October 9, 2018 . 12 min read
Crypto Chat #59
On Governance Pt. 2:
Last week I featured Vlad Zamfir’s Governance 101 article, which outlined the various challenges – with emphasis on coordination mechanisms (or lack thereof) – to governing decentralized systems.
Governance 101 was great.
The follow up, My Intentions for Blockchain Governance, is even better, with Vlad setting out various avenues for governance that he wholeheartedly intends to avoid, namely:
1. Autonomous, AI driven blockchains
2. Blockchain Governance Capture via a inevitably plutocratic one-token-one-vote scheme
3. Internet Censorship as Blockchain Governance.
So what alternatives are amenable to Mr. Zamfir?
Blockchain should be governed on a basis of global cooperation between self-selecting members and entities from the global public. If it turns out to be infeasible for us to govern the blockchain voluntarily using a system that is open to global public participation, then I hope that international law will govern the blockchain.
I intend to have blockchains serve as global, public utilities with no owners, and I intend for them to be governed ethically and in a way that puts the interests of the public before the interests of other participants in blockchain governance.
Developers can implement changes, but they (generally) do not want to have the authority to alter balances or to interfere with the execution of smart contracts. I think we need to figure out how we are going to come up with (at least) a process for dealing with malicious users and applications that has the legitimacy required for developers to implement its decisions without having to exercise too much power, but which also cannot be captured.
In an ideal world Vlad’s vision for blockchain governance would be ideal.
Who could possibly rail against the notion of blockchains as global, public utilities governed by a diverse group of ethically-motivated volunteers? Who could possibly disagree that engineers should be relieved of the burden of governance? Who could possibly dismiss the necessity of building mechanisms to deal with malicious behaviour and users?
Probably only the most ardent polemicists, and who cares what they have to say anyway?
And yet, from my perspective, the path to blockchain Utopia is not only fraught with dangerous obstacles but literally non-existent. In today’s increasingly fractured society, who decides the definition of ‘ethical’? Who decides what constitutes ‘malicious behaviour’? How is International Law at all compatible with pseudonymous systems?
In his response to Zamfir’s treatise, Vitalik Buterin articulates a similar viewpoint, taking issue not with the notion of ‘minimizing negative externalities’ but instead ‘the feasibility of the proposed methods’.
Some highlights below:
If hard forks start to be actively used as a mechanism for interfering with “undesired applications” (assuming we can effectively coordinate on what those are), then developers of undesired applications will either (i) move to other blockchains that *do* follow stronger immutability norms, or (ii) modify themselves in ways that allow them to dodge hard forks.
A hard fork is NOT “smart” enough to be able to catch all of these contingencies.
The DAO hard fork gave many people the impression that, if the political will was there, hard forks could be used as a remedial tool...an impression that I think is incorrect.
Loss recovery is better addressed through better secure and user-friendly wallet infrastructure.
The EOS community actually tried to implement what is essentially EIP 867 with the ECAF. This was met with wide disapproval by a community that was in favor of “governed blockchains” from the start, and the ECAF is no longer active.
Most of the “truly dangerous” applications depend heavily on oracles. Centralized oracles can be pressured to lie in limited cases in order to interfere with applications when needed for the social interest, and decentralized oracles (eg. Augur) probably should adopt governance norms much closer to what you advocate to avoid being used for harm.
So what does this all mean for the future of Ethereum’s governance?
My impression is that the ever-nebulous ‘community’ would currently side with Vitalik’s ‘immutability-maximalism’, preferring to outsource formal governance procedures away from the protocol layer to dApps and Layer 2.
And yet, lest we forget, there are still myriad protocol ‘governance’-related decisions to be made before we can get to a fully mature, self-sufficient state, namely regarding the implementations of Casper, Sharding, privacy, monetary policy etc. Indeed, it remains an open question as to whether the core protocol will ever stop innovating, and if so, whether it can do so effectively without a requisite governance structure.
‘No formal governance’ is clearly still a form of governance that needs to be formalized.
But what if, by definition, it can’t be? Where does that leave the future of this industry, of these public utilities that, if all goes well, stand to play a non-insignificant role in the development of global society? What if 'formalized non-formal governance' is the missing ‘middleware’ that ultimately dooms this experiment to fail?
Tune in next week for more rhetorical questions.
Sia, a blockchain for decentralized storage, has hard forked their hashing algorithm in order to render Bitmain and Innosilicoin-produced ASICs useless.
The catch? Following the hard fork, miners will only be able to continue mining using Obelisk hardware. Obelisk miners are manufactured by Sia’s parent company, Nebulous.
I’m yet to wrap my head around this one.
David Vorick, Sia co-founder, said in his announcement that the move is intended to “send a clear message to all future Sia ASIC manufacturers: we will not tolerate an abusive ASIC monopoly” while concurrently moving to monopolize the Sia ASIC landscape.
Of course, one would think and hope that Vorick has Sia's best intentions at heart. But what if he didn't? And even if he did, is this the type of decision making that we want to characterize ostensibly decentralized networks? Did someone say 'single point of failure'?
The Sia team was quick to point out that no one is being forced to accept the hard fork. Indeed, it is the nature of hard forks that they must be accepted by those running nodes.
And yet I don’t buy it.
The ecosystem is still too nascent for non-professional splinter groups to successfully maintain pre-fork chains. From a technical standpoint maintaining a pre-fork chain is non-trivial. From a community development standpoint maintaining a pre-fork chain is non-trivial. Telling people that they have a range of options in the form of non-trivial options is disingenuous.
Moral of the story? Let’s just avoid this nonsense altogether and transition to Proof of Stake.
Most Read From CC#58:
Bitcoin (BTC) continued to bore this week, remaining firmly put within the $6,500-$6,700 range. I would make a joke about Bitcoin being a stable coin but that wouldn’t be funny.
Let’s take a look at volatility with Twitter user @Rptr45:
1. There have only been 5 distinct periods of lower vol than the current regime, with none of those coming since September 2016.
2. BTC has gone from a 90% vol asset for the duration of 2017 to a <50% vol asset over the past 3 months. The vol profile now looks analogous to high beta tech stocks.
3. Expect a big move in the subsequent 6-12 months and know that low vol can persist for longer than you’d think.
For those looking for a better sense of whether BTC is over/undervalued, look no further than the Market Value to Realized Value Ratio (MVRV Ratio) proposed by Murad Mahmudov and David Puell.
Several weeks ago I featured Realized Cap, incorrectly noting that it provides insight into a market’s liquidity (my mistake). Rather, realized cap adjusts for lost coins and coins used for holding, providing a sense as to the sum of entries whereby users began seeing BTC’s potential as a long-term store of value.
MVRV is calculated by dividing market value by realized value on a daily basis.
Any MVRV above 3.7 denotes overvaluation, with a value below 1 denoting undervaluation. Today’s MVRV sits close to 1.
The value of MVRV as a price indicator becomes further apparent when you map MV over RV. When MV sits above RV we can deduce that the network is overvalued based on its stable fair value and vice versa for when MV sits below RV.
Of course, while MVRV could arguably be considered a fundamental indicator (vs. technical) it still doesn’t provide much insight into the fundamental strength of the network in question. Moreover, as with any popular metric, MVRV is likely to be the victim of spoofing if it ever gains significant traction.
For those interested in exploring chain security as it relates to monetary policy, I unbiasedly recommend my own metric, the Fee Ratio Multiple.
Ether (ETH) likewise remains locked in a tight range. All eyes on BTC.
Meanwhile, many congratulations due to the team at Prysmatic Labs, who have released a demo client for Ethereum 2.0, which includes implementations of the beacon chain and validator registration contract.
BarryWhiteHat continues to plow forward with zero-knowledge based research, this week proposing a SNARK-based side chain that could scale up to 17,000 transactions per second.
However, all is not well as analysts have started to notice instances of Spy Mining.
Spy Mining takes place when miners try to find the next block before they have even had time to download and verify the previous block. Instead they simply take the block header from the previous block, insert a coinbase transaction for themselves, and mine an otherwise empty block, thus improving their chances of securing the full block reward.
There are a couple interesting things to be said of Spy Mining.
The first is that a Nash Equilibrium should emerge over time that makes Spy Mining less profitable than regular mining: if all miners start Spy Mining then the block production time margins disappear, resulting in the same proportionate revenue share as before without the additional transaction fee revenue.
The second thing to note is that Spy Mining is not possible under Ethereum’s Proof of Stake (PoS) implementation, as block production is no longer a competition – instead, validators take turns to produce blocks and collect the requisite block rewards and transaction fees. As a result, the immediate equilibrium under PoS is for transactions to be included within blocks.
Taking inspiration from the success of the Brexit referendum, the Federal Reserve Board has invited public comment on the actions it could take to support faster payments in the United States.
Someone tell them about GeminiUSD!
David Swensen, Chief Investment Officer of Yale University’s $30bn endowment, has ostensibly swallowed the ‘red pill’, participating in crypto asset fund Paradigm’s $400m raise. Paradigm is led by Coinbase co-founder Fred Ehrsam, former Sequoia Capital partner Matt Huang, and Charlie Noyes, crypto-wunderkid and former Principal at Pantera Capital.
Swensen is widely regarded as one of the smartest money managers in the world, known for taking long-term view of markets. His interest in the crypto asset markets will likely be noted by other institutional funds, both private and sovereign.
As several analysts have noted, investing in a fund is not the same as market buying crypto assets. Nonetheless, this is a powerful first step in the right direction and another sign of the increasing maturation of this asset class. The only question remaining is which university endowment will be next.
ARKInvest provides novel insight into Bitcoin’s mining pool landscape.
The top four mining pools currently capture more than 51% of the network’s hash power, opening up the network to the threat of a double-spend attack.
However, in the context of the Herfindahl-Hirschman Index, a common heuristic for determining market concentration, today’s mining marketplace remains competitive, with an HHI average since 2013 of 1228.
The crypto asset exchange rat race is heating up: TD Ameritrade has partnered with DRW, Virtue Financial, and Susquehanna International Group to launch a marketplace for individuals and institutions to trade both spot and derivatives tied to digital assets. ErisX will go live in Q2 2019.
ErisX is expected to serve as a strong competitor to New York Stock Exchange-led Baakt, which has similarly intends to offer a trading platform for physically delivered futures.
With the imminent emergence of Baakt and ErisX, the question on everyone’s (maybe not everyone) mind is how industry-native incumbent’s will continue to compete.
Gemini, New York’s leading crypto asset exchange, is clearly up for the fight, this week announcing that they have obtained insurance coverage for the digital assets it holds in custody. Previously, Gemini had only provided FDIC insurance for dollar deposits up to $250,000.
Providing insurance for digital assets effectively mitigates exchange-based counter-party risk, which has incessantly reared its head over recent years through various high profile hacks.
One might now expect:
1. Gemini customers to keep their digital assets on exchange, opening up further liquidity from limit orders
2. Gemini to be well positioned to dominate the increasingly competitive, and likely profitable, Staking-as-a-Service (StaaS) landscape.
Bonus: A staking-implementation framework from Chorus One, a StaaS project led by Epicentre’s Meher Roy.
A heartfelt confession from Angus Champion de Crespigny, formerly of the Ernst & Young Blockchain Division, on how he lost his faith in private blockchains.
The ETHSanFrancisco hackathon has produced two Harberger Tax-inspired projects.
First, what is the Harberger Tax (HT)?
HT is a concept popularized by Glen Weyl in his seminal work, Radical Markets. The intellectual foundations of HT are based on the notion that property rights lead to monopolistic and inefficient markets.
HT is a novel tax system whereby members of a society appraise their own property (among other items) and pay a fixed annual tax based on that valuation. The catch is that property owners must be willing to sell their property at any time to the first bidder.
If a valuation is above the fair value, the property owner will have to pay a disproportionately large amount of tax. If a valuation is set too low, the property owner is at risk of having to sell their property for less than it is worth.
As a result of these mechanisms, it is expected that the Harberger Tax produces fair valuations for property and thus may increase allocative efficiency of resources. The tax itself is paid into a pool, which is then shared equitably amongst the society-in-question's members.
For more on Radical Markets, I recommend buying the book and/or listening to one of Glen Weyl’s lectures.
Back to San Francisco!
Radical Pixels allows users to set pixels on a communal canvas. Users collectively discover the fair utility value of each pixel. Values are taxed and fund the ‘commons’, which is currently set as the Fellowship of Ethereum Magicians, a community of Ethereum developers that handle the EIP process.
Harberger Ads prices ad space using HT. The model allows a website owner (among other advertisement spaces) to monetize their site “without the complications of preset complex advertising pricing models and contracts, and without the need to work with a centralized high-commission marketing platform.”
Bonus: Joey Krug, founder of Augur, casually spun up an Augur wrapper for DyDx Protocol, allowing users to access margin for Augur markets.
This is but one of infinite bi-products that will emerge from the intersection of novel open-source financial protocols. Exciting to say the least.
Musteka is an ultra-light light client for Ethereum produced by the team at MetaMask, which will help relieve reliance on full node providers like Infura.
Similarly to Coda, Musteka turns a users browser into a partial Ethereum node, which then gossips with peers to create a p2p network of full Ethereum nodes.
Moloch is a grant-making decentralized autonomous organization (DAO) built to overcome the ‘tragedy of the commons’ afflicting core protocol development.
By pooling ETH and ERC20 tokens, teams building on Ethereum can collectively fund open-source work that is in their collective interest to see built.
The DAO operates through the submission, voting on, and processing of a series of membership proposals: a $5,000 deposit is required for new membership proposals in order to combat spam.
The voting period includes a 7 day grace period, which protects Moloch members from 51% attacks and supporting proposals they vehemently oppose. Further, the game theory behind the voting process system means that members are disincentivized from voting yes on proposals that they believe will make other members quit the organization.
I am very excited to see how Moloch functions in the wild.
Numerai, a project that crowdsources trading strategy from thousands of data scientists, has announced a new product, Erasure, that will allow users to sell their strategy data directly to hedge funds.
Aptly dubbed Tinder for Markets, Erasure promises to disrupt the data silos currently managed by hedge funds and permit portable reputation for data scientists.
The dirty secret here is that data scientists have to stake Numerai’s native token, NMR, in order to submit predictions. It remains unclear to me as to why this same dynamic could not be replicated with a more liquid, less volatile coin/token like ETH or DAI.
Michael Wellman, a Computer Science professor quoted in the Bloomberg article, captures this sentiment brilliantly with the following line: “You can set up a marketplace without setting up your own economy.”
To make matters worse, NMR has “one of the loosest policies on inflation” of all token-based projects, with all newly issued tokens directed towards the foundation. At current circulating supply NMR has a ~$7m network value and a fully diluted network value of $110m, a multiple of 15.7x.
In other words, NMR is currently structured in a way that is beneficial to the company rather than investors and/or users.
Still stuck on consensus?
Jordan Clifford of Scalar Capital has you covered with this overview of BFT, pBFT, PoW, PoS, dPoS, PoST, and DAG.
Bonus: The ever-brilliant breakdown from the team at Mechanism Labs with various relevant protocol examples.
Bonus Pt. 2: Awesome Cryptoeconomics, a curated list of awesome resources for cryptoeconomics research.