffuse messages. A public blockchain uses a hop-by-hop “gossip” way of communication to distribute data in blocks for the blockchain to log immutably in its public writing database (DB). The use of the word “gossip” as a technology descriptor, although sounding a little graphical, precisely describes how a blockchain DB is robustly and reliably secured by the distributed participants in the blockchain network. Each participant hop-by-hop receives and forwards a block, and deposits the block’s payload logs in its local duplicated copy of the DB. The gossip algorithm for Bitcoin can be referred to as “gossip-on-syntax” in that, a block qualifies gossiping if an easy hash validating returns YES for passing a {\em work} threshold. This easy syntax checking gossip function is serviceable by almost all participants. It instructs gossipers to assist only participants who are computationally powerful. Because a syntax gossip formulation cannot perceive the network topology structure of powerful miners, Bitcoin gossipers are unfortunately responsible helpers for a so-called “one-CPU-51\%-of-the-votes” attacker.

A more resounding manifestation for Bitcoin’s low utilization of its participants is by the well-known slowness to lookup Bitcoin’s ledger. A newly found block extending the Bitcoin blockchain needs a long delay, in specific after having been buried under 7 new blocks extending the chain, for the block to gain confirmed trustworthiness. So a user looking up a new transaction logged in this block must wait that long time of about 70 minutes. In need of a long delay for slowly confirming trustworthiness, Bitcoin’s ledger, i.e., the UTXO Set, has never got a trusted operator to perform any DB management (DBMS) work. Such a no-trusted-operator-to-manage ledger DB, after having grown to a large size, has to keep occupying the RAM as a whole data chunk. To input/output the ledger (which data chunk?) between the RAM and external storage space would take too long time. Consequently, most of the Bitcoin participants and/or users are unable to validate the ledger’s correctness as they cannot afford a huge RAM to conduct the validation job. Nakamoto’s “one-CPU-one-vote” ideal turned out to also necessarily mean “one-hugely-expensive-RAM-one-vote”.

Let surrendering blockchain control to a small fraction of the blockchain participants be a have-to-accept way of life with public blockchain. A so-called Proof-of-Stake (PoS) model of block generation has appeared for improving blockchain efficiency and without wasting mining electricity. The PoS model bases blockchain security on a conventionally meaningful business belief that the richer the more responsible. Let a rich minority criterion qualified node deposit a sum of money, called security stake, in a PoS money staking mechanism. The more amount the participant, aka stakeholder, deposits, the more chance for its turn to generate a block. Thus, the gossip formulation for the PoS model can be referred to as “gossip-on-affluence”, meaning a block qualifies gossiping as long as it origins from a rich stakeholder. The security stake is for punitive confiscation or destroy (in PoS programmers’ jargon, “money slashing”) when a block generator is found responsible for either having composed an erroneous block, or upon a timeout having not conducted its turn of blocking duty. A PoS version of “Ethereum2” \cite{Ethereum2} would be the most eminent PoW-to-PoS switch to take place in a near future of writing this paper.

The job of PoS block generation is easily done by signing a digital signature using a cryptographic (private) key. That of PoS block validation needs also producing a digital signature for proving a permissioned stakeholder’s entitlement. Without these PoS proofs, the communication traffic for PoS blocking and/or validating would spam the network. Long term protection for, and frequent use of, a signature private key form single-point-of-failure (SPOF) challenges for each of the stakeholders. These SPOF points, although being distributed to the stakeholders, combine to amplify the frequency for centralized SPOF states that a PoS blockchain is always in. If a private key can be worth very expensively for a stakeholder who very likely is a business organization investor in a PoS blockchain (venture loving organizations are the targeted participants for a PoS blockchain to recruit), the key has no value for an (in-organization) attacker who maybe managed to access the private key only for to cause disruption to the PoS blockchain. These frequency amplified and centralized SPOF breakdown scenarios for PoS blockchains are at best nothing-at-stake ones, a little bit of irony for PoS’ naming.

There are other disruption complications for a PoS blockchain. E.g., money slashing can hardly be done by a consensus algorithm due to a well-known academic conclusion, named “Fischer-Lynch-Paterson (FLP) Impossibility”~\cite{FLP}, being the following statement. There can never exist a network distributed algorithm for its network distributed executors to input messages and output a consensus decision. If money slashing has ever been in practice by any PoS blockchain, it necessarily is by some unspecified manual operation. Still manual slashers would need to figure out an equitable exchange rate between an “error” and a money amount to slash. The quoted “error” here may well be an alleged one for the money slashers to take pains to discern.

\subsection{Public Blockchain Status Quo Summary}

To this end a summary can be stated as follows. Known public blockchain algorithms for lining up blocks have only used small fractions of the blockchain participants as functional and dependable nodes. The remainder larger fractions of the blockchain participants are only brutally syntax instructed or affluence stipulated to play trivial roles as if they were unintelligent. Low utilization of the blockchain participants has unfortunately wasted the collectively very strong capabilities that a majority of the blockchain participants can aggregate.

\subsection{No Brutal Instruction for Modest Blockchain Participants}

A blockchain should have the following property: A majority set of its participants should not be told, mandated, instructed, stipulated, manipulated, controlled, or even implicitly influenced, by the remainder set, i.e., a minority, of the participants to enter into a collective behavior. This blockchain property motivates intelligent uses of a majority of the blockchain participants. It requires careful algorithmic design. In fact, the PoW formulation of gossip-on-syntax, and the PoS stipulation of gossip-on-affluence, are two examples of brutal mandating a majority of the blockchain participants entering into collective behaviors. The PoW unresourceful nodes are algorithm designed to gossip syntax “correct” blocks, whereas the PoS proletarian nodes are rule stipulated to gossip stakeholders composed blocks. Both suffer undesirable consequences as a result of their unintelligent using a majority of the blockchain participants.

The present paper will take a “gossip-on-semantics” approach to using modest participants in public blockchains. Each modest participant upon receipting a gossip message will perform a calculation using its local information state and reach a semantics decision on whether to update the local information state and forward the gossip message on, or to discard the gossip message.

Without much ado for now, it should already be self-evident that such a local state calculation reached decision can provide a modest blockchain participant with strong resistance to influence, manipulation, control, etc. from computationally powerful and/or monetarily affluent participants.

\subsection{Organization of the Work Presentation}

The remainder of this paper is organized as follows. In Section~2, a novel notion of gossip-on-semantics is formulated to achieve truly honest majority control by most of the blockchain participants. In Section~3, a new blockchain is constructed with reasoning why, how and what it can benefit from the gossip-on-semantics way of honest majority control. In Section~4, a system architecture level of generalization for blockchains is provided with concrete descriptions on its construction. Finally, this paper presentation concludes in Section~5 where a couple of questions regarding well-known blockchain attacks are asked for inviting answers and/or exploration discussions from interested readers.