What is Proof of Reputation?

GoChain uses a Proof of Reputation (PoR) consensus model that depends on the reputation of the participants to keep the network secure. A participant (a block signer) must have a reputation important enough that they would face significant financial and brand consequences if they were to attempt to cheat the system. This is a relative concept as almost all businesses would suffer significantly if they were caught attempting to be deceitful, but larger companies will typically have more to lose and thus are chosen over companies with less to use (smaller businesses).

Once a company proves reputation and passes verification, they may be voted into the network as an authoritative node and at this point, it operates like a Proof of Authority network (PoA), where only authoritative nodes can sign and validate blocks (more on PoA) below.

Reputation is critical to a business. A business that acts in an unethical way suffers on many levels including financial (fines, loss of revenue), valuation decreases, branding (distrust) and public relations. Trust is cornerstone to a successful business and once a brand loses trust with their customers, it can take years to recover if ever.

Let’s use the Volkswagen emissions scandal as an example. They operated alone to deceive the public and their own customers. Once caught, it was both a financial and public relations disaster that cost them $25 Billion, a 50% drop in stock price, and their brand lost a lot of trust. Their reputation was put to the test and they paid dearly for it. Proof of Reputation relies on this risk of significant loss to enable trust in the network.

PoR is more attractive to the broader business community than untrusted networks based on PoW or PoS. Risk-adverse companies will be able to rely on known brands in the same way they trust companies like Visa, Inc. or JPMorgan Chase & Co. In PoR, everyone knows exactly who they are trusting with their data.

Reputation is impossible to measure precisely but we are weighing several important metrics in our decision:

1) Market Cap

2) Publicly Traded

3) Brand Significance

Companies with a large market cap have more to lose than small cap companies. We use this metric when evaluating companies because the value of the member companies needs to be in parity with the value of the processing network to disincentivize cheating. We next look at whether a company is publicly or privately held. The effect of reputation affects publicly traded companies more directly and immediately than private companies. An unethical decision on the network can impact a public company’s stock price in minutes.

Finally, we give preference to companies which require strong public brands for their business. For example, a loss of reputation for a company like Coca-Cola or Apple will be more impactful than for a coal mining company.

Our current authorized signers represent over $100 billion in enterprise value across 8 countries. The latest list of Authorized Signers can be viewed on our network statistics page.

Authorized signers are trusted nodes that create blocks, sign them, and distribute them to other nodes. Similar to miners in a Proof-of-Work (PoW) system in that they create blocks and sign them but without the mining cost.

A list of authorized signers is maintained on the blockchain. Only authorized nodes can sign blocks and all blocks are verified that this is true by checking the signer is in the authorized list. The signing algorithm is essentially the same signature algorithm as PoW but with a different set of headers. PoW-specific headers will be removed and additional headers added to enable voting.

Given N authorized signers, a signer may only sign a block every (N/2) + 1. This ensures that someone would need to control > 50% of signers to perform a malicious attack [6].

This means that more than half of the authorized signers on the network would have to collude to attack or control the network. Given that GoChain’s authorized signers have a combined market cap of over $100 billion as of Q1 2020, there is little to no incentive for them to ever consider this.

Additionally, since all signers are known entities, bad actors risk being sued in court or prosecuted criminally. This is all to say that the risk of a 51% attack against the network is extremely unlikely.

By using trusted nodes, transactions can be verified very quickly and the volume of transactions the network can handle increases by orders of magnitude. Similar to systems we use every day that can handle high volumes, like a Google search or Visa payments, those systems can handle high load only because they trust the servers and the network they are running on.

Other factors such as block size and gas limits are artificially low because of the computation power required by PoW. By trusting the consensus nodes we can increase the volume of the network by 100x more than Ethereum can currently handle. Trusted consensus is used outside cryptocurrencies in systems such as etcd which can reach 141,578 transaction per second on a 3-node cluster using modest hardware [?]. Improving throughput is critical as the growth rate of Ethereum is skyrocketing to an unsustainable rate. Ethereum runs at 13 tx/second right now; we are targeting 1,300 tx/second at mainnet launch.

The two major parameters we can tweak are block size (gas limit) and block times. Due to the fact that we have a relatively small set of signers (vs the number of miners in PoW) with known capabilities, we are able to increase the block size drastically and reduce the block times. This alone greatly increases the number of transactions per second. As stated above, the reason you can’t increase block size in PoW is that it makes the hashing algorithm too hard and too expensive. GoChain does not have that limitation.

Using a trusted network of authoritative nodes means that there will be no mining. No mining means there will be no battle between computers to win blocks and therefore no wasted energy. Nodes will only require a small fraction of this energy to process transactions, run smart contracts, and verify blocks.

Ethereum’s estimated energy consumption at the time of this writing is 14 TWh and rising [32]. Assuming 450W power usage per server [31], our 50-node cluster will only use 197.1 MWh or 0.001% of the energy of the Ethereum network.

The storage requirements to store the entire blockchain is quite large– Ethereum size is hundreds of gigabytes and it’s growing rapidly. It can take hours or days to synchronize to a new node which makes it impractical for the average user that just wants to send a transaction. There are newer modes you can run to reduce the size such as fast and light mode, which reduce the size drastically and that is a good step in the right direction.

Since GoChain will be handling 100x more transaction volume, storage becomes much bigger– potentially 100x bigger. As of this writing, Ethereum transactions average 174 bytes and 700,000 transactions occur per day [29] which produce 120.4MB of block data per day or 43.9GB per year. Increasing throughput by 2 orders of magnitude will generate 4.4TB of block data per year. Propagating this across all 23,000 nodes in the Ethereum network [30] would require 101 petabytes.

By limiting the set of nodes operating on the dataset we reduce the network traffic and storage requirements. Checkpointing allows nodes to only store the small fraction of the total blockchain that is required for current processing. Current cloud pricing of $0.022 USD per GB/month [33] makes storing a copy of the blockchain history only $1,161.60 USD per year of block data.

Proof of Authority (PoA) works great in a private network, which is what it was originally intended for, where you know and trust the nodes you add to the network. The advantages gained with PoA are hard to ignore so using it for consensus in a public network is a great thing to solve for. The problem is it doesn’t work in a public setting without something at stake. There are a couple of other companies that are attempting to use PoA in a public network, but there are several issues with these networks.

One such network is using a network of 12 US Notary Publics. While this obviously has a serious centralization problem, the more important problem is the security of the network.

First, there is a disparity between the net worth of the network versus the market cap of the network. This is what Proof of Stake (PoS) attempts to solve. Assuming an average net worth of an individual in the United States is $68,828 [18], the total net worth of the validators is $825,936:

12 ∗ $68, 828 = $825,936

Even if the number of validators increased by an order of magnitude, the total net worth of the validators is a tiny fraction of the $6.8T in transactions processed by Visa, Inc. every year. This disparity introduces a strong incentive for bribery.

Second, validators must post their physical address publicly which opens the potential for intimidation or physical threats. A terrorist organization or rogue state can mount an attack on a large scale financial system by controlling half of these validators.

Finally, most individuals lack the experience and infrastructure to run a secure transaction processing system. This significantly increases the network’s exposure to malicious hacking.

Rating reputation

There are a few factors to take into account:

• Valuation — the higher the value of a company means the more financial risk.
• Brand — companies like Coca Cola or Apple are very high on the brand scale, almost to the point where you could say brand is everything to them. A coal mining company does not have the same kind of brand risk.

Unfortunately these things are not an exact science and need to be evaluated by humans.