ARK — Not the Arkenstone of Interoperability

17.07.2018

TL;DR. Ecology:3. Technology:4. Decentralization:6. Valuation:4. Rating:4/10

Introduction

ARK took the step in to the public spotlight with a bitcointalk.org thread by user techbytes on the 15th of October 2016. The project gravitates around the ARK cryptocurrency as well as its platform that is primarily focused on mass adoption of blockchain technology. That adoption approach is somewhat reflected in the fact that an extensive signature bounty campaign was initiated immediately. A Twitter campaign was also initiated. To fund the project, an ICO-like Token Exchange Campaign (TEC) was held, mainly benefiting Lisk contributors due to a huge bonus not fully valid for BTC or ETH contributors. As the TEC was held in November 2016, 93.75M tokens seems to have been allocated to investors, and 18.75M to the ARK team. Around 1M tokens were allocated to escrow agents, around 2.5M to bounties, and finally around 9M went to what is called the ArkShield - a strategic reserve fund for the ARK ecosystem. The initial total supply landed on 125M and has since expanded somewhat due to steady issuance in the form of block rewards.

One of the main innovations that is hoped to encourage adoption is something the ARK team calls SmartBridge, a protocol for cross-chain communication, meaning independent blockchains may be able to 'speak' to each other through ARK software. An example of this would be an ARK transaction that triggers a transaction on the Ethereum network from A to B. If this transaction is for a utility token issued on ETH, it should be obvious that it simplifies a great deal for users, who then don't have to buy the different utility tokens manually themselves. ARK could in this way be used as a middle layer. Without digging in to the technicals, BD Ratings understands that the main goal is similar to that of Polkadot, AION or Cosmos, but it is not yet deployed in a decentralized manner.

Being an independent cryptocurrency, the ARK code base is derived from Lisk, Crypti and BitShares. Initially, it seems it was a Lisk copy in all but name (which itself was a fork of Crypti), and the former Lisk developer François-Xavier Thoorens joined the ARK team from the very start. Later on, the code have been more and more rewritten, and will be fully novel once the second version of mainnet is deployed after some more testing. The blockchain utilizes a modified version of BitShares' Delegated Proof of Stake (DPoS) consensus algorithm with not 101 but 51 delegated block producers that validates transactions to earn transaction fees as well as block rewards. Some of the ARK developers seems to have been involved with Lisk delegation earlier and used that experience when constructing the ARK delegation model with its lowered delegator count and rewritten delegator voting rules. One aim seems to be a push for a cartel-resistant DPoS model.

ARK mainnet launched on the 21st of March 2017. In November 2017 it was announced that ARK, instead of establishing a Swiss Foundation like many other projects, created a French Cooperative (SCIC). A second version of mainnet was released for testing on 14th of June 2018 and is expected to within a couple of months be pushed to mainnet. With this new version, ARK will also be able to interpret smart contracts written in Solidity, which possibly will pave way to a more decentralized interoperability on the chain.

Ecology

BD Ratings is of the opinion that the early ARK bounty campaigns were unprofessional. Paying users to change BCT signatures encourages short-sighted profit motives instead of organically building a loyal community with real knowledge of the chain. The Twitter campaign promoted spam, as well as intruded on users to the point that they had to give access to their Twitter accounts to a third party that could update profiles and tweet for them. You can call it bribed sock-puppetry. The fact that the ARK team had way higher bounties for marketing than they had for bug-hunting is quite telling where their focus has been. More price-focus is seen by prominent team membercannabananathat openly mentioned the Lisk contribution bonus would drive up the price of Lisk as around 20% of the total Lisk supply were to hopefully be collected by the Ark team during the TEC. Incidentally, he later regrettedthe TEC tie to Lisk, as Lisk is not as stable as the larger BTC or ETH.

ARK has established bounties also for the purpose of enlarging the community with actual use-cases like clients written in new programming languages, payment plugins, IoT applications and supply chain management. This is a positive thing and how 'ecological' bounties ought to work.

Looking at the on-chain activity that is currently going on, it seems there are very few actual transactions, and much more block rewards that explain the total activity. This is somewhat expected as the point-click-blockchain functionality is not fully deployed yet, and neither is chain interoperability functionality. Low activity is also a sign of the times with the current bear market.

If interoperability through decentralized SmartBridge 'tech' becomes successful, it is obviously a big win for ARK as it could possibly establish itself as the main protocol for cross-blockchain communication before other competitors. A high usage of such a protocol, combined with other factors such as low inflation, a secure and decentralized chain, properly aligned delegate incentives etc, would establish a sound ecology and a base for being a sound value protocol such as Bitcoin or Ethereum. It seems however that there are some practical thresholds for this cross-chain communication. Chains that are to 'speak' with ARK and thus with other chains seems to in some cases have to implement changes in the core code - something which will be difficult to push through on the larger chains. This ought to be solved by modified clients instead of modified core protocols, where for example a modified Ethereum Parity client also stores or listens in on the ARK chain, which makes it able to interact with that. How ARK will act with regards to this issue remains to be seen, and it might be one reason why there seems to have been a small shift away from interoperability development to development that focuses on mainstream adoption or simply rewriting the code base. On Reddit, BCT or on the ARK blog, BD Ratings see no clear description on how the interoperability layer will work with external chains but it will surely be taken in to account in future ARK ratings updates as soon as more is known.

Grade: 3

Reasons: Low on-chain activity. Bad bounty policies. Core functionality still not widely deployed. An apparent shift from interoperability goals.

Technology

ARC is fully open source and is developed by multiple developers. Extensive bug- and developer bounties are open for anyone to participate in, which is a very good thing. They were even extended indefinitely and will run as long as ARK has funds.

The ARC motto of "Point. Click. Blockchain." needs to be carefully interpreted as we have seen the disasters coming from hastily published code (the DAO). ARK obviously has a utility approach to its blockchain; often emphasized is the need for helping blockchain technology reach the mainstream. But what exactly is 'blockchain technology', and do everyone really need it? A blockchain is a database; most if not all value from it derives from decentralization. Any company that deploys its own blockchain (connected to ARK) is not really doing anything revolutionary unless they construct that blockchain to be decentralized and secure as well. A 'blockchainize everything' attitude will probably turn out to be a waste of time for many participants.

Reviewing the ARK team's roadmap (in lack of a better word), it is also here evident that they want to focus on these sandboxed blockchains to solve the scaling problem that Bitcoin and Ethereum sporadically have run in to. Again, while sounding good initially, BD Ratings wonders if not these chains would be centralized in most cases, as they don't utilize the decentralized ARK mainnet directly. The point of an Ethereum ERC20 token for example is that the standard code will run as proclaimed, without interference since the contracts are deployed on Ethereum. If such ERC20-like contracts were to be published on small blockchains that communicates with ARK, they could also be susceptible to changes should a centralized delegate body want it. It might be that the ARK team knows this well, but anyway see this point-click-blockchain approach as something positive for applications that customers accept will be centralized. The plug-and-play approach may be an easy way for businesses to opt-in to the ARK chain for its payment infrastructure.

Another issue with the mass production of blockchains is security. To put it bluntly, BD Ratings thinks that highly valuable blockchains ought to treat their code as if it are to run software for a space rocket, airplane or nuclear power reactor. Finding bugs or incentive weaknesses in the core protocol, as well as a constant push for decentralizing the chain, are magnitudes more important than focusing on a point-click-blockchain philosophy. It should be added that ARK does focus on the core code, as seen in the recent release of ARK Core v2, so we might experience a motto more for marketing rather than something deeply rooted within the team.

The ARC Core v2 that was released recently potentially lifts the maximum transactions per second to 19, which is good for a decentralized setup. As the ARK team outlines in the blog article, any higher number than that means a trade-off on decentralization. It is good that they understand this and are not afraid to defend that position.

Grade: 4

Reasons: Focus on blockchain mass production. Quite high transaction throughput. No decentralized interoperability.

Decentralization

The ARC client has multiple implementations, strengthening the general decentralization of the chain. Among other languages, ARK is implemented on JavaScript, GoLang and Python. Critics of multiple implementations mean that the chain runs a risk of consensus failure if the clients were to differ slightly. From a decentralization point of view however, having core clients in multiple languages means a broader base of developers, less dependencies on specific persons, a more thorough vetting of the core code and possibly a broader base of users that may run the software.

One issue worth mentioning is possible cartel-like groups forming among the 51 delegates. Vitalik Buterin have warned about aspects of DPoS earlier, and there seems to be signs of cartels forming on certain chains like Lisk. Intertwined with this problem is the reality of voter apathy, where regular token holders for multiple reasons don't vote for delegates. This results in active delegates or delegate groups having unproportional powers over who among delegate candidates should be elected. There will always be some forces tugging towards a degree of decentralization since too heavy chain control will lower the value of the token that delegates get paid with, but a cartel like structure seems likely as well. Some evidence for this would be if ARK delegates starts to lower their payouts to delegate voters, which seems to have happened in Lisk. So far however, the structure of ARK seems to have had no implication on centralizing the chain.

Parts of the ARK community showed strong proof of a decentralization mindset when a user proposed that the ARK core team through their money reserve ArkShield should vote in the Ark Community Fund (ACF) as a delegate. The ACF is a community initiative and stand separate from the ARK team. Despite this separation, ARK reddit users rejected this idea as they have a notion that for example the Lisk team is influencing the delegator voting process in a centralized manner. They want the developer team to stay out of the delegator voting process. When looking at the delegator data, it hints that votes are not concentrated to a few delegators, which is logical since voters want to maximize their part of block rewards.

Grade: 6

Reasons: Open source. Decentralization ethos. Still no apparent signs of cartels forming within the 51 delegator group. Multiple client implementations.

Valuation

With under a USD 200M valuation and relatively low inflation, ARK is not in any extreme position when comparing to other interoperability projects like AION with its more than USD 400M valuation. ARK however has diverted from the goal of blockchain interoperability - while AION, Cosmos and Polkadot seems to stay focused on just that. Both AION and Polkadot seems to have large and technically competent teams as well. BD Ratings thinks that the general valuation of interoperability projects will blow up massively in the future as such tokens start forming layers with extremely strong ecology as they tie the popular chains together.

If ARK was to maintain a strong focus on decentralized interoperability technology and let go of the point-click-blockchain nonsense, the overall rating would be higher.

Grade: 4

Reasons: Strong competitors that offers a better risk/reward ratio.