in , , ,

On the Subject of Nodes

How blockchain fundamentals are holding Kin back

One of the great debates in blockchain centers around what it means to be decentralized, whether it’s a necessary trait for the application you’re developing, and how a blockchain’s centralization should affect the regulatory status of the cryptoasset(s) running on top of it. While many followers of the blockchain space would identify as decentralization ‘maximalists,’ touting the utopian anarchic virtues of their favorite project, others would argue that centralization is a sliding scale, and that none of the major blockchains in production today are truly as decentralized as they claim to be. Take Bitcoin, for example, where the world’s longest-running and most valuable blockchain has one-third of its hashpower consolidated under one mining pool operator, Bitmain, nearly putting one multi-billion dollar entity in control of the protocol.

Or take two of the other top 10 cryptocurrencies by market capitalization, XRP and Stellar, which have been criticized relentlessly since their inception, for adopting federated Byzantine fault-tolerant models of consensus, as opposed to the more popular ‘trustless’ Proof-of-Work (PoW) or Proof-of-Stake (PoS) models. Where PoW grants power and incentives to the users that run the most powerful machines supporting the network, and PoS grants power and incentives to the wealthiest users supporting the network, federated models optimize for network efficiency by employing ideas like flexible trust and quorum slices. While I’m not savvy enough to go much deeper into consensus theory, the short story is that decentralization maximalists typically aren’t fans of consensus models that require ‘trust’ in any specific node on the network, regardless of the blockchain’s application.

Consensus and Kin

So when the development team behind the Kin cryptocurrency, the Kin Foundation, announced several months ago that it would ‘fork’ the Stellar blockchain to use in parallel with the existing ERC20 token, and that the ONLY node operator at the outset of the new blockchain’s production network would be Kik, the chat app that launched the cryptocurrency, the crypto community’s negative reaction didn’t come as much of a surprise. While Foundation (and Kik) CEO and Founder Ted Livingston later clarified that this wouldn’t be a problem for the blockchain’s end-users because they already trust the app’s developers, and that the network would grow to be more decentralized as additional apps and partners joined the ecosystem, the maximalists didn’t want to hear any of it.

Fast forward five months, after the maximalists and some pre-sale investors have exited a large portion of their Kin position, bringing down the value of the Kin token in conjunction with an extended bear market, and the topic of nodes has surfaced once again. After the release of the atomic swap between ERC20 Kin and the token on the new blockchain was indefinitely delayed, despite already having the technical procedure developed and audited, the project’s followers wondered why. And in recent updates posted by Livingston and community manager Yoel Rivelis, it was revealed that the atomic swap won’t be cleared for release by the Foundation’s legal team until the Kin blockchain federation has assembled at least seven full nodes (operated by independent entities) to run the network. In other words, the Foundation has come to the conclusion that based on their blockchain’s consensus model, they can’t reasonably claim that the blockchain is decentralized enough to link to the ERC20 token until they have seven nodes. I’ll say more on why that probably is, in a moment.

This revelation, which comes at the tail end of a Q3 which saw several major announcements from the Kin Foundation, including the launch of the Kinit survey rewards app, the launch of Kin inside of beauty app Perfect365, and the hiring of former Twitch exec Matt DiPietro as CMO, may also explain other undelivered promises from the project.

The Liquidity

For example, a recurring problem for early adopters and followers has been the available market liquidity of the Kin token. During the token’s “distribution event,” community staff assured prospective investors that they had received indication from multiple exchanges that planned to list the token shortly after launch. The Jaxx wallet even formally announced they would support Kin (and presumably, would offer it on their built in Shapeshift exchange as well). And in the year following those statements, the largest exchange overall that has listed the token, HitBTC, only offers one trading pair, and they aren’t even Kin’s largest exchange by volume.

While Livingston claimed that the Foundation had de-prioritized listing the token on exchanges until there was a call to action for developers and advertisers to buy and sell the asset, according to the community staff, it had become a priority as early as late July. And yet, two months later, the token remains unlisted on any additional major exchanges, despite the project’s high profile and connections to various exchanges at the board of directors level.

While many high-volume cryptocurrency exchanges, both in the United States and around the world do not have very strict criteria for asset listing, other than substantial application fees (to the tune of millions of USD), others hold themselves to a high standard of eligibility based on the project’s fundamentals. Perhaps the most sought-after exchange, for its retail customer base and direct pairs to fiat currency, Coinbase publishes strict eligibility criteria (which they call the Digital Asset Framework), which include concepts such as decentralization and token utility. While the ERC20 version of the Kin asset is fully decentralized (at least, as far as the industry at large is concerned), without at least seven nodes, the Kin blockchain is not. And without the aforementioned atomic swap, the ERC20 Kin asset has arguably zero utility, as it isn’t connected to the app ecosystem where Kin is earned and spent.

The Regulators

Coinbase isn’t the only organization involved in crypto that has a problem with tokens that lack the combination of decentralization and utility. The Securities and Exchange Commission of the United States (SEC) regulates the sale of securities assets to and from citizens of the US. The SEC has recently developed a greater interest in enforcing securities laws in the cryptocurrency space, particularly with respect to tokens sold in an initial coin offering event (ICO), which is how the Kin Foundation raised their development funds. After issuing guidance on non-compliant token sales such as The DAO, and taking enforcement action against sales in-progress like Munchee, and completed token sales like Centra, the SEC has been intensely deliberating amongst themselves and other US regulatory bodies to develop a better framework for how the laws should apply to crypto assets.

https://steemitimages.com/0x0/https://cdn.steemitimages.com/DQmesqJsffFW5afTEKJk3thtZJDEGrB1MyFkwF6BKmpxvJa/image.png
William Hinman, SEC Head of Division of Corporation Finance

In June, the head of the SEC’s Division of Corporation Finance issued an unofficial statement that Bitcoin and Ethereum are not securities, and that the decentralized status of their blockchains was a key determining factor in reaching his conclusion. The SEC, which has had an open investigation into Kin’s token sale since shortly after the conclusion of the sale (alongside investigations into dozens of other tokens), is keeping a close eye on how the Kin Foundation conducts itself, and may be watching how it proceeds towards decentralizing its blockchain, and whether it succeeds in providing meaningful utility to the Kin token. Exchanges based in the US, such as Coinbase, Gemini, Bittrex and others, are likely to be wary of listing assets that may fit the SEC’s fuzzy criteria for a security token. And even if the atomic swap were achieved with only one node running the Kin blockchain, the utility of the token would remain limited to that centralized chain, therein not qualifying the ERC20 asset for real utility.

The Partners

Kin has also had a hard time onboarding major partners, as well as smaller developers in the absence of any programmatic incentives for integrating their cryptocurrency. It is plausible that some app companies, who likely follow the crypto space to some degree, aren’t sold on the idea of implementing a currency over which so much power is held by one or two entities, or which is still lacking so many of the fundamental infrastructural features necessary to make it all ‘work.’ This presents something of a ‘chicken and egg’ scenario, in which exchanges, regulators, apps and investors are hesitant to partner with a blockchain so centralized and feature-incomplete today, which means they won’t run nodes for Kin, which means the blockchain won’t become decentralized enough to unblock those missing features.

Naturally, the ‘seven nodes’ requirement raises several key questions, the first of which we (at NuFi) feel we already know the answer to.

So, why does Kin need at least seven nodes?

I’ll defer to Adam to comment on this:

While it’s important that Kin not be recognized as a security, it is also important that the network not be recognized as a Money Services Business (MSB) by FINCEN.

As we noted within “How Does the Kin Consensus Protocol (KCP) Work?” the Kin network will need a series of federated validator nodes within the network to create balanced quorum slices who can ultimately ensure >66% accepting votes in a network consensus.

So why seven?

Having 7 nodes ensures that no entity controls more than 20% of the vote. Which seems to be the magic number the Kin Foundation believes results in the network not being considered a money services business.

What is so special about 7 nodes and the 20% number?

For the Kin Consensus Protocol to successfully validate a transaction, the network must reach a consensus of >66% of votes. These votes are voted on by overlapping quorum slices, where within each one of those quorum slices a >66% or greater vote must take place.

Since members who follow a quorum slice can have their vote changed by the quorum slice they follow, it actually takes significantly less than 66% of voting power to influence the network.

In fact, if a single actor (entity or user) were to control 20% of the votes in a Federated Byzantine Agreement network (like Kin or Stellar), and all quorum slices within that network overlapped, that it is almost mathematically impossible for the network to vote the same way as the actor who controls 20%. In order to defeat the vote of the 20% actor, every other tangential quorum slice would have to cast their primary vote against the vote of that actor. If any single node within the network that is in a tangential quorum slice were instead to vote in favor of that vote, or vote to accept that vote, or to fail to vote, it would create a domino effect of quorum slices changing their votes due to the level of influence this node has.

The only other way around this would be to isolate that node (or those nodes) in a specific quorum slice, which runs the risk in turn of leaving us with disjointed quorums which result in a broken network.

Given this, anyone who controlled more than 20% of the federated nodes that were default to a Federated Byzantine Agreement network would have the power to:

  1. Always get their vote approved even if it was the initial minority vote.
  2. Hold the network hostage with fractured quorum slices.

In any scenario in which a minority entity (or entities), or a minority of the voting nodes can exercise control over the majority, the network is no longer decentralized and therefore can not be considered exempt as a money services business.

At 7 federated nodes, we are able to create interdependent quorum slices, where no one node has excessive voting power and the majority favor always plays out within these votes.

Who is running a node today?

As mentioned above, the only confirmed node operator to date is Kik Interactive, developers of the Kik chat app (and current parent company of the not-for-profit Kin Foundation). It’s possible that the two other apps that have partnered with the project, IMVU and Perfect365, are also running (or planning to run) nodes, but we aren’t clear on that.

Who will run nodes in the future, and when?

It’s possible that the Foundation has stipulated in its terms of partnership with apps like IMVU and Perfect365 that they are to run full nodes for the network as soon as they’re ready for integration, but this has never been stated. As for smaller developers and followers of the project, the Foundation hasn’t yet made it clear what the operating costs to run a node will be, and they also haven’t published all of the code necessary to get a full node running on the new blockchain. A preliminary documentation FAQ uncovered on Github a month ago estimated a cost of upwards of $2000/month for Kin blockchain nodes. So, in the meantime, Kin may need four additional major app partners to run nodes in order to achieve their goal of seven.

Other blockchains have thousands of nodes. How does the Kin blockchain have fewer than seven nodes after a whole year since raising $100 million and beginning development?

While Kin started raising funds over a year ago, they didn’t shift blockchain strategy to forking Stellar, thus needing to build their own network of nodes, until May. And because these nodes need to be sufficiently independent of each other in order to truly decentralize the network, they can’t just buy Amazon AWS instances around the world and claim decentralization. They also may be bound by what incentives or funding they can offer node operators, as the Foundation paying for others to run nodes could easily raise eyebrows over the threat of collusion. Still, with nearly five months behind them, and a major fundraise completed, it is concerning that the Foundation hasn’t yet been able to onboard more than just two apps to participate in the ecosystem, and presumably, in consensus as well.


Notice:

NuFi.io is an independent publisher aimed at providing quality journalism in the cryptocurrency space. NuFi is not associated with, paid by, or employed by any cryptocurrency project. We rely on subscriptions from readers like you!

To help keep NuFi.io ad-free and creating quality content consider:

[thrive_leads id=’3175′]

Dillon King

Written by Dillon King

Dillon is a NICET-certified fire sprinkler designer by day and a Kin Ambassador by night... and also day. He mined his first Bitcoins in 2013 and promptly lost them in the MTGOX hack. Today, he's a crypto minimalist, but fascinated by the space nonetheless.

Leave a Reply

How does the Kin Consensus Protocol (KCP) work?

What is Metcalfe’s Law and how does it apply to crypto?