[Note: the content below was originally sent to clients and contacts on a private weekly note from Post Oak Labs on July 15, 2018.]
Earlier this week, the Mueller investigation indicted a dozen GRU officers as part of its investigation into the 2016 elections.
In the indictment, the DoJ alleges that these officers used bitcoin to finance some of its operations. This was not limited to simply exchanging bitcoins for services, but also mining them. It is unclear how many bitcoins were mined or which specific mining pool was involved.
If you have read my articles and papers in the past, this is an issue I and others have raised with respect to FMI: the possibility of illicit actors not only running infrastructure but profiting and having the ability to launder proceeds of crime. See “know your miner” in Chapter 3.
For example, in early 2015, after publishing Consensus as a service, several coin journalists chain’splained to me that it is not a problem if North Korea or other actors were running mining pools that regulated institutions used to process financial instruments. This was back in the heyday of maximalism — the view that everything would run on top of Bitcoin, laws be damned.
Turns out, they were probably wrong because financial institutions likely would be violating AML / OFAC / sanctions check requirements if they were sending payments to pools/miners that were sanctioned and/or located in sanctioned countries. Vendors such as Symbiont eventually shifted to non-public chain infrastructure because of this legal constraint too (though they originally started by using Bitcoin).
An ironic thing that most of the ideological bitcoin proponents miss is: that savvy state actors could be using the infrastructure nominally built by anarchists… in order to carry out the state-sponsored activities (such as what the GRU allegedly did, but also less sophisticated operations).
Why did the GRU use bitcoins? According to the indictment, to avoid direct relationships with traditional financial institutions. We can only speculate at this time for other reasons but consider that if you mine a coin, a 3rd party cannot immediately track the purchase of newly minted coins… because they haven’t been purchased. This is one reason why “virgin” coins carry a premium over others. For instance, Blocktrail provided the service (although it has since removed its announcement).
In the future, perhaps mining equipment manufacturers could be subpoenaed to learn their customer list, but keep in mind that there is a secondary market for miners as well, and some of those have ended up in both North Korea and Russia.
Anyone have a guess for how much state-sponsored activity comprises cryptocurrency networks today?
[Disclaimer: The views expressed below are solely my own and do not necessarily represent the views of my clients. I currently own no cryptocurrencies.]
As a follow-on to my previous book reviews, an old colleague lent me a copy of Cryptoassets by Chris Burniske and Jack Tatar.
Overall they have several “meta” points that could have legs if they substantially modify the language and structure of multiple sections in the book. As a whole it’s about on par with the equally inaccurate “Blockchain Revolution” by the Tapscotts.
As I have one in my previous book reviews, I’ll go through and provide specific quotes to backup the view that the authors should have waited for more data and relevant citations as some of their arguments lack definitive supporting evidence.
In short: hold off from buying this edition.
If you’re interested in understanding the basics of cryptocurrencies but without the same level of inaccuracies, check out the new The Basics of Bitcoins and Blockchains by Antony Lewis. And if you’re interested in the colorful background of some of the first cryptocurrency investors and entrepreneurs, check out Digital Gold by Nathaniel Popper.
Another point worth mentioning at the beginning is that there are no upfront financial disclosures by the authors. They do casually mention that they have bitcoin once or twice, but that’s about it.
I think this is problematic because it is not being transparent about potential conflicts of interest (e.g., promoting financial products you may own and hope to see financial gain from).
For instance, we learned that Chris Burniske carried around a lot of USD worth of cryptocurrencies on his phone from a NYTarticle last year:
But a particularly concentrated wave of attacks has hit those with the most obviously valuable online accounts: virtual currency fanatics like Mr. Burniske.
Within minutes of getting control of Mr. Burniske’s phone, his attackers had changed the password on his virtual currency wallet and drained the contents — some $150,000 at today’s values.
Some quick math for those at home. The NYT article above was published on August 21, 2017 when 1 BTC was worth about $4,050 and 1 ETH was worth about $314. So Burniske may have had around 37 BTC or 477 ETH or a combination of these two (and other coins).
That is not a trivial amount of money and arguably should have been disclosed in this book and other venues (such as op-eds and analyst reports).1 In the next edition, they should consider adding a disclosure statement.
A final comment is that several reviewers suggested I modify the review below to be (re)structured like a typical book review — comparing broad themes instead of a detailed dissection — after all who is going to read 38,000+ words?
That is a fair point. Yet because many of the points they attempt to highlight are commonly repeated by promoters of cryptocurrencies, I felt that this review could be a useful resource for readers looking for different perspective to the same topics frequently discussed in media and at events.
Note: all transcription errors are my own.
On p. xi, the authors wrote:
When embarking on our literary journey, we recognized the difficulty in documenting arguably the world’s fastest moving markets. These markets can change as much in a day – up or down – as the stock market changes in a year.
It is only mentioned in passing once or twice, but we know that market manipulation is a real on-going phenomenon. The next edition could include a subsection of cryptocurrencies and ICOs that the CFTC and SEC – among other regulators – have identified and prosecuted for manipulation. More on that later below.
On p. xiv, Brian Kelly wrote in the Foreword
The beauty of this book is that it takes the reader on a journey from bitcoin’s inception in the ashes of the Great Financial Crisis to its role as a diversifier in a traditional investment portfolio.
A small quibble: Satoshi actually began writing the code for Bitcoin sometime in mid-2007, before the GFC took place. It may be a chronological coincidence that it came out when it did, especially since it was supposed to be a payment system, which is just one small function of a commercial bank.23
On p. xv Kelly writes:
As with any new model, there are questions about legality and sustainability, but the Silicon Valley ethos of “break things first, then ask for forgiveness” has found its way to Wall Street.
There are also two problems with this:
Both the SEC and CFTC – among other federal agencies – were set up in the past because of the behavior that Kelly thinks is good: “break things first, then ask for forgiveness” is arguably a bad ethos to have for any fiduciary and prudential organizations.4
Any organization can do that, that’s not hard. Some have gotten away with it more than others. For instance, Coinbase was relatively loose with its KYC / AML requirements in 2012-2014 and has managed to get away with it because it grew fast enough to become an entity that could lobby the government.
On p.xv Kelly writes
“Self-funded, decentralized organizations are a new species in the global economy that are changing everything we know about business.”
In point of fact, virtually all cryptocurrencies are not self-funded. Even Satoshi had some kind of budget to build Bitcoin with. And basically all ICOs are capital raises from external parties. Blockchains don’t run and manage themselves, people do.
On p. xv Kelly writes:
“These so-called fat protocols are self-funding development platforms that create and gain value as applications are built on top.”
The fat protocol thesis has not really born out in reality, more on that in a later chapter below. While lots of crytpocurrency “thought leaders” love to cite the original USV article, none of the platforms are actually self-funded yet. They all require external capital to stay afloat because insiders cash out for real money.5 And because there is a coin typically shoehorned at the protocol layer, there is very little incentive for capable developers to actually create apps on top — hence the continual deluge of new protocols each month — few actors want to build apps when they can become rich building protocols that require coins. More on this later.6
On p. xxii the authors write:
“… and Marc Andreessen developing the first widely used web browser, which ultimately became Netscape.”
A pedantic point: Marc Andreessen was leader of a team that built Mosaic, not to take away from that accomplishment, but he didn’t single handedly invent the web browser. Maybe worth rewording in next edition.
On p. xxiii they write:
Interestingly, however, the Internet has become increasingly centralized over time, potentially endangering its original conception as a “highly survivable system.”
This is a valid point however it glosses over the fact that all blockchains use “the internet” and also — in practice — most public blockchains are actually highly centralized as well. Perhaps that changes in time, but worth looking at “arewedecentralizedyet.”
On p. xxiii they write:
Blockchain technology can now be thought of as a general purpose technology, on par with that of the steam engine, electricity, and machine learning.
This is still debatable. After all, there is no consensus on what “blockchains” are and furthermore, as we have seen in benchmark comparisons, blockchains (however defined) come in different configurations. While there are a number of platforms that like to market themselves as “general purpose,” the fact of the matter is that there are trade-offs based on the user requirements: always ask who the end-users and the use-cases a blockchain was built around are.
On p. xxiv they cite Don and Alex Tapscott. Arguably they aren’t credible people on this specific topic. For example, their book was riddled with errors and they even inappropriately made-up advisors on their failed bid to launch and fund their NextBlock Global fund.
On p. xxiv the authors write:
Financial incumbents are aware blockchain technology puts on the horizon a world without cash – no need for loose bills, brick-and-mortar banks, or, potentially, centralized monetary policies. Instead, value is handled virtually through a system that has no central authority figure and is governened in a centralized and democratic manner. Mathematics force order in the operations. Our life savings, and that of our heirs, could be entirely intangible, floating in a soup of secure 1s and 0s, the entire system accessed through computers and smartphones.
This conflates multiple things: digitization with automation.7 Retail banking has and will continue its march towards full digital banking. You don’t necessarily need a blockchain to accomplish that — we see that with Zelle’s adoption already.8
Also, central banks are well aware that they could have some program adjust interest rates, but discretion is still perceived as superior due to unforeseen incidents and crisis. 9
On p. xxv they write:
The native assets historically have been called cryptocurrencies or altcoins but we prefer the term cryptoassets, which is the term we will use throughout the book.
The term seems to have become a commonly accepted term but to be pedantic: most owners and users do not actually utilize the “cryptography” part — because they house the coins in exchanges and other intermediaries they must trust (e.g., the user doesn’t actually control the coin with a private key).10
And as we continue to see, these coins are easily forkable. You can’t fork physical assets but you can fork and clone digital / virtual ones. That’s a separate topic though maybe worth mentioning in the next edition.
On p. xxv they write:
It’s early enough in the life of blockchain technology that no books yet have focused solely on public blockchains and their native cryptoassetss from the investing perspective. We are changing that because investors need to be aware of the opportunity and armed both to take advantage and protect themselves in the fray.
Might be worth rewording because in Amazon there are about 760 books that pop up when “investing in cryptocurrencies” is queried. And many of those predate the publication of Cryptoassets. For instance, Brian Kelly, who wrote the Forward, published a fluffy coin promotion book a few years ago.
On p. xv they write:
Inevitably, innovation of such magnitude, fueled by the mania of making money, can lead to overly optimistic investors. Investors who early on saw potential in Internet stock encountered the devastating dot-com bubble. Stock in Books-A-Million saw its price soar by over 1,000 percent in one week simply by announcing it had an updated website. Subsequently, the price crashed and the company has since delisted and gone private. Other Internet-based high flyers that ended up crashing include Pets.com, Worldcom, and WebVan. Today, none of those stocks exist.
So far, so good, right?
Whether specific cryptoassets will survive or go the way of Books-A-Million remains to be seen. What’s clear, however, is that some will be big winners. Altogether, between the assets native to blockchains and the companies that stand to capitalize on this creative destruction, there needs to be a game plan that investors use to analyze and ultimately profit from this new investment theme of cryptoassets. The goal of this book is not to predict the future – it’s changing too fast for all but the lucky to be right- but rather to prepare investors for a variety of futures.
Even for 2017 when the book was publish, this statement is lagging a bit because there were already several “coin graveyard” sites around. Late last month Bloomberg ran a story: more than 1,000 coins are dead according to Coinopsy.
It is also unclear, “that some will be big winners.” Maybe modify this part in the next version.11
On p. xxvi they write:
“One of the keys to Graham’s book was always reminding the investor to focus on the inherent value of an investment without getting caught in the irrational behavior of the markets.”
There is a healthy debate as to whether cryptocurrencies and “cryptoassets” have any inherent value either.12 Arguably most coins traded on a secondary market depend on some level of ‘irrational’ behavior: many coin holders have short time horizons and want someone else to help push up the price so they can eventually cash out.13
On p. 3 they write:
In 2008, Bitcoin rose like a phoenix from the ashes of near Wall Street collapse.
The Bitcoin whitepaper came out on October 31, 2008 and Satoshi later said that he/she had spent the previous 18 months coding it first before writing it up in a paper. The authors even discuss this later on page 7. Worth removing in next edition.
On p. 3 they write:
Meanwhile, Bitcoin provided a system of decentralized trust for value transfer, relying not on the ethics of humankind but on the cold calculation of computers and laying the foundation potentially to obviate the need for much of Wall Street.
This is not quite true. At most, Bitcoin as it was conceived and as it is today — is a relatively expensive payment network that doesn’t provide definitive settlement finality.15 Banks as a whole, do more than just handle payments — they manage many other services and products. So the comparison isn’t really apples-to-apples.
Note: banks again as a whole spend more on IT-related systems than nearly any other vertical — so there is already lots of “cold calculation” taking place within each of these financial institutions.16
Now, maybe blockchain-related ideas replace or enhance some of these institutions, but it is unlikely that Bitcoin itself as it exists today, will do any of that.
On p. 5 they write:
What people didn’t realize, including Wall Street executives, was how deep and interrelated the risks CMOs posed were. Part of the problem was that CMOs were complex financial instruments supported by outdated financial architecture that blended and analog systems.
There were a dozen plus factors for how and why the GFC arose and evolved, but “outdated financial infrastructure” isn’t typically at the top of the list of culprits. Would blockchain-like systems have prevented the entire crisis? There are lots of op-eds that have made the claim, but the authors do not really provide much evidence to support the specific “blended” argument here. Perhaps worth articulating in its own section next time.
Speaking of which, also on p. 5 they write:
Whether as an individual or an entity, what’s now clear is that Satoshi was designing a technology that if existent would have likely ameliorated the toxic opacity of CMOs. Due of the distributed transparency and immutable audit log of a blockchain, each loan issued and packaged into different CMOs could have been documented on a single blockchain.
This seems to conflate two separate things: Bitcoin as Satoshi originally designed it in 2008 (for payments) and later what many early adopters have since promoted it as: blockchain as FMI.18
Bitcoin was (purposefully) not designed to do anything with regulated financial instruments, it doesn’t meet the PFMI requirements. He was trying to build e-cash that didn’t require KYC and was difficult to censor… not ways to audit CMOs. If that was the goal, architecturally Bitcoin would likely look a lot different than it did (for instance, no PoW).
And lastly on p. 5 they write:
This would have allowed any purchaser to view a coherent record of CMO ownership and the status of each mortgage within. Unfortunately, in 2008 multiple disparate systems – which were expensive and therefore poorly reconciled – held the system together by digital strings.
Interestingly, this is the general pitch for “enterprise” blockchains: that with all of the disparate siloed systems within regulated financial institutions, couldn’t reconciliation be removed if these same systems could share the same record and facts on that ledger? Hence the creation of more than a dozen enterprise-focused “DLT” platforms now being trialed and piloted by a slew of businesses.
This is briefly discussed later but the next edition could expand on it as the platforms do not need a cryptocurrency involved.19
On p. 7 they write:
By the time he released the paper, he had already coded the entire system. In his own words, “I had to write all the code before I could convince myself that I could solve every problem, then I wrote the paper.” Based on historical estimates, Satoshi likely started formalizing the Bitcoin concept sometime in late 2006 and started coding around May 2007.
Worth pointing out that Hal Finney and Ray Dillinger — and likely several others – helped audit the code and paper before any of it was publicly released.
On p. 8 they write:
Many years later people would realize that one of the most powerful use cases of blockchain technology was to inscribe immutable and transparent information that could never be wiped from the face of digital history and that was free for all to see.
There appears to be a little hyperbole here.
Immutability has become a nebulous word that basically means many different things to everyone. In practice, the only thing that is “immutable” on any blockchain is the digital signature — it is a one-way hash. All something like proof-of-work or proof-of-stake does are decide who gets to vote to append the chain.
Also, as mentioned above, there are well over 1,000 dead coins so it is actually relatively common for ‘digital history’ to effectively be wiped out.
On p. 8 they write:
A dollar invested then would be worth over $1 million by the start of 2017, underscoring the viral growth that the innovation was poised to enjoy.
Hindsight is always 20-20 and the wording above seems to be a little unclear with dates. As often as the authors say “this is not a book endorsing investments,” other passages seem do just the opposite: by saying how smart you would’ve been if you had bought at a relative low, during certain (cherry picked) dates.
Also, what viral growth? What are the daily active and monthly active user numbers they think are occurring on these chains? In later chapters, they do cite some on-chain activity but this version lacks specific DAU / MAU that would strengthen their arguments.20 Worth revisiting in the next edition.
On p. 8 they write:
Diving deeper into Satoshi’s writings around the time, it becomes more apparent that he was fixated on providing an alternative financial system, if not a replacement entirely.
This isn’t quite right. The very first thing Satoshi tried to build was a marketplace to play poker which was supposed to be integrated with the original wallet itself.
A lot of the talk about “alternative financial system” is arguably revisionist propaganda from folks like Andreas Antonopoulos who have tried to rewrite the history of Bitcoin to conform with their political ideology.
Readers should also check out MojoNation and what that team tried to accomplish.
On p. 9 they write;
While Wall Street as we knew it was experiencing an expensive death, Bitcoin’s birth cost the world nothing.
There are at least two issues that can be modified for the future:
Wall Street hasn’t died, maybe parts of the financial system are replaced or removed or enhanced, but for better and worse almost 10 years since the collapse of Lehman, the collective financial industry is still around.
Bitcoin cost somebody something, there were opportunity costs in its creation. And as we now know: the ongoing environmental impact is enormous. Yet promoters typically handwave it away as a “cost of doing anarchy.” Thus worth rewording or removing in the next edition.
On p. 9 they also wrote:
It was born as an open-source technology and quickly abandoned like a motherless babe in the world. Perhaps, if the global financial system had been healthier, there would have been less of a community to support Bitcoin, which ultimately allowed it to grow into the robust and cantankerous toddler that it currently is.
This prose sounds like something from Occupy Wall Street and not something found in literature to describe a computer program.
For example, there are lots of nominally open source blockchains, hundreds or maybe even thousands.21 That’s not very unique (it is kind of expected since there is a financial incentive to clone them).
And again, Satoshi worked on it for at least a couple years. It’s not like he/she dropped it off at an orphanage after immediate gestation. This flowery wording acts like a distraction and should be removed in the next edition.
On p. 12 they write:
Three reputable institutions would not waste their time, nor jeopardize their reputations, on a nefarious currency with no growth potential.
There is a bit of an unnecessary attitude with this statement. The message also seems to go against the criticism earlier in the book towards banks. For instance, the first chapter was critical of the risks that banks took leading up to the GFC. You can’t have it both ways. In the next edition, should either remove this or explain what level or risk is appropriate.
Also, what is the “growth potential” here? Do the authors mean the value of a coin as measured in real money? Or actual usage of the network?
Lastly, the statement above equates the asset value growth (USD value increases) with a bank’s interest. Bank’s do not typically speculate on the price, they usually only care about volumes which make revenues. A cryptocurrency could go to $0.01 for all they care; and if people want to use it then they could consider servicing it provided the bank sees an ability to make money. For example, UK banks did not abandon the GBP even though it lost 20% of its value in 2016 following the Brexit referendum.
On p. 12 they write:
Certainly, some of the earliest adopters of Bitcoin were criminals. But the same goes for most revolutionary technologies, as new technologies are often useful tools for those looking to outwit the law.
This is a “whataboutism” and is actually wrong. Satoshi specifically says he/she has designed Bitcoin to route around intermediaries (like governments) and their ability to censor. It doesn’t take too much of a stretch to get who would be initially interested in that specific set of payment “rails” especially if there is no legal recourse.22
On p. 12 they also write:
We’ll get into the specific risks associated with cryptoassets, including BItcoin, in a later chapter, but it’s clear that the story of bitcoin as a currency has evolved beyond being solely a means of payment for illegal goods and services. Over 100 media articles have jumped at the opportunity to declare bitcoin dead, and each time they have been proven wrong.
The last sentence has nothing to do with the preceding sentence, this is a non sequitur.
Later in the book they do talk about other use cases but the one that they don’t talk about much is how — according to analytics — the majority of network traffic in 2017 was users moving cryptocurrencies from one exchange to another exchange.
For example, about a month ago, Jonathan Levin from Chainalysis did an interview and mentioned that:
So we can identify, it is quite hard to know how many people. I would say that 80% of transactions that occur on these cryptocurrency ledgers have a counterparty that is a 3rd party service. More than 80%.
Maybe mention in the second edition: the unintended ironic evolution of Bitcoin has had… where it was originally designed to route around intermediaries and instead has evolved into an expensive permissioned-on-permissionless network.23
On p. 13 they write:
It operates in a peer-to-peer manner, the same movement that has driven Uber, Airbnb, and LendingClub to be multibillion-dollar companies in their own realms. Bitcoin lets anyone be their own bank, putting control in the hands of a grassroots movement and empowering the globally unbanked.
Not quite. For starters: Uber, Airbnb, and LendingClub all act as intermediaries to every transaction, that’s how they became multibillion-dollar companies.
Next, Bitcoin doesn’t really let anyone be their own bank because banks offer a lot more products and services beyond just payments. At most, Bitcoin provides a way of moving bitcoins you control to someone else’s bitcoin address (wallet). That’s it.24
And there is not much evidence that Bitcoin or any cryptocurrency for that matter, has empowered many beyond relatively wealthy people in developed or developing countries. There have been a few feel-good stories about marginalized folks in developing countries, but those are typically (unfortunately) one-off theatrics displaying people living in squalor in order to promote a financial product (coins). It would be good to see more evidence in the next edition.
For more on this topic, recommend listening to LTB episode 133 with Richard Boase.
On p. 13 they write:
Decentralizing a currency, without a top-down authority, requires coordinated global acceptance of a shared means of payment and store of value.
Readers should check out “arewedecentralizedyet” which illustrates that nearly all cryptourrencies in practice have some type of centralized, top-down hierarchy as of July 2018.
On p. 13 they write:
Bitcoin’s blockchain is a distributed, cryptographic, and immutal database that uses proof-of-work to keep the ecosystem in sync.
Worth modifying because the network is not inherently immutable — only digital signatures have “immutability.”25 Also, proof-of-work doesn’t keep any “ecosystem” in sync. All proof-of-work does is determine who can append the chain. The “ecosystem” thing is completely unrelated.
On p. 15 they write:
There is no subjectivity as to whether a transaction is confirmed in Bitcoin’s blockchain: it’s just math.
This isn’t quite true.26 Empirically, mining pools have censored transactions for various reasons. For example, Luke-Jr (who used to run Eligius pool) thinks that SatoshiDice misuses the network; he is also not a fan of what OP_RETURN was being used for by Counterparty.
Also, humans control pools and also manage the code repositories… blockchains don’t fix and run themselves. So it’s not as simple as: “it’s just math.”
On p. 15 they write an entire paragraph on “immutability”:
The combination of globally distributed computers that can cryptographically verify transactions and the building of Bitcoin’s blockchain leads to an immutable database, meaning the computers building Bitcoin’s blockchain can only do so in an append only fashion. Append only means that information can only be added to Bitcoin’s blockchain over time and cannot be deleted – an audit trail etched in digital granite. Once information is confirmed in Bitcoin’s blockchain, it’s permanent and cannot be erased. Immutability is a rare feature in a digital world where things can easily be erased, and it will likely become an increasingly valuable attribute for Bitcoin over time.
This seems to have a few issues:
As mentioned several times before in this review, “immutability” is only a characteristic of digital signatures, which are just one piece of a blockchain. Recommend Gwern’s article entitled “Bitcoin-is-worse-is-better” for more details.
Empirically lots of blockchains have had unexpected and expected block reorgs and hard forks, there is nothing fundamental to prevent this from happening to Bitcoin. See this recent article discussing a spate of attacks on various PoW coins: Blockchain’s Once-Feared 51% Attack Is Now Becoming Regular
The paragraph above ignores the reality that well over 1,000 blockchains are basically dead and Bitcoin itself had a centralized intervention on more than one occasion, such as the accidental hardfork in 2013 and the Bitcoin block size debate from 2015-2018.
On p. 15 they introduce us to the concept of proof-of-work but don’t really explain its own origin as a means of combating spam email in the 1990s.
For instance, while several Bitcoin evangelists frequently (mistakenly) point to Hashcash as the original PoW progenitor, that claim actually legitimately goes to a 1993 paper entitled Pricing via Processing or Combatting Junk Mail by Cynthia Dwork and Moni Naor. There are others as well, perhaps worth adding in the next edition.27
On p. 16 they write:
Competition for a financial rewad is also what keeps Bitcoin’s blockchain secure. If any ill-motivated actors wanted to change Bitcoin’s blockchain, they would need to compete with all the other miners distributed globally who have in total invested hundreds of millions of dollars into the machinery necessary to perform PoW.
This is only true for a Maginot Line attack (e.g., attack via hashrate).28 There are cheaper and more effective out of band attacks, like hacking BGP or DNS. Or hacking into intermediaries such as exchanges and hosted wallets. Sure the attacker doesn’t directly change the blocks, but they do set in motion a series of actions that inevitably result in thefts that end up in blocks further down the chain, when the transactions otherwise wouldn’t have taken place.
On p. 17 they write:
The hardware runs an operating system (OS); in the case of Bitcoin, the operating system is the open-source software that facilitates everything described earlier. This software is developed by a volunteer group of developers, just as Linux, the operating system that underlies much of the cloud, is maintained by a volunteer group of developers.
This isn’t quite right in at least two areas:
Linux is not financial market infrastructure software; Bitcoin originally attempted to be at the very least, a payments network. There are reasons why building and maintaining FMI is regulated whereas building an operating system typically isn’t. It has to do with risk and accountability when accidents happen. That’s why PFMI exists.
On p. 17 they discuss “private versus public blockchains”:
The difference between public and private blockchains is similar to that between the Internet and intranets. The internet is a public resource. Anyone can tap into it; there’s not gate keepers.
This is wrong. All ISPs gate their customers via KYC. Not just anyone can set up an account with an ISP, in fact, customers can and do get kicked off for violating Terms of Service.
“The Internet” is just an amalgamation of thousands of ISPs, each of whom have their own Terms of Service. About a year ago I published an in-depth article about why this analogy is bad and should not be use: Intranets and the Internet.
On p. 18 they write:
Public systems are ones like BItcoin, where anyone with the right hardware and software can connect to the network and access the information therein. There is no bouncer checking IDs at the door.
This is not quite right. The “permissionless” characteristic has to do with block making: who has the right to vote on creating/adding a new block… not who has the ability to download a copy of the blockchain. Theoretically there is no gatekeeper for block making in Bitcoin. Although, there are explicit KYC checks on the edges (primarily at exchanges).
In practice, the capital and knowledge requirements to actually create a new mining pool and aggregate hashpower that is sufficiently capable of generating the right hash and “winning” the scratch-off lottery is very high, such that on a given month just 20 or so block makers are actually involved.29
While there is no strict permissioning of these participants (some come and go over the years), it is arguably a de facto oligopoly based on capital expenditures and not some type of feel-good meritocracy described in this book.30
On p. 18 they write:
Private systems, on the other hand, employ a bouncer at the door. Only entities that have the proper permissions can become part of the network. These private systems came about after Bitcoin did, when enterprises and businesses realized they liked the utility of Bitcoin’s blockchain, but weren’t comfortable or legally allowed to be as open with he information propagated among public entities.
This is not nuanced enough. What precisely is permissioned on a “permissioned” blockchain is: who gets to do the validation.
While there are likely dozens of “permissioned” blockchain vendors — each of which may have different characteristics — the common one is that the validators are KYC’ed participants. That way they can be held accountable if there is a problem (like a fork).
For example, many enterprises and businesses tried to use Bitcoin, Ethereum, and other cryptocurrencies but because these blockchains were not built with their use cases in mind, unsurprisingly found that they were not a good fit.
This is not an insult: the “comfort” refrain is tiring because there have been a couple hundred proofs-of-concept on Bitcoin – and variants thereof – to look into whether those chains were fit-for-purpose… and they weren’t. This passage should be reworded in the second edition.
On p. 18 they write:
Within financial services, these private blockchains are largely solutions by incumbents in a fight to remain incumbents.
Maybe that is the motivation of some stakeholders, but I don’t think I’ve ever been in a meeting in which the participants (banks) specifically said that. It would be good to have a citation added in the next edition. Otherwise, as Hitchens said: what can be presented without evidence can be dismissed without evidence.
On p. 18 they write:
While there is merit to many of these solutions, some claim the greatest revolution has been getting large and secretive entities to work together, sharing information and best practices, which will ultimately lower the cost of services to the end consumer. We believe that over time the implementation of private blockchains will erode the position held by centralized powerhouses because of the tendency toward open networks. In other words, it’s a foot in the door for further decentralization and the use of public blockchains.
This is a “proletariat” narrative that is frequently used in many cryptocurrency books. While there is a certain truth to an angle – collaboration of regulated entities that normally compete with one another – many of the vendors and platforms that they are piloting are actually “open.”
Which brings up the euphemism that some vocal public blockchain promoters like to stake a claim in… the ill-defined “open.” For instance, coin lobbyists such as Coin Center and coin promoters such as Andreas Antonopoulos regularly advertise that they are experts and advocates of “open” chains but their language is typically filled with strawmen.
For instance, enterprise-specific platforms such as Fabric, Corda, and Quorum are all open sourced, anyone can download and run the code without the permission of the vendors that contribute code or support to the platforms.
Thus, it could be argued that these platforms are “open” too… which they are.
But it is highly unlikely that ideological advocates would ever defend or promote these platforms, because of their disdain and aversion to platforms built by financial organizations. 31
Lastly, this “foot in the door” comment comes in all shapes and sizes; sometimes coin promoters use “Trojan horse” as well. Either way it misses the point: enterprises will use technology that solves problems for them and will not use technology that doesn’t solve their problem.
In practice, most cryptocurrencies were not designed – on purpose – to solve problems that regulated institutions have… so it is not a surprise they do not use coin-based platforms as FMI. It has nothing to do with the way the coin platforms are marketed and everything to do with the problems the coins solve.
On p. 19 they write:
Throughout this book, we will focus on public blockchains and their native assets, or what we will define as cryptoassets, because we believe this is where the greatest opportunity awaits the innovative investor.
The authors use the term “innovative investor” a dozen or more times in the book. It’s not a particularly useful term.32
Either way, later in the book they don’t really discuss the opportunity cost of capital: what are the tradeoffs of an accredited investor who puts their money long term into a coin versus buys equity in a company. Though, to be fair, part of the problem is that most of the companies that actually have equity to buy, do not publish usage or valuation numbers because they are still private… so it is hard to accurately gauge that specific trade-off.33
On p. 19 they write about Bitcoin maximalism (without calling it that):
We disagree with that exclusive worldview, as there are many other interesting consensus mechanisms being developed, such as proof-of-stake, proof-of-existence, proof-of-elapsed time, and so on.
Proof-of-existence is not a consensus mechanism. PoE simply verifies the existence of a file at a specific time based on a hash from a specific blockchain. It does not provide consensus. This should be reworded in the next edition.
Furthermore, neither proof-of-stake or proof-of-elapsed-time are actual consensus mechanisms either… they are vote ordering mechanisms — a mechanism to prevent or control sybil attacks. 34 See this excellent thread from Emin Gun Sirer.
On p. 22 they write:
Launched in February 2011, the Silk Road provided a rules-free decentralized marketplace for any product one could imagine, and it used bitcoin as the means of payment.
This isn’t quite true. Certain guns and explosives were considered off-limits and as a result “The Armory” was spun off.
On p. 22 they write:
Clearly, this was one way that Bitcoin developed its dark reputation, though it’s important to know that this was not endorsed by Bitcoin and its development team.
Isn’t Bitcoin — like all cryptocurrences — supposed to be decentralized? So how can there be a singular “it” to not endorse something?35
On p. 22 they write:
The drivers behind this bitcoin demand were more opaque than the Gawker spike, though many point to the bailout of Cyprus and the associated losses that citizens took on their bank account balances as the core driver.
This is mostly hearsay as several independent researchers have tried to identify the actual flows coming into and going out of Cyprus that are directly tied to cryptocurrencies and so far, have been unable to.36
On p. 23 they write about Google Search Trends:
We recommend orienting with this tool even beyond cryptoassets, as it’s a fascinating window into the global mesh of minds.
Incidentally, despite the authors preference to the term “cryptoassets” — according to Google Search Trends, that term isn’t frequently used in search’s yet.
This diversity has led to tension among players as some of these cryptoassets compete, but this is nothing like the tension that exists between Bitcoin and the second movement.
Another frequent name typically used to call “the second movement” was Bitcoin 2.0.
For example, back in 2014 and 2015 I interviewed a number of project organizers and attempted to categorize them into buckets, including things like “commodities” and “assets.” See for instance my guest presentation in 2014 at Plug and Play: (video) (slides).
This label isn’t frequently used as much anymore, but that’s a different topic entirely.
On p. 25 they write an entire section entitled: Blockchain, Not Bitcoin
The authors stated:
Articles like one from the Bank of England in the third quarter of 2014 argued, “The key innovation of digital currencies is the ‘distributed ledger,’ which allows a payment system to operate in an entirely decentralized way, without intermediaries such as banks. In emphasizing the technology and not the native asset, the Bank of England left an open question whether the native asset was needed
The term blockchain, independent of Bitcoin, began to be used more widely in North America in the fall of 2015 when two prominent financial magazines catalyzed awareness of the concept.
Let’s pull apart the problems here.
First, the “blockchain not bitcoin” mantra was actually something that VCs such as Adam Draper pushed in the fall of 2015.
For instance, in an interview with Coindesk in October 2015 he said:
“We use the word blockchain now. I say bitcoin, and they think that’s the worst thing ever. It just feels like they put up a guard. Then, I switch to blockchain and they’re very attentive and they’re very interested.”
Draper seems ambivalent to the change, though he said he was initially against using it, mostly because he believes it’s superficial. After all, companies that use the blockchain as a payments rail, the argument goes, still need to interface with its digital currency, which is the mechanism for transactions on the bitcoin blockchain.
“When we talk about blockchain, I mean bitcoin,” Draper clarifies. “Bitcoin and the blockchain are so interspersed together, the incentive structure of blockchain is bitcoin.”
Draper believes it’s mostly a “vernacular change”, noting the ecosystem has been through several such transitions before. He rifles off the list of terms that have come and gone including cryptocurrency, digital currency and altcoin.
“It’s moved from bitcoin to blockchain, which makes sense, it’s the underlying tech of all these things,” he added. “I think in a lot of ways blockchain is FinTech, so it will become FinTech.”
If you’re looking for more specific examples of companies that began using “blockchain” as a euphemism for “bitcoin” be sure to check out my post: “The Great Pivot.”
The authors also fail to identify that there were lots of early stage vendors and entrepreneurs working in the background on educating policy makers and institutions on what the vocabulary was and how the various moving pieces worked throughout 2015.
Check out my own paper covering this topic and a handful of vendors in April 2015: Consensus-as-a-service. This paper has been cited dozens of times by a slew of academics, banks, regulators, and so forth. And contra Draper: you don’t necessarily need a coin or token to incentivize participants to operate a blockchain.37
On p. 26 they write:
A private blockchain is typically used to expedite and make existing processes more efficient, thereby rewarding the entities that have crafted the software and maintain the computers. In other words, the value creation is in the cost savings, and the entities that own the computers enjoy these savings. The entities don’t need to get paid in a native asset as reward for their work, as is the case with public blockchains.
First, not all private blockchains are alike or commoditized.
Two, this statement is mostly true. At least those were the initially pitches to financial institutions. Remember the frequently cited Oliver Wyman / Santander paper from 2015? It was about cost savings. Since then, the story has evolved to also include revenue generation.
For more up-to-date info on the “enterprise” blockchain world, recommend reading:
On the other hand, for Bitcoin to incentivize a self-selecting group of global volunteers, known as miners, to deploy capital into the mining machines that validate and secure bitcoin transactions, there needs to be a native asset that can be paid out to the miners for their work. The native asset builds out support for the service from the bottom up in a truly decentralized manner.
This may have been true in January 2009 but is not true in July 2018. There are no “volunteers” in Bitcoin mining as running farms and pools have become professionalized and scaled in industrial-sized facilities.
Also, that last sentence is also false: virtually every vertical of involvement is dominated by centralized entities (e.g., exchanges, hosted wallets, mining manufacturing, etc.).
On p. 27 they write:
Beyond questioning the need for native cryptoassets – which would naturally infuriate communities that very much value their cryptoassets – tensions also exist because public blockchain advocates believe the private blockchain movement bastardizes the ethos of blockchain technology. For example, instead of aiming to decentralize and democratize aspects of the existing financial services, Masters’s Digital Asset Holdings aims to assist existing financial services companies in adopting this new technology, thereby helping the incumbents fight back the rebels who seek to disrupt the status quo.
Ironically, virtually all major cryptocurrency exchanges now have institutional investors and/or partnerships with regulated financial institutions.38 Like it or not, but the cryptocurrency world is deep in bed with the very establishment that it likes to rail at on social media.
Also, Bitcoin again is at most a payments network and does not actually solve problems for existing financial service providers on their many other lines of business.
On p. 27 they write:
General purpose technologies are pervasive, eventually affecting all consumers and companies. They improve over time in line with the deflationary progression of technology, and most important, they are a platform upon which future innovations are built. Some of the more famous examples include steam, electricity, internal combustion engines, and information technology. We would add blockchain technology to this list. While such a claim may appear grand to some, that is the scale of the innovation before us.
If you’re not familiar with hyperbole and technology, I recommend watching and reading the PR for the Segway when it first came out. Promoters and enthusiasts repeatedly claimed it would change the way cities are built. Instead, it is used as a toy vehicle to shuffle tourists around at national parks and patrol suburban malls.
Maybe something related to “blockchains” is integrated into various types of infrastructure (such as trade finance), but the next edition should provide proof of some actual user adoption.
For example, the authors in the following paragraph say that “public blockchains beyond Bitcoin that are growing like gangbusters.”
Which ones? In the approximately 9 months since this book was published, most “traction” has been issuing ICOs on these public blockchains. Currently the top 3 Dapps at the time of this writings, run decentralized exchanges… which trade ICO tokens. Now maybe that changes, that is totally within the realm of possibility.39 But let’s take the hype down a few notches until consistent measurable user growth is observed.
On p. 28 they write:
The realm of public blockchains and their native assets is most relevant to the innovative investor, as private blockchains have not yielded an entirely new asset class that is investable to the public.
The wording and attitude should be changed for the next edition. This makes it sound as if the only real innovation that exists are network-based coins that a group of issuers continually create and that you, the reader, should buy.
By downplaying opportunities being tackled by enterprise vendors, the statement glosses over the operating environment enterprise clients reside in and how they must conduct unsexy due diligence and mundane requirements gathering because they have to follow laws and regulations otherwise their customers won’t use their specific platforms.
These same vendors could end up “tokenizing” existing financial instruments, it just takes a lot longer because there are real legal consequences if something breaks or forks.40
On p. 28 and 29 they ask “where is blockchain technology in the hype cycle.”
This section could be strengthened by revisiting and reflecting on the huge expectations that these coin projects have raised and were raising at the time the book was first being written. How were expectations eventually managed?
Specifically, on p. 29 they write:
While it’s hard to predict where blockchain technology currently falls on Gartner’s Hype Cycle (these things are always easier in retrospect), we would posit that Bitcoin is emerging from the Trough of Disillusionment. At the same time, blockchain technology stripped of native assets (private blockchain) is descending from the Peak of Inflated Expectations, which it reached in the summer of 2016 just before The DAO hack occurred (which we will discuss in detail in Chapter 5).
The first part is probably wrong if measured by actual usage and interest (as shown by the Google Search image a few sections above).41
The second part of the paragraph is probably right, though the timing was probably a little later: likely in the last quarter of 2016 when the first set of pilots turned out to require substantially larger budgets. That is to say, in order to be put platforms into production most small vendors with short runways realized they needed more capital and time to integrate solutions into legacy systems. In some cases, that was too much work and a few vendors pivoted out of enterprise and created a coin or two instead.42
On p. 31 they write:
Yes, the numbers have changed a lot since. Crypto moves fast.
This isn’t a hill I want to die on, but historically “crypto” means cryptography. Calling cryptocurrencies “crypto” is basically slang, but maybe that’s the way it evolves towards.
On p. 32 they write:
Historically, crypotassets have most commonly been referred to as cryptocurrencies, which we think confuses new users and constrains the conversation on the future of these assets. We would not classify the majority of cryptoassets as currencies, but rather most are either digital commodities (cryptocommodities), provisioning raw digital resources, or digital tokens (cryptotokens), provisioning finished digital goods and services.
They have a point but a literature review could have been helpful at showing this categorization is neither new nor novel.
In 2014, an academic paper was published that attempted to categorize Bitcoin from an ontological perspective. Based on the thought process presented in that paper, the Dutch authors concluded that Bitcoin is a money-like informational commodity. It isn’t money and isn’t a currency (e.g., isn’t actually used).434445
On p. 32 they write:
In an increasingly digital world, it only makes sense that we have digital commodities, such as computer power, storage capacity, and network bandwidth.
This book only superficially explains each of these and doesn’t drill down into why these “digital commodities” can’t be priced in good old fashioned money or why an internet coin is needed. If this is a good use case, is it just a matter of time before Blizzard and Steam get on board? Maybe worth looking at what entertainment companies do for the next edition.
On p. 33 they write about “why crypto” as shorthand for “cryptoassets” instead of “cryptography.”
For historical purposes, Matt Blaze, the most recent owner of crypto.com, provides a good explanation that could be included or cited next edition: Exhaustive Search Has Moved.
On p. 35 they write:
Except for Karma, the problem with all these attempts at digital money was that they weren’t purely decentralized — one way or another they relied on a centralized entity, and that presented the opportunity for corruption and weak points for attack.
This seems to be conflating two separate things: anonymity with electronic cash. You can have one without the other and do.46
Also, the BIP process is arguably a weak point for attack.47
On p. 35 they write:
One of the most miraculous aspects of bitcoin is how it bootstrapped support in a decentralized manner.
The fundamental problem with this statement is that it is inaccurate.48 Large amounts of centralization continues to exist: mining, exchanges, BIP vetting, etc.
On p. 35 they write:
Together, the combination of current use cases and investors buying bitcoin based on the expectation for even greater future use cases creates market demand for bitcoin.
Is that a Freudian slip?
Speculators buy bitcoin because they think can sell bitcoins at a higher price because a new buyer will come in at a later date and acquire the coins from them.49
For example, last month Hyun Song Shin, the BIS’s economic adviser and head of research, said:
“If people pay to hold the tokens for financial gain, then arguably they should be treated as a security and come under the same rigorous documentation requirements and regulation as other securities offered to investors for a return.”
In the United States, recall that one condition for what a security is under the Howeyframework is an expectation of profit.
Whether Bitcoin is a security or not is a topic for a different post.50
On p. 36 they write:
For the first four years of Bitcoin’s life, a coinbase transaction would issue 50 bitcoin to the lucky miner.
On November 28, 2012, the first halving of the block reward from 50 bitcoin to 25 bitcoin happened, and the second halving from 25 bitcoin to 12.5 bitcoin occurred on July 9, 2016. The thrid will happen four years from that date, in July 2020. Thus far, this has made bitcoin’s supply schedule look somewhat linear, as shown in Figure 4.1.
Technically incorrect because of the inhomogeneous Poisson process and the relatively large amounts of hashrate that came online, the first “4 year epoch” was actually less than 4 years.
Whereas the genesis block was released in January 2009, the first halving should have occurred in January 2013, but instead it took place in November 2012. Similarly, the second halving should have — if rigidly followed — taken place in November 2016, but actually occurred in July 2016 because even more hashrate had effectively accelerated block creation a bit faster than expected.
On p. 36 they write:
Based on our evolutionary past, a key driver for humans to recognize something as valuable is its scarcity. Satoshi knew that he couldn’t issue bitcoin at a rate of 2.6 million per year forever, because it would end up with no scarcity value.
Maybe Satoshi did or did not think this way, but irrespective of his or her view, having a finite amount of something means there is some amount of scarcity… even if it is a relatively large amount. Now this discussion obviously leads down the ideological road of maximalism which we don’t have time to go into today.52 Suffice to say that bitcoin is fundamentally not scarce due to its inability to prevent forks that could increase or decrease the money supply.
On p. 37 they write:
Long term, the thinking is that bitcoin will become so entrenched within the global economy that new bitcoin will not need to be issued to continue to gain support. At that point, miners will be compesnated for processing transaction and securing the network through fees on high transaction volumes.
This might happen but hasn’t yet.
For instance, Kerem Kaskaloglu (see p. 71) created a cartoon model to show what this should look like.
Notice how reality doesn’t stack up to the idealized version (yet)?
On p. 39 they write about BitDNS, Namecoin, and NameID:
Namecoin acts as its own DNS service, and provides users with more control and privacy.
In the next edition they should mention how Namecoin ended up having one mining pool that consistently had over 51% of the network hashrate and as a result, projects like Onename moved over to Bitcoin and then eventually its own separate network altogether (Blockstack).
On p. 41 they write:
This is an important lesson, because all cryptocurrencies differ in their supply schedules, and thus the direct price of each cryptoasset should not be compared if trying to ascertain the appreciation potential of the asset.
One way to strengthen this section is to provide a consistent model or methodology to systemically value a coin that doesn’t necessarily involve future demand from new investors. Maybe in the second edition they could provide a way to compare or at least say that no valuation model works yet, but here is a possible alternative?
On p. 42 they write:
A word to the wise for the innovative investor: with a new cryptocurrency, it’s always important to understand how it’s being distributed and to whom (we’ll discuss further in Chapter 12). If the core community feels the distribution is unfair, that may forever plague the growth of the cryptocurrency.
If a cryptocurrency or “cryptoasset” is supposed to be decentralized, how can it have a singular “core” community too?
In practice, most retail buyers of coins don’t seem to care about centralization or even coin distribution. Later in the book they mention Dash and its rapid coin creation done in the first month. Few investors seem to care. 53
On p. 42 they write:
Ripple has since pivoted away from being a transaction mechanism for the common person and instead now “enables banks to send real-time international payments across network.” This focus plays to Ripple’s strengths, as it aims to be a speedy payment system that rethinks correspondent banking but still requires some trust, for which banks are well suited.
If readers have time, I recommend looking through the marketing material of OpenCoin, Ripple Labs, and Ripple from 2013-2018 because it has changed several times.54 Currently there are a couple of different products including xRapid and xCurrent which are aimed at different types of users and as a result, the passage above should be updated.
On p. 43 they write:
Markus used Litecoin’s code to derive Dogecoin, thereby making it one more degree of separation removed from Bitcoin.
This is incorrect. Dogecoin was first based off of Luckycoin and Luckycoin was based on a fork of Litecoin. The key difference involved the erratic, random block reward sizes.
On p. 45 they write about Auroracoin.
Auroracoin is a cautionary tale for both investors and developers. What began as a seemingly powerful and compelling use case for a cryptoasset suffered from its inability to provide value to the audience it sought to impact. Incelanders were given a cryptocurrency with little education and means to use it. Unsurprisingly, the value of the asset collapsed and most considered it dead. Nevertheless, cryptocurrencies rarely die entirely, and Auroracoin may have interesting times ahead if its developer team can figure out a way forward.
Over 1,000 other coins have died, so “rarely” should be changed in the next edition
Why does a decentralized cryptocurrency have a singular development team, isn’t that centralization?
On p. 46 they write:
Meanwhile, Zcash uses some of the most bleeding-edge cryptography in the world, but it is one of the youngest cryptoassets in the book and suitable only for the most experienced cryptoasset investors.
In the next edition it would be helpful to specifically detail what makes someone an experienced “cryptoasset” investor.
On p. 46 they write:
Adam Back is considered the inspiration for Satoshi’s proof-of-work algorithm and is president of Blockstream, one of the most important companies in the Bitcoin space.
While Hashcash was cited in the original Satoshi whitepaper, recall above, that the original idea can be directly linked to a 1993 paper entitled Pricing via Processing or Combatting Junk Mail by Cynthia Dwork and Moni Naor. Also, it is debatable whether or not Blockstream is an important company, but that’s a different discussion altogether.
On p. 46 they write:
Bitcoin and the permissionless blockchain movement was founded on principles of egalitarian transparency, so premines are widely frowned upon.
What are the founding principles? Where can we find them? Maybe it exists, but at least provide a footnote.55
On p. 47 they write:
While many are suspicious of such privacy, it should be noted that it has tremendous benefits for fungibility. Fungibility refers to the fact that any unit of currency is as valuable as another unit of equal denomination.
Monero’s supply schedule is a hybrid of Litecoin and Dogecoin. For monero, a new block is appended to its blockchain every 2 minutes, similar to Litecoin’s 2.5 minutes.
In the next edition I’d tighten the language a little because a new monero block is added roughly or approximately every 2 minutes, not exactly 2 minutes.
On p. 48 they write:
By the end of 2016, Monero had the fifth largest network value of any cryptocurrency and was the top performing digital currency in 2016, with a price increase over the year of 2,760 percent. This clearly demonstrates the level of interest in privacy protecting cryptocurrency. Some of that interest, no doubt, comes from less than savory sources.
That is a non sequitur.
Where are the surveys of actual Monero purchasers during this time frame and their opinions for why they bought it? 56
For instance, in looking at the two-year chart above, how much on-chain activity in 2016 was due to speculators interest in “privacy” versus coin flipping? It is impossible to tell. Even with analytics all you will be able to is link specific users with purchases. Intent and motivation would require surveys and subpoenas; worth adding if available in the next edition.
On p. 48 they write:
Another cryptocurrency targeting privacy and fungiblity is Dash.
Is Dash really fungible though? That isn’t explored in this section. Plus Dash has a CEO… how is that decentralized?
On p. 49 they write:
In fact, Duffield easily could have relaunched Dash, especially considering the network was only days old when the instamine began to be widely talked about, but he chose not to. It would have been unusual to relaunch, given that other cyrptocurrencies have done so via the forking of original code. The creators of Monero, for example, specifically chose not to continue building off Bytecoin because the premine distribution had been perceived as unfair.
How is this not problematic: for a “decentralized” cryptocurrency to be controlled and run by one person who can unilaterally stop and restart a chain?
It actually is common, that’s the confusing part. Why have regulators such as FinCEN and the SEC not provided specific guidance (or enforcement) on the fact that one or a handful of individuals actually are unlicensed / non-exempted administrators of financial networks?
On p. 49 they write:
The Bitcoin and blockchain community has always been excited by new developments in anonymity and privacy, but Zcash took that excitement to a new level, which upon issuance drove the price through the roof.
Putting aside the irrational exuberance for Zcash itself, why do the authors think so many folks are vocal about privacy and anonymity?
Could it be that a significant portion of the coins are held by thieves of exchanges and hosted wallets who want to launder them? Here are a few recent examples:
Through his time at DigiCash and longstanding involvement in cryptography and cryptoassets, Zooko has become one of the most respected members in the community.
Let’s put aside Zooko and Zcash. The phrase, “the community” frequently appears in this book and similar books. It is an opaque, ill-defined (and cliquish) term that is frequently used by coin promoters to shun certain people that do not promote specific policies (and coins).57 It’s a term that should be clearly defined in the next edition.
On p. 50 they write:
While it is still early days for Zcash, we are of the belief that the ethics and technology chops of Zooko and his team are top-tier, implying that good things lie in wait for this budding cryptocurrency.
The statement above seems like an endorsement. Did either of the authors own Zcash just as the book came out? And what are the specific ethics they speak of? And why do the authors call it a cryptocurrency instead of a “cryptoasset”?
On p. 51 they write:
For example, the largest cryptocommodity, Ethereum, is a decentralized world computer upon which globally accessible and uncensored applications can be built.
How is it a commodity? Maybe it is and while they use a lot of words in this chapter, they never really precisely why it is in a way that makes much sense. Recommend modifying the first few pages of this chapter.
On p. 52 they write about “smart contracts” and mention Nick Szabo.
For a future edition I recommend diving deeper into the different uses and definitions of smart contracts. Also could be worth following Tony Arcieri suggestion:
I really like “authorization programs” but people really seem married to the “smart contract” terminology. Never mind Martin Abadi’s work on authorization languages (e.g. Binder) predates Nick Szabo’s “smart contracts” by half a decade…
For instance, there has been a lot of work done via the Accord Project with Clause.io and others such as IBM and R3. Also worth looking into Barclay’s and UCL’s effort with the Smart Contract Templates. A second edition that aims to be up-to-date should look at these developments and how they have evolved from what Abadi and Szabo first proposed.
On p. 53 they mentioned that Counterparty “was launched in January 2014.” Technically that is not true. The fundraising (“proof-of-burn”) took place in January and it was the following month that it “launched.”
On p. 54 they write:
The reason Bitcoin developers haven’t added extra functionality and flexibility directly into its software is that they have prioritized security over complexity. The more complex transactions become, the more vectors there are to exploit and attack these transactions, which can affect the network as a whole. With a focus on being a decentralized currency, Bitcoin developers have decided bitcoin transactions don’t need all the bells and whistles.
This is kind of true but also misses a little history.
For instance, Zerocoin was first proposed as an enhancement directly built into Bitcoin but key, influential Bitcoin developers who maintained the repository, pushed back on that for various technological and philosophical reasons. As a result, the main authors of that proposal went on to form and launch Zcash.58
On p. 56 they write:
Buterin understood that building a system from the ground up required a significant amount of work, and his announcement in January 2014 involved the collaboration of a community of more than 15 developers and dozens of community members that had already bought into the idea.
I assume the authors mean, following the Bitcoin Miami announcement in January 2014, but they don’t really say. I’m not sure how they arrive at the specific headcount numbers they did above, would be good to add a footnote in the future.
On p. 56 they write:
The ensuing development of the Bitcoin software before launch mostly involved just two people, Satoshi and Hal Finney.
This assumes that Satoshi is not Hal Finney, maybe he was. But it should also include the contributions of Ray Dillinger and others.
On p. 56 they write:
Buterin also knew that while Ethereum could run on ether, the people who designed it couldn’t, and Ethereum was still over a year away from being ready for release. So he found funding through the prestigious Thiel Fellowship.
This is inaccurate.
After reading this, I reached out to Vitalik Buterin and he said:59
That’s totally incorrect. Like the $100k made very little difference.
So that should be corrected in the next version.
On p. 57 they write:
Ethereum democratized that process beyond VCs. For perspective on the price of ether in this crowdsale, consider that at the start of April 2017, ether was worth $50 per unit, implying returns over 160x in under three years. Just over 9,000 people bought ether during the presale, placing the average initial investment at $2,000, which has since grown to over $320,000.
There are a few issues with this:
Ethereum did a small private and a larger public sale. We do have the Terms and Conditions of the public sale but we do not know how many participated in the private sale and under what terms (perhaps the T&Cs were identical).
Over the past 12 months there has been a trend for the “top shelf” ICOs to eschew a public sale (like Ethereum did) and instead, conduct private placement offerings with a few dozen participants at most… typically VCs and HNWIs.
There are lots of dead ICOs. One recent study found that, “56% of crypto startups that raise money through token sales die within four months of their initial coin offerings.” Ethereum is definitely an exception to that and should be highlighted as such.
On p. 57 they write:
The extra allocation of 12 million ether for the early contributors and Ethereum Foundation has proved problematic for Ethereum over time, as some feel it represented double dipping. In our view, with 15 talented developers involved prior to the public sale, 6 million ether translated to just north of $100,000 per developer at the presale rate, which is reasonable given the market rate of such software developers.
Who are these 15 developers, why is that the number the authors have identified?
Also, how much should FOSS developers be compensated and/or the business model around that is a topic that isn’t really addressed at all in this book, yet it is a glaring omission since virtually all of the projects they talk about are set up around funding and maintaining a FOSS team(s). Maybe some findings will be available for the next version.
On p. 57 they write:
That said, the allocation of capital into founders’ pockets is an important aspect of crowdsales. Called a “founder’s reward,” the key distinction between understandable and a red flag is that founders should be focused on building and growing the network, not fattening their pockets at the expense of investors.
Because coins do not typically provide coin holders any type of voting rights, it is legally dubious how you can hold issuers and “founders” accountable.60
That is why, as mentioned above, there has been an evolution of terms and conditions such that early investors in a private placement for coins may have certain rights and that the founders have certain duties that are all legally enforceable (in theory).
Because no one is publishing these T&Cs, it is hard to comment on what are globally accepted practices… aside from allowing early investors liquidity on secondary markets where they can quickly dump coins.61
Without the ability to legally hold “founders” accountable for enriching themselves at the expense of the project(s), the an interim solution has been to get on social media and yell alot… which is really unprofessional and hit or miss. Another solution is class action lawsuits, but that’s a different topic.
Also, I put the “founders” into quotes because these seem to be administrators of a network, maybe in the next edition they will be described as such?
On p. 58 they write:
Everyone trusts the system because it runs in the open and is automated by code.
There is lots of different types of open source code that runs on systems that are automated. For instance, the entire Linux, Apache, and Mozilla worlds predate Bitcoin. That isn’t new here.62
Readers and investors shouldn’t just trust code because someone created a GitHub repo and said their blockchain is open and automated.63
On p. 59 they write:
Most cryptotokens are not supported by their own blockchain.
This is actually true and problematic because it creates centralization risks and the ability for one party to unilaterally censor transactions and/or act as administrators.
For instance, a few days ago, Bancor had a bug that was exploited and about $13.5 million in ETH were stolen… and Bancor was able to freeze the BNT. That’s because BNT is effectively a centrally administered ERC20 token on top of Ethereum.
Ignoring for the moment whether or not BNT is or is not a security, this is not the first time such issuance and centralization has occurred. See the colored coin mania from 2014-2015.
On p. 60 they write about The DAO:
Over time, investors in these projects would be rewarded through dividends or appreciation of the service provided.
They mention regulators briefly later on – about SEC views – but most of the content surrounding crowdsales was non-critical and borderline promotional.64 Might be worth adding more meat around this in the next edition.
On p. 61 they write about The DAO:
The hack had nothing to do with an exchange, as had been the case with Mt. Gox and other widely publicized Bitcoin-related hacks. Insted, the flaw existed in the software of The DAO.
However, a hard fork would run counter to what many in the Bitcoin and Ethereum communities felt was the power of a decentralized ledger. Forcefully removing funds from an account violated the concept of immutability.
Just a few pages earlier the authors were saying that the lead developer behind Dash should have restarted the network because that was common and now they’re saying that doing a block reorg is no bueno. Which is it?
Why should the reader care what a nebulously defined “community” says, if it is is not defined?
The reason we have codes of conduct, terms of service, and EULAs is to specifically answer these types of problems when they arise.
Since public blockchains are supposed to be anarchic, the lack of formal governance is supposed to be a feature, right? That’s a whole other topic but suffice to say that these two sentences should be reworded in the next edition to incorporate the wisdom found in the Lexicon paper.
On p. 62 they write:
Many complained of moral hazard, and that this would set a precdent for the U.S. government or other powerful entities to come in someday and demand the same of Ethereum for their own interests. It was a tough decision for all involved, including Buterin, who while not directly on The DAO developer team, was an admistrator.
This is the first and only time they point out that key participants collectively making governance decisions are administrators… a point I have been highlighting throughout this review.
I don’t think it is fair to label Vitalik Buterin as a singular administrator, because if he was, he wouldn’t have had to ask exchanges to stop trading ether and/or The DAO token. Perhaps he was collectively involved in that process, but mining pool operators and exchange managers are arguably just as important if not more so. See also: Sufficiently Decentralized Howeycoins
On p. 62 they write:
While hard fork are often used to upgrade a blockchain architecture, they are typically employed in situations where the community agrees entirely on the beneficial updates to the architecture. Ethereum’s situation was different, as many in the community opposed a hard fork. Contentious hard forks are dangerous, because when new software updates are released for a blockchain in the form of a hard fork, there are then two different operating systems.
A few things:
Notice the continued use of an ill-defined “the community”
How is agreement or disagreement measured? During the Bitcoin block size debate, folks tried to use various means to express interest, most of which resulted in sybil attacks such as retweets and upvotes on social media by an army of bots.
Is any fork non-contentious. Surely if we looked hard enough, we could always find more than a handful of coin owners and/or developers that disagreed with the proposal. Does that mean you should ignore them? Whose opinion matters? These types of questions were never really formally answered either in the case of the Bitcoin Segwit / Bitcoin Cash fork… or in the Ethereum / Ethereum Classic / The DAO fork. Governance is pretty much an off-chain popularity contest, just like voting for politicians.65
On p. 63 they write:
The site for Ethereum Classic defines the cryptoasset as “a continuation of the original Ethereum blockchain–the classic version preserving untampered history; free from external interference and subjecitve tampering of transactions.”
This could be revised since Ethereum Classic itself has now had multiple forks.
As mentioned in a previous post last year:
Ethereum Classic: this small community has held public events to discuss how they plan to change the money supply; they video taped this coordination and their real legal names are used; only one large company (DCG) is active in its leadership; they sponsor events; they run various social media accounts
There has been lots of external interference, that’s been the lifeblood of public blockchains… because they don’t run themselves, people run and administer them.
Continuing on p. 63 they write:
While many merchants understably complain about credit card fees of 2 to 3 percent, the “platform fees” of Airbnb, Uber, and similar platform services are borderline egregious.
Maybe they are, maybe they are not.66 What is the right fee they should be? Miners take a cut, exchanges take a cut, developers take a cut via “founder’s funds.”
The next edition should give a step-by-step comparison to show why fee structures are egregious (maybe they are, it just is not clear in this book).
On p. 64 they wrote about Augur. Incidentally, Augur finally launched in early July while writing this review. I have an origin story but will keep that for later.
On p. 65 they wrote about Filecoin:
For example, a dApp may use a decentralized cloud storage system like Filecoin to store large amounts of data, and another cryptocommodity for anonymized bandwidth, in addition to using Ethereum to process certain operations.
A couple thoughts:
That’s the theory, though Filecoin hasn’t launched yet — why do they get the benefit of the doubt yet other projects don’t?
There is no price or use comparison in this chapter or elsewhere… the book could be strengthened if it provided more evidence of adoption because we have seen that running decentralized services such as Tor or Freenet have been less than spectacular.
On p. 65 they write:
Returning to the fundamentals of investment theory will allow innovative investors to properly position their overarching portfolio to take advantage of the growth of cryptoassets responsibly.
It is still unclear what an “innovative investor” is — at least the way these authors describe it.67
On p. 69 Tatar writes:
Not only did I decide to inveset in bitcoin, I decided to place the entirety of that year’s allocation for my Simplified Employee Pension (SEP) plan into bitcoin. When I announced what I had done in my article “Do Bitcoin Belong in your Retirement Portfolio?,” it created a stir online and in the financial planning community.
This was one of just a couple places where the authors actually disclose that they own specific coins, next edition they should put it up front.
On p. 70 Tatar writes:
Was I chasing a similar crash-and-burn scenario with bitcoin? Even my technologically and investment savvy son, Eric, initially criticized me about bitcoin. “They have these things called dollar bills, Dad. Stick to using those.”
Eric is probably right: that the authors of this book accepted traditional money for their book (Amazon doesn’t currently accept cryptocurrencies).
Based on their views presented in this book, the authors probably don’t spend (many) coins they may have in the portfolio, instead holding on to them with the belief that other investors will bid up the price (measured in actual money).
On p. 77 they write about the GFC prior to 2008:
Becoming a hedge fund manager became all the rage for business-minded students when it was revealed that the top 25 hedge fund managers earned a total of $22.3 billion in 2007 and $11.6 billion in 2008.
Coincidentally a similar “rage” for running cryptocurrency-related funds has occured in the past 18 months, especially for ICOs.
More than two hundred “funds” quickly popped up in order to gobble up coins during coin mania. At least 9 have closed down through April and many more were down double digits due to a bear market (and not hedging).
On p. 83 they write:
Bitcoin is the most exciting alternative asset in the twenty-first century, and it has paved the way for its digital siblings to enjoy similar success.
It is their opinion that this is the case, but the authors don’t really provide a lot of data to reinforce it yet, other than the fact that there have been some bull runs due to exuberance.68 Worth rewording in the next edition.
On p. 83 they write:
Because bitcoin can claim the title of being the oldest cryptoasset…
Similarly, I (Chris) didn’t even consider investing in bitcoin when I first heard about it in 2012. By the time I began considering bitcoin for my portfolio in late 2014, the price was in the mid $300s, having increased 460,000-fold from the initial exchange rate.
I believe this is the only time in the book that Burniske discloses any coin holdings.
On p. 85 they make some ridiculous comparison with the S&P 500, DJIA, NASDAQ 100… and Bitcoin.
The former three are indices of multiple regulated securities. The latter is just one coin that is easily influenced and manipulated by external unaccountable parties. How is that an apples to apples comparison?
On p. 87 they continue by comparing Bitcoin with Facebook, Google, Amazon, and Netflix.
Again, these are regulated securities that reflect cash flows and the financial health of multinational companies… Bitcoin has no cash flows and isn’t (yet) setup to be a company… and isn’t regulated (no KYC/AML at the mining farm or mining pool level).
Bitcoin was originally built to be an e-cash transmission network, a decentralized MSB.69 How is comparing it with non-MSBs a useful comparison?
On p. 88 they write:
Remember that, as of January 2017, bitcoin’s network value was 1/20, 1/22, 1/3, and 1/33 that of the FANG stocks respectively. Therefore, if bitcoin is to grow to a similar size much opportunity remains.
This whole section should be probably be modified because these aren’t apples-to-apples comparisons. FANG stocks represent companies that have to build and ship multiple products in order to generate continuous revenue.
With Bitcoin, it is bitcoin that is the product, nothing else is being shipped nor is revenue being generated70
Maybe the price of a bitcoin — as measured with actual money — does reach a 1:1 or even surpass the stocks above. But a new version of this book could be strengthened with an outline on how it could do so sustainably.
The authors do have a couple narrow, daily volatility charts in the book, but none that provide a similar wideview comparison with something that is remotely comparable (Bitcoin versus Twitter doesn’t make any sense).
On p. 101 they write:
Cryptoassets have near-zero correlation to other captial market assets.
In contrast, the past few years have been more nuanced: bitcoin’s volatily has calmed, yet it retains a low correlation with other assets.
That first part is untrue, as shown by the chart above from JP Koning. The second part is relative.72
On p. 107 they write:
The Securities and Exchange Commission has thus far steered clear of applying a specific label to all cryptoassets, though in late July 2017 it did release a report detailing how some cryptoassets can be classified as securities, with the most notable example being The DAO.
That’s pretty much the extent of the authors analysis of the issue. Granted they aren’t lawyers but this is a pretty big deal, maybe in the next edition beef this up?
On p. 107 they write:
While it’s a great validation of cryptoassets that regulators are working to provide clarity on how to classify at least some of them, most of the existing laws set forth suffer from the same flaw: agencies are interpereting cryptoassets through the lens of the past.
From this wording it seems that the authors want laws changed or modified to protect their interests and the financial interests of their LPs. This isn’t the first or last time that someone with a vested interest lobbies to get carve outs, exceptions, or entire moratoriums.
Maybe that it is deserved, but it’s not well-articulated in this chapter other than to basically call regulators “old-fashioned” and out of touch with technology.73 Could be worth rethinking the wording here.
On p. 107 they write:
Just as there is diversity in equities, with analsts segmenting companies depending on their market capitalization, sector, or geography, so too is there diversity in cryptoassets. Bitcoin, litecoin, monero, dash, and zcash fulfill the three definitions of a currency: serving as a means of exchange, store of value and unit of account.
This is empirically incorrect. None of these coins functions as a unit of account, they all depend on and are priced in… actual money.74
There are lots of reasons for why this is case but that is beyond the scope of this review. 7576
On p. 110 they write about ETFs:
It should be noted that when we talk about asset classes we are not doing so in the context of the investment vehicle that may “house” the underlying asset, whether that vehicle is a mutual fund, ETF, or separately managed account.
They don’t really discuss it in the book, but just so readers are aware, there have been about 10 Bitcoin-only ETFs proposed in the US, all of which have been rejected by the SEC (or applications were voluntarily removed).
Curious to know why? See the March 10, 2017 explanation from the SEC.
Note: this hasn’t stopped sponsors from re-applying. In the process of writing this review, the CBOE filed for a Bitcoin ETF.
On p. 111 they write:
Much of the thinking in this chapter grew out of a collaboration between ARK Invest and Coinbase through late 2015 and into 2016 when the two firms first made the claims that bitcoin was ringing the bell for a new asset class.
Just to be clear: the joint paper they published in that time frame was a bit superficial as it lacked actual user data from Coinbase exchanges (both GDAX and the consumer wallet). I pointed that out back then and this book is basically an expanded form of that paper: where is specific usage data on Coinbase? The only way we have learned any real user numbers about Coinbase is from an IRS lawsuit.
For instance, a future edition should try to differentiate on-chain activity that is say, gambling winnings or miners payouts from exchange arbitrage or even coin shuffling. Their analysis should be redone once they remove the noise from the signal (e.g., not all transactional activity is the same).
This is a real challenge and not a new issue. For instance, see: Slicing data.
On p. 112 they write:
Cryptoassets adhere to a twenty-first century model of governance unique from all other asset classes and largely inspired by the open source software movement. The procurers of the asset and associated use cases are three pronged. First, a group of talented software developers decide to create the blockchain protocol or distributed application that utilizes a native asset. These developers adhere to an open contributor model, which means that over time any new developer can earn his or her way onto the development team through merit.
There is no new governance model.
In practice, changes are done via social media popularity contests. We saw that with the Bitcoin blocksize debate and Ethereum hard fork. And in some ways, strong vocal personalities (and cults of personality) is how other FOSS projects (like Python) are managed and administered.
The fluffy meritocracy feel-goodism is often not the order of the day and we see this in many projects such as Bitcoin where the commit access and BIP approval process is limited to a small insular clique.
The 4 point plan above is a much more accurate break down of how most coin projects are setup.
On p. 112 they write:
However, the developers are not the only ones in charge of procuring a cryptoasset; they only provide the code. The people who own and maintain the computers that run the code–the-miners–also have a say in the development of the code because they have to download new software updates. The developers can’t force miners to update software. Instead, they must convince them that it makes sense for the health of the overall blockchain, and the economic health of the miner, to do so.
But in many projects: developers and miners are one in the same. This is why it is so confusing to not have seen additional clarity or guidance from FinCEN because of how centralized most projects are in practice.
These companies often employ some of the core developers, but even if they don’t, they can assert significant influence over the system if they are a large force behind user adoption.
Maybe that is the case for some cryptocurrencies.78 Should “core” developers be licensed like professional engineers are?
Also, isn’t their statement above evidence that most projects are fairly centralized because the division of labor results in specialization?
On p. 113 they write:
These users are constantly providing feedback to the developers, miners, and companies, in whose interest it is to listen, because if users stop using the cryptoasset, then demand will go down and so too will the price. Therefore, the procurers are constantly held accountable by the users.
Except this isn’t what happens in practice.
Relatively little activity takes place at all on most of these coin platforms and most of what does occur involves arbitrage trading and/or illicit activity.
This activity seems to have little direct connection to the price of the coin because the price of the coin is still largely determined by the whims of speculative demand.
For instance, above is a two-year transactional volume chart for bitcoin. The price of bitcoin in the summer of 2016 was in the $600-$700 range whereas it is 10x that today. Yet daily transaction volume is actually lower than it was back then. Which means: the two are separate phenomenon.
Also, arguably the only direct way coin owners can — in practice — hold maintainers accountable is via antics on social media. That is why control of a specific reddit, Telegram, or Twitter account is very important and why hackers target those channels in order to influence prices.
On p. 113 they write about supply schedules:
For example, with oil, there’s the famous Organization of the Petroleum Exporting Countries (OPEC), which has had considerable control over the supply levels of oil.
Inadvertently they actually described how basically all proof-of-work coins operate: via a small clique of known miners and mining pools. A cartel?
While these miners have not yet increased or decreased the supply of bitcoins, mining is a specialized task that requires certain capital and connections in order to be successful at. These participants could easily collude to change the money supply, censor transactions, etc. and there would be no immediate legal recourse.
On p. 115 they write:
Cryptoassets, like gold, are often constructed to be scarce in their supply. Many will be even more scarce than gold and other precious metals. The supply schedule of cryptoassets typically is metered mathematically and set in code at the genesis of the underlying protocol or distributed application.
How to measure scarcity here?
Despite what alchemists tried for centuries to do: aside from particle accelators, on Earth the only way of increasing the supply of gold and silver is via digging it out of the ground. For cryptocurrencies, it is relatively easy to fork and clone both code and chains. Digital scarcity for most — if not all — public chains, seems to be is a myth.
In the next edition, maybe remove the “backed by maths” trope? None of these chains run themselves, they all depend on humans to run the equipment and maintain the code.
On p. 115 they write:
As discussed earlier, Satoshi crafted the system this way because he needed initially to bootstrap support for Bitcoin which he did by issuing large amounts of the coin for the earliest contributors. As Bitcoin matured, the value of its native asset appreciated, which means less Bitcoin is over eight years old, it provides strong utility to the world beyond as an investment, which drive demand.
Satoshi likely mined around 1 million bitcoins for himself/herself. Because of how centralized and small the network originally was in 2009, he/she probably could have unilaterally stopped the network and relaunched it and effectively removed that insta-mine. 79
In addition, there was almost no risk to either be a developer or a miner… the entry/exit costs were very low… so why did he issue large amounts of coins for these contributors?80
Also, how does it provide strong demand beyond investment? How many people do the authors know regularly use Bitcoin itself for retail payments?81
Also, through Bitcoin’s evolution, arguably some of its utility was removed by going down a specific block size path. The counterargument is that payments will be done via some other networks (such as Lightning) attached to Bitcoin, but as of this writing, that hasn’t panned out.
One last comment about this passage, FOSS is historically charity work and difficult to build a sustainable operation. A couple notable exceptions are Red Hat and SUSE (which was just acquired by EQT).
On p. 115 they write:
The Ethereum team is currently rethinking that issuance strategy due to an intended change in its consensus mechanism.
In the second edition is it possible to be consistent on this one point: how is an “official” or “centralized” development team congruent with the idea of having a “decentralized ecosystem”?
Also, the administrators of Ethereum Classic modified the money supply last year and most folks were blasé. Where is the relevant FinCEN guidance?
On p. 115 they write:
Steemit’s team pursued a far more complicated monetary policy with its platform, composed of steem (STEEM), steem power (SP), and steem dollars (SMD).
They have also chosen to modify their monetary policy post-inception.
The authors of this book need to be consistent in their wording because in other places they criticize centralized financial institutions but do not criticize centralized monetary supply decision of coin makers. Also, again, why or how does a decentralized project have a singular team?
On p. 116 they write:
Crypotassets can be likened to silicon. They have come upon the scene due to the rise of technology, and their use cases will grow and change as technology evolves. Currently, bitcoin is the most straightforward, with its use case being that of a decentralized global currency. Ether is more flexible, as developers use it for computational gas within a decentralized world computer.
This isn’t a good analogy. Silicon exists as a naturally occurring element… whereas cryptocurrencies do not naturally arise — humans create them.
In addition, bitcoin is arguably not the most straightforward due to a long divorce and schism process the past three years. One distinct group of promoters calls it “digital gold” and another distinct group calls it a “payment system” — the two groups are almost violently opposed to one another’s existence.
On p. 116 they write:
Then there are the trading markets, which trade 24/7, 365 days a year. These global and eternally open markets also differentiate cryptoassets from other assets discussed herein.
The FX markets are open globally almost 24/6 for most of the year, so that’s not really a braggable claim.82 There are legal, regulatory, and practical reasons why most capital markets operate in the time windows they do… it is not because of some technological limitation. Worth rewording in the next edition.
On p. 116 they write:
In short, the use cases for cryptoassets are more dynamic than any preexisting asset class. Furthermore, since they’re brought into the world and then controlled by open-source software, the ability for cryptoassets to evolve is unbounded.
In the next edition, maybe remove the pomp and circumstance unless there is actual data to back up the platitudes. We can all easily conjure up lots of potential use cases for just about any type of technology, but unless they are built and used, the hype should be turned down a few notches.
Also, there are many other open sourcesoftware projects that have actually shipped frequently used productivity tools and no one is yelling from the mountain tops about how they have unbounded potential. How are internet coins any different?
On p. 117 they write:
Cryptoassets have two drivers of their basis of value: utility and speculative.
In theory, perhaps. But in practice, most coins just have potential utility because with few exceptions, most buyers typically hold with the expectation the coin will appreciate. Maybe that change in the future.
On p. 117 the write:
For example, Bitcoin’s blockchain is used to transact bitcoin and therefore much of the value is driven by demand to use bitcoin as a means of exchange.
Perhaps, though in the next edition recommend modifying the wording to include: “… as a means of exchange or investment…” Currently, we know a large portion of activity is likely movement (arbitrage) between exchanges.8384
But even ignoring this data (from analytics companies) this scenario has been diced-up elsewhere:
Speculative value is driven by people trying to predict how widely used a particular cryptoasset will be in the future.
If there are systematic surveys of actual buyers and sellers perhaps add those in the second edition.85
On p. 118 they write:
With cryptoassets, much of the speculative value can be derived from the development team. People will have more faith that a cryptoasset will be widely adopted if it is crafted by a talented and focused development team. Furthermore, if the development team has a grand vision for the widespread use of the cryptoasset, then that can increase the speculative value of the asset.
This is false.
For starters, the value of a new coin is almost entirely a function of the marketing effort from the coin issuers: that’s why nearly all ICOs carve out a portion of their funding pie to market, promote, and advertise… spreading the sexy gospel of the new coin.
This is a big bucks opaque industry, with all sorts of shenanigans that take place just to get listed on secondary markets… with coin issuers paying more than $1 million to get listed.
While $1 million or even $3 million may sound like a lot to get listed, the issuers know it is worth it because the retail speculators on the other end will at least temporarily pump the coin price up often long enough for the original insiders and investors to cash out.
Now the coin issuers may talk a big game and at eloquent length about how their grand vision: that their coin will end world hunger and save the environment, but they often have no ability to execute and build the product(s) they claimed in their whitepaper.
As mentioned above, one recent study found that, “56% of crypto startups that raise money through token sales die within four months of their initial coin offerings.”
Also, how does a decentralized cryptocurrency have an official singular development team?
On p. 118 they write:
As each cryptoasset matures, it will converge on its utility value. Right now, bitcoin is the furthest along the transition from speculative price support to uility price support because it has been around the longest and people are using it regularly for its intended utility use case.
And what is its intended use case? The maximalist vision (digital gold) or the originalist payments vision?
On p. 118 they write:
For example, in 2016, $100,000 of bitcoin was transacted every minute, which creates real demand for the utility of the asset beyond its trading demand. A great illustration of bitcoin’s price support increasingly being tied to utility came from Pantera Capital, a well-respected investment firm solely focused on cryptoassets and technology. in Figure 8.2 we can see that in November 2013 bitcoin’s speculative value skyrocketed beyond its utility value, which is represented here by transactions per day using Bitcoin’s blockchain (CAGR is the compound annual growth rate).
But this didn’t happen.
Pantera has a habit of cherry picking dates and using different types of graphs (such as log versus linear) in order to talk its book.
For instance, they conjured up and pushed the “bitcoin absorbs the value of gold” narrative back in late 2014. Then a year later, they became part of the “great pivot” by rebranding everything “blockchain” instead of bitcoin.
Putting those aside, the transactional part of the graph (Figure 8.2) from Pantera was published in early 2017 and has not held up to further scrutiny by mid-2018.
Perhaps for some unknown reason the up-and-to-the-right hockey stick graph that Pantera tried to create with its dotted lines will germinate. But for now, as of this writing, their transactional / utility thesis is incorrect.
Why? Because the assumptions were the same as the authors of this book: they assume retail or institutional users will flock to using bitcoin for non-speculative reasons, but that has not occurred yet.
On p. 119 they write:
Speculative value diminishes as a cryptoasset matures because there is less speculation regarding the future markets the cryptoasset will penetrate. This means people will understand more clearly that demand for the asset will look like going forward. The younger the cryptoasset is, the more its value will be driven by speculative vlaue, as shown in Figure 8.3. While we expect cryptoassets to ossify into their primary use cases over time, especially as they become large system that supports significant amounts of value, their open-source nature leaves open the possiblity that they will be tweaked to pursue new tangential use cases, which could once again add speculative value to the asset.
Their wording in this and other passages has definitive certainty without any hedging.
This is unfounded. Recall, what can be presented without evidence can be dismissed without evidence. This also makes a circular argument that the next edition needs to provide evidence for (or just remove it).
On p. 122 the write:
For example, currently the bond markets are undergoing significant changes, as a surprising amount of bond trading is still a “voice and paper market,” where trades are made by institutions calling one another and tangible paper is processed. This makes the bond market much more illiquid and opaque than the stock market, where most transactions are done almost entirely electronically: With the growing wave of digitalization, the bond markets are becoming increasingly liquid and transparent. The same can be said of markets for commodities, art, fine wine, and so on.
In re-reading this I can’t tell if the authors recognize that the bond market, as well as all of the other markets listed, started out in pre-electronic and even pre-industrial times.
That’s not to defend the status quo, only that if modern day trading platforms and automation existed a couple hundreds years ago, it is likely that bonds trading would have migrated much earlier than 2018… maybe even on a blockchain!
On p. 122 they write:
Cryptoassets have an inherent advantage in their liquidity and trading volume profile, because they are digital natives. As digital natives, cryptoassets have no physical form, and can be moved as quickly as the Internet can move the 1s and 0s that convey ownership.
This is conflating digitization/digitalization with blockchains. You can have one without the other and in fact, do.
For instance, with US equities, beginning in the ’60s through the ’70s, stocks were dematerialized then immobilized in CSDs and ownership is now transferred electronically.86
Perhaps there is something to be said about this market infrastructure further evolving in time with a blockchain of some kind.
For example in the US, the DTCC (a large CSD) has:
Virtually every major CSD, stock exchange, and clearing house has likewise publicly opined or participated in some blockchain-related initiatives. But that is a separate topic maybe worth looking into for the next edition.
On p. 123 they write:
Even though they are growing at an incredible clip, separation between cryptoasset markets and traditional investor capital pools still largely remains the case.
How much real money has actually entered the cryptocurrency market?
There have been several attempts to quantify it and it is still rather small, maybe up to $10 billion came in during 2017.
On p. 125 they write:
For example, in 2016, Monero experienced a sizeable increase in notoriety–largely because its privacy features began to be utilized by a well-known dark market–which sent its average trading volume skyrocketing. In December 2015, daily volume for the asset was $27,300, but by December 2016 it was $3.25M, well over a hunderfold increase. The price of the asset had appreciated more than 20-fold in the same period, so some of the increase in trading volume was due to price appreciation, but clearly a large amount was due to increased interest and trading activity in the asset.
But how do the authors know this “clearly” was the case? Did they do some random sample surveys? The next edition they need to prove their assumption, not just make them. After all, it is hard — perhaps impossible — to externally ascertain what is going on at an exchange simply by looking at self-published volumes.
Also, the exchanges that these coins trade on are still typically unregulated, with little optics into how often manipulation occurs. That is why a number of them have been subpoenaed by various governmental bodies; in the US this includes the SEC, CFTC, IRS, FBI, and even separate states acting in coordination.
On p. 129 they write:
From these trends, we can infer that this declining volatility is a result of increased market maturity. Certainly, the trend is not a straight line, and there are significant bumps in the road, depending on particular events. For example, monero had a spike in volatility in late 2016 because it experienced a significant price rise. This shows volatility is not only associated with a tanking price but also a skyrocketing price. The general trend, nonetheless, is of dampening volatility […].
This is not true either. Maybe there are cherry picked dates in which there is relatively lower volatility than normal, but this year alone prices as measured in real money, declined between 60-100% for basically all crypotocurrencies and this involved a roller coaster to achieve.
In fact, in the process of writing this review, there were multiple days in which prices increased 5-10% for most coins and then a few days later, saw the same size of loses. Erratic volatility has not disappeared.
On p. 133 they write:
Despite the many PBOC interventions, Chinese citizens used bitcoin to protect themselves against the erosion in value of their national currency.
Who in China did this?
I have spent an enormous amount of time visiting China the past several years on business trips and not once did someone say they had shifted their wealth from RMB into bitcoin because of RMB depreciation. There are many speculators and miners, but to my knowledge there has not been a formal survey of buyers and their motivations… and the result being because of RMB depreciation.
The next edition should either remove this statement or add a citation.
On p. 134 they write:
As bitcoin rose and fell, so too did these assets. This reinforces the need for the innovative investor to become knowledgeable about these assets’ specific characteristics and recognize where correlations may or may not occur.
Recommend removing “innovative investor” in this location.87
On p. 137 they write:
On its path to maturity, bitcoin’s price has experienced euphoric rise and harrowing drops, as have many cryptoassets. One of the most common complaints among bitcoin and cryptoasset naysayers is that these fluctuations are driven by the Wild West nature of the markets, implying that cryptoassets are a strange new breed that can’t be trusted. While each cryptoasset and its associated markets are at varying levels of maturity, associating Wild West behavior as unique to cryptoasset markets is misleading at best.
No it isn’t. The authors do not even define or provide some kind of way to measure “maturity.” This paragraph creates a strawman.
The burden-of-proof rests on the party making the positive claim. In this case, the party claiming that a coin is becoming mature must provide objective evidence this is taking place. Should reword in the next edition.
On p. 138 they write:
Broadly, we categorize five main patterns that lead to markets destabilizing: the speculation of crowds, “This time is different,” Ponzi schemes, Misleading information from asset issuers, Cornering.
Those are valid patterns, in full agreement here. But this edition does not help in dispelling these problems and arguably even contributes to some of the speculative frenzy.
On p. 138 they write:
Sometimes they do this to capitalize on short-term information they believe will move the market, other times they do it because they expect to ride the momentum of the market, regardless of its fundamentals. In short, they try to profit within the roller-coaster ride.
What are the fundamentals of any coin described in this book? Next edition, clearly write out 5-10 if possible.
On p. 139 they write:
As America was struggling through the Great Depression, which many pinned on the stock market crash of 1929, there was strong resentment against speculators. Every crisis loves a scapegoat.
And in Bitcoinland there is no difference. Bitcoiners love to blame: bankers, the Illuminati, naysayers, concern trolls, academics, the government, Jamie Dimon, big blockers, small blockers, weak hands, statists, other coins, China, George Soros, Warren Buffett, Mike Hearn… virtually every month there is a new boogeyman to blame something on. I’ve even been blamed many times and I’m not involved at all in the market.
On p. 143 they write:
Cheap credit often fuels asset bubbles, as seen with the housing bubble that led to the financial crisis of 2008. Similarly, cryptoasset bubbles can be created using extreme margin on some exchanges, where investors are effectively gambling with money they don’t have.
Fully agree, good point.
On p. 144 they write:
The best way to avoid getting burned in this manner is to do proper due diligence and have an investment plan that is adhered to.
Fully agree, good point.
On p. 145 they write:
The key to understanding bitcoin’s value is recognizing it has utility as “Money-over-Internet-Protocol”( MoIP)–allowing it to move large amounts of value to anyone anywhere in the world in a matter of minutes–which drives demand for it beyond mere speculation.
This might be partially true but is has the same feel-good narrative that folks like Andreas Antonopoulos have been getting paid handsomly to regurgitate. Bitcoin (the network) does not move anything beyond bitcoins (the coin). Users still have to convert bitcoins into actual money at end points.
Converting a large amount — greater than $10,000 — will likely require KYC and AML and maybe even sanctions checks. This adds time and money which is one of the reason why the remittance use-case didn’t really get much traction after the hype in 2014 – 2015 and why companies such as Abra had to pivot a few times.
With that said, their metapoint is valid on the edges: despite the frictions that may exist, some participants are willing to go through this experience in order to gain more anonymity for uses they might not otherwise be able to do using traditional methods.88
Over the past three years there has also been an expansion of country- and region-based payment schemes worldwide to achieve near-real-time transfers, with Europe being one of the most significant accomplishments.89
In parallel, there are on-going experimentation and scaling of private blockchain-based ‘rails’ like Swift gpi or Alipay with GCash which have a potential to surpass volumes of the Bitcoin network.90
On p. 145 they write:
When Mt. Gox was established, bitcoin finally became accessible to the mainstream.
Up until recently it was difficult for even diehard users to get onboarded onto most exchanges. And specifically in 2010 with the launch of Mt. Gox, Jed McCaleb used Paypal to help facilitate the transfer of money… until Paypal dropped Mt. Gox because of too many chargebacks. To get money into and out of Mt. Gox often was a frictionfull task, unless you lived in Japan.
On p. 149 they write:
As shown in Figure 10.4, steem’s price in bitcoin terms would fall from its mid-July peak by 94 percent three months later, and by 97 percent at the end of the year. This doesn’t mean the platform is bad. Rather, it shows the speculation and excitement about its prospects fueled a sharp rise and fall in price.
In hindsight, everything is 20-20. The same truism in their last sentence can be said just about with every coin that sees the meteoric rise that Steemit did in 2016.91
On p. 150 they write:
While zcash has since stabilized and continues to hold great promise as a cryptoasset, its rocky start was caused by mass speculation.
Do the authors own any Zcash (or other cryptocurrencies mentioned in this book besides bitcoin)?
In late 2016 there were oodles of “thought leaders” talking about how Zcash was — for a moment — valued at a trillion dollars because of the very thin supply that was trading on exchanges. It was a headscratching meme that illustrates a shortcoming to the common “market cap” valuation mehtod.92
On p. 152 they write:
The idea of valuation, which we will tackle in the next chapters, is a particularly challenging one for cryptoassets. Since they are a new asset class, they cannot be valued as companies are, and while valuing them based on supply and demand characteristics like that of commodiites has some validity, it doesn’t quite suffice.
Then why spend an entire chapter (Chapter 7) comparing coins such as bitcoin, to companies and their stock?
You can’t have it both ways. Either heavily modify Chapter 7 in the next edition, or remove this comment.
On p. 155 they write:
Given the emerging nature of the cryptoasset markets, it’s important to recognize that there is less regulation (some would say none) in this arena, and therefore bad behavior can persist for longer than it may in more mature markets.
And there are now full-time lobbyists and trade associations — sponsored by donors whom have benefited from this unregulated / underregulated market — that actively push back against sensible regulations being applied. But that’s a different conversation beyond this post.
On p. 155 they write:
As activity grows in bitcoin and crypotasset markets, investors must look beyond the madness of the crowd and recognize that there are bad actors who seek easy prey in these young markets.
Even for a book published in late 2017, this is pretty much lip service. Volumes of books can be written about the shenanigans within nearly every public ICO and high-profile coin project. The authors should either modify the statement above or ideally expand it to detail specific egregious examples besides just OneCoin.
While a truly innovative crypotasset and its associated architecture requires a heroic coding effort from talented developers, because the software is open source, it can be downloaded and duplicated. From there, a new cryptoasset can be issued wrapped in slick marketing. If the innovative investors doesn’t do proper due diligence on the underlying code of read other trusted sources who have, then it’s possible to fall victim to a Ponzi scheme.
Enough with the “heroic” adjectives, let’s not put anyone on a pedestal, especially if the platform is not being used by anyone besides speculators and illicit actors.
Secondly, a minor grammar question: other uses of “open-source” in this book have a dash and the one above does not.
Millions of dollars poured into OneCoin, whose technology ran counter to the values of the cryptoasset community: its software was not open source (perhaps out of fear that developers would see the holes in its design), and it was not based on a public ledger, so no transactions could be tracked.
First, what are the “values” that the “community” has? Are these explicity written somewhere? Who decided those?
Second, those actually don’t sound too uncommon.
For instance, one recent study found: “Security researchers have found, on average, five security flaws in each cryptocurrency ICO (Initial Coin Offering) held last year. Only one ICO held in 2017 did not contain any critical flaws.”
And remember, these projects are “open source” yet most buyers and investors didn’t bother looking at the code. OneCoin is par for the course.
On p. 159 they write:
The swift action revealed the strength of a self-policing, open-source community in pursuit of the truth.
In my most popular post last year, I went through in detail explaining how self-policing is an oxymoron in the cryptocurrency world.
For example, “the community” actively listed OneCoin on secondary markets and profited from its trading. Did exchange operators return those gains to victims? In addition, “the community” has thus far, not set up any self-regulating organization (SRO) that has any ability or teeth to enforce a code-of-conduct.
In fact, it was agencies from Sweden, the UK, and other governments that acted and cracked down on OneCoin… not a collective effort from exchanges or VCs or twitter personalities.
On p. 159 they explain googling for code on GitHub:
If nothing pops up with signs of the code on GitHub, then the cryptoasset is likely not open source, which is an immediate red flag that a cryptoasset and investment should be avoided.
Sure, but it doesn’t include the fact(s) that even in 2017 we knew that many coin projects had bugs in it… because there is no incentive to independently audit this code or to publish it in an objective manner.
For example, often when someone tries to help highlight problems, they are demonized as a “concern troll” as the coin tribes brigade their Twitter and reddit threads. There are a couple of sites like ConcourseQ that now do help highlight problems, but most “crypto thought leaders” on social media spend their time rallying retail investors to buy coins instead of busting or calling out the legitimate coin scams.
On p. 161 they write about John Law:
Fortunately, today it’s quite easy to find information on just about anyone through Google searches.
Yes and no. And that still doesn’t act as a shield against fraud. The founders of Centra had shady, criminal pasts but were still able to raise more than $30 million in an ICO. Their misdeeds only became widely known after a New York Timesarticle explored it… this was not a story that was investigated by any of the “coin media” who collectively have a vested interested not to “self-police” the market they cover.
As with most panics, the contagion spread from the Gold Exchange. Because of Gould’s cornering of the market, stock prices dropped 20 percent, a variety of agricultural exports fell 50 percent in value, and the national economy was disrupted for several months. Gould exited with a cool $11 million profit from the debacle, and scot-free from legal charges. It is all too common that character like Gould escape unscathed by the havoc they create, which then allows them to carry on with their machinations in other markets.
These kinds of panics and manipulation are part and parcel to retail traders on cryptocurrency exchanges. Scapegoats and the blame game consist of a myriad of boogeymen — but typically the culprits are never found.93
On p. 167 they write:
In addition to miners, in Dash there are entities called masternodes, which are also controlled by people or groups of people. Masternodes play an integral role in performing near instant and anonymous transaction with Dash.
Putting aside whether Dash is or is not anonymous… the fact that the authors state that humans play a direct role in running the infrastructure raises a bunch of questions that I have repeated in this review.
How are these participants held accountable? How is governance managed? Have these participants registered with FinCEN? Why or why not?
On p. 168 they write about the Bitcoin Rich List:
Another 116 addresses hold a total of 2.87 million bitcoin, or 19 percent of the total outstanding, which is sizeable. Unlike dash, however, these holders aren’t necessarily receiving half the newly minted bitcoin, and so their ability to push the price upward is less.
Should there be a thorough investigation of how any one party or set of parties can artificially move prices around based on control of the money supply? In our current real-world framework, there are frequent public hearings and audits done. When will minters of cryptocurrencies be publicly audited?
On p. 171 they write:
Each cryptoasset is different, as are the goals, objectives, and risk profiles of each investor. Therefore, while this chapter will provide a starting point, it is by no means comprehensive. It’s also not investment advice.
Throughout the book the authors have repeatedly endorsed or not-endorsed specific coins. The second edition needs to be a lot more consistent.
On p. 172 they write:
Currently, there is no such thing as sell-side research for cryptoassets, and this will require innovative investors to scour through the details on their own or rely on recognized thought leaders in the space.
This is a sad truth: it is nearly impossible to get neutral, objective research on any coin that has been created.
Why? Because all coin holders basically have an incentive to promote and advertise the coins they own and talk down other coins they perceive as competition. Paying “researchers” has happened and will continue to do so.
Also, here’s another appearance of “innovative investor” — can that be removed altogether?
And lastly, how to know who the “recognized thought leaders” are? Based on the amount of twitter followers they have? That has been gamed. Based on how popular their Youtube account is? That has been gamed.
For example, these two article explain some of this payola world:
It’s unclear if this is due to lobbying efforts or maybe the researchers owned a bunch of EOS coins. At this time, the EOS block producing and arbitrator framework are both broken. Block producers paused the network a few weeks ago and the arbitrators / constitutions will probably be scrapped.
How can this rating system be trusted?
On p. 173 they write about white papers:
Any cryptoasset worth its mustard has an origination white paper. A white paper is a document that’s often used in business to outline a proposal, typically written by a thought leader or someone knowledgeable on a topic. As it relates to cryptoassets, a white paper is the stake in the ground, outlining the problem the asset addresses, where the asset stands in the competitive landscape, and what the technical details are.
During the Consensus event this past May, someone accidentally dropped a napkin on the floor and someone loudly said: watch out, that’s the latest multimillion dollar white paper.
And that’s the situation where we are in now. Readers: the passage above was not at all critical of the real mess we are in today. For instance, Tron literally plagiarized in its whitepaper, raised a ton of money in its ICO and recently bought BitTorrent.
There is no direct connection between a “good” or “bad” whitepaper and the performance of the coin. Retail investors do not typically care and haven’t done much research. Yet another reason agencies such as the SEC will be overwhelmed in the coming years due to rampant fraud and deceit. Worth looking into the next edition.
On p. 173 they write:
Some of these white papers can be highly technical, though at the very least perusing the introduction and conclusion is valuable.
This seems like an incongruent statement compared to other advice in the book about doing deep research. Recommend revising.
On p. 174 they write:
A number of cryptoasset-based projects focus on social networks, such as Steemit and Yours, the latter of which uses litecoin. While we admire these projects, we also ask: Will these networks and their associated assets gain traction with competitors like Reddit and Facebook? Similarly, a cryptoasset service called Swarm City (formerly Arcade City) aims to decentralize Uber, which is already a highly efficient service. What edge will the decentralized Swarm City have over the centralized Uber?
And that in a nutshell is why the second edition of the book arguably needs to be slimmed down by 25%+. Virtually all of the use cases in this book are simply potential use cases and have shown little or even no traction in reality. For example, if the authors were as critical to Bitcoin and Zcash as they were to Swarm City then the second edition might be perceived as more balanced.
Specifically, in their promotion of Bitcoin as a payments platform, they have not done a deep dive into other existing payment networks, such as Visa or an RTGS from a central bank.94 They should do that in the next edition otherwise these come across as one-sided arguments.
Also, Yours switched from Litecoin over to Bitcoin Cash last year (around the time the book was published) and Swarm City is still not very active at the time this review was written.
On p. 175 they write about The Lindy Effect
The same applies to cryptoassets. The longest-lived cryptoasset, bitcoin, now has an entire ecosystem of hardware, software developers, companies, and users built around it. Essentially, it has created its own economy, and while a superior cryptocurrency could slowly gain share, it would have an uphill battle given the foothold bitcoin has gained.
This is untrue in theory and practice.
While maximalists would vocally claim that there can only be one-chain-to-rule-them-all, there is no real moat that Bitcoin has to prevent users from exiting or switching to other platforms (see discussion on substitute goods).
In practice, effectively all proof-of-work cryptocurrencies depend on external capital to stay afloat, often in the form of venture capital. ((See Robert Sams on rehypothecation, deflation, inelastic money supply and altcoins)) Part of the reason is that miners need to pay their bills in traditional currency and therefore must liquidate some or all of their coins to do so. Another issue is that because many participants think or believe that coin prices as measured in real money will increase in the future, they hold. Yet the expenses of service providers (exchanges, wallets, etc.) typically need to be paid with traditional money.
As a result, this creates sell-side pressure. And unlike the traditional FX market which has “natural” buyers in the form of international merchants and multinational corporations: there still is no “natural” buyers of cryptocurrencies outside of illicit activity (e.g., darknet market participants).
To compound this situation is that there is still no real circular flow of income, no real economy for any of these cryptocurrencies.95 And with the exception of a few cases each year, miners typically do not directly invest their coin holdings into companies, so crypotcurrency-related startups are dependent on foreign currency.
On p. 175 they write:
The demise of The DAO significantly impacted Ethereum (which The DAO was built on), but through leadership and community involvement, the major issues were addressed, and as of April 2017 Ethereum stands solidly as the second largest cryptoasset in terms of network value.
In the second edition, could the authors explicitly lay out how they define “leadership” in this context as well as what the “community” is? If it is singular and centralized, how is that fitting for an entity that is supposed to be decentralized?
Also, for readers interested in The DAO, here’s a short fiery thread on that topic.
On p. 176 they discuss “utility value and speculative value”
For bitcoin, its utility is that it can safely, quickly, and efficiently transfer value to anyone, anywhere in the world.
That may have been the original vision expressed in the whitepaper but it is not what the maximalists now claim Bitcoin is. Who’s promotion around utility is something we should take into consideration?
Also, considering how easy and common it is to hack cryptocurrency intermediaries such as exchanges, I think it is debatable that Bitcoin is “safe” for unsophisticated retail users, but that’s a separate topic.
On p. 176 they write:
The merchants wants to use bitcoin because it will allow her to transfer that money within an hour as opposed to waiting a week or more. Therefore, the Brazilian merchant buys US$100,000 worth of bitcoin and sends it ot the Chinese manufacture.
They explain a little more but the difficulties with this example starts here. The authors only focus on the bitcoins themselves, they don’t explore the actual full lifecycle that international merchants and manufacturers have to go through in order to exchange bitcoins into real money that they can use to pay bills.
That is to say: the Brazilian merchant and Chinese manufacture do not hold onto coins, so it is not just a matter of how fast they can send or receive the coins. What ultimately matters to them is how quickly they can receive the real money from a bank.
So the next edition needs to include the full roundtrip costs and frictions including the on-ramps and off-ramps into the traditional financial system. This is why many Bitcoin remittance companies struggled and ultimately had to pivot out of that cross-border use case (such as Abra). For the next edition, a side-by-side cost comparison would be helpful.96
On p. 177 they write:
That means on average each of these addresses is holding US$5.5 million worth of bitcoin, and it’s fair to assume that these balances are not those of merchants waiting for their transactions to complete. Instead, these are likely balances of bitcoin that entities are holding for the long term based on what they think bitcoin’s future utility value will be. Future utility value can be thought of as speculative value, and for this speculative value investors are keeping 5.5 million bitcoin out of the supply.
This seems like euphemisms. We understand that time preferences and discounted utility come into dramatic effect here. Maybe worth rewording?
For example, a large portion of those coins could be permanently destroyed (e.g., someone deleted the private key or threw away the hard drive). Though a significant portion could also be maximalists holding onto their coins with the hope that other investors create sufficient demand to move the price — as measured in real money — upward and upward. So they can then cash out.
If daily and weekly anecdotes on twitter and reddit are any indication, that’s arguably the real utility value of most coins, not just bitcoin. And there is some analytics to back up that argument too.
On p. 177 they write:
At the start of April 2017, there were just over 16 million bitcoin outstanding. Between international merchants needing 10 million bitcoin, and 5.5 million bitcoin held by the top 1,000 investors, there are only roughly 500,000 bitcoin free for people to use.
Citation needed. If the authors have any specific information that can share with the audience about any of these numbers, that’d be very helpful. Especially regarding the merchants needing 10 million bitcoin. If anything, there may be fewer merchants actively accepting bitcoin today than there were a couple years ago.
On p. 177 they write:
If demand continues to go up for bitcoin, then with a disinflationary supply schedule, so too will its price (or velocity).
It would be good to see what the authors think the velocity of bitcoin is. I’ve tried to track down and write about it in the past. See all of Chapter 9.
On p. 177 they write:
In other words, those investors no longer feel bitcoin has any speculative value left, and instead its price is only supported by current utility value.
As mentioned above, it would be helpful in the next edition if the authors included specific definitions and characteristics in a chart for what utility versus speculative value are.
Also, I don’t endorse the post in its entirety, but about five years ago Rick Falkvinge wrote an interesting note about the transactional value from illicit activity as it relates to Bitcoin. That has some actual data in it (though very old now).
On p. 178 they write:
For bitcoin, instead of looking at the “domestically produced goods and services” it will purchase in a period, the innovative investor must look at the internationally produced goods and services it will prucahse. The global remittances market–currently dominated by companies that provide the ability for people to send money to one another internationally–is an easy graspable example of service within which bitcoin could be used.
This whole section should probably be culled because this isn’t really a viable, scalable use case that bitcoin itself can solve.
For example, between 2014-2016, tens of millions of dollars were invested in more than a dozen “rebittance” companies (Bitcoin-focused remittance) and most either failed or pivoted.
Those that still exist had to build additional services and bitcoin were a means to an end. In all cases, these companies had to build their own cryptocurrency exchange and/or partner with several cryptocurrency exchanges in order to liquidate the coins — they need to hedge and limit their exposure to volatility. Bitcoin also doesn’t solve for the last-mile problem at all… but that is a separate topic.97
On p. 179 they write:
If each bitcoin needs to be worth $952 to service 20 percent of the remittance market and $11,430 to service the demand for it as digital gold, then in total it needs to be worth $12,382. There is no limit to the number of use cases that can be added in this process, but what is extremely tricky is figuring out the percent share of the market that bitcoin will ultimately fulfill and what the velocity of bitcoin will be in each use case.
This is highly debatable. And it is exactly what Pantera stated four years ago. Sources should be cited in the next edition; and also provide a velocity estimate for the potential use cases.
On p. 180 they write:
Taking the concepts of supply and demand, velocity, and discounting, we can figure out what bitcoin’s value should be today, assuming it is to serve certain utility purposes 10 years from now. However, this is much easier said than done, as it involves figuring out the sizes of those markets in the future, the percent share that bitcoin will take, what bitcoin’s velocity will be, and what an appropriate discount rate is.
An actual asset would certainly need these blanks filled, but Bitcoin doesn’t behave like a normal asset. For instance, it goes through enormous speculative bubbles and busts. It reached just under $20,000 per coin in mid-December last year not for any utility reason but pure speculation… yet many of the “thought leaders” at the time said it was because new buyers were going to use it for its utility.
On p. 180 they write:
Already there have been reports, such as those from Spence Bogart at Needham & Company, as well as Gil Luria at Webush, that look at the fundamental value of bitcoin.
I’ve read most of their reports, they’re nearly all based on edge-case assumptions or one-off anecdotes that never saw much traction (such as remittances). In addition, arguably both of their analysis may have been colored by their coin investments at the time they published their work. That’s not to say their material is discredited but I would discount some of their cryptocurrency-related reports.98
On p. 180 they write:
The valuations these analysts produce can be useful guides for the innovative investor, but they should not be considered absolute dictations of the truth. Remember, “Garbage in, garbage out.” We suspect that as opposed to these reports remaining proprietary, as is currently the case with much of the research of equities and bonds, many of these reports will become open-source and widely accessible to all levels of investors in line with the ethos of cryptoassets.
This has not happened. If anything, the market has been flooded with junk marketing material that masquerades as “research.” Universities are now getting funded by coin issuers and asked to co-publish papers. Even if there are no explicit shenanigans going on, there is now a shadow of doubt that hangs over these organizations.
Also, the next edition needs to define what “the ethos of cryptoassets” is somewhere up front. And dispense with “innovative investor”?99
On p. 182 they write about getting to know “the community and the developers”:
In getting to know the community better, consider a few key points. How committed is the developer team, and what is their background? Have they worked on a previous cryptoasset and in that processrefined their ideas so that they now want to alunch another?
If information cannot be found on the developers, or the developers are overtly anonymous, then this is a red flag because there is no accountability if things go wrong.
Satoshi clearly wouldn’t have been able to pass this test. Nor BitDNS originally (which later became Namecoin).
It is a double-standard to want accountability here yet promote an ill-defined “decentralization” throughout this book. You really can’t have it both ways.
Remember, the reason why administrators and operators of financial market infrastructure are heavily regulated is to hold participants legally responsible and accountable for when mistakes and accidents occur.
Cryptocurrencies were designed to be anarchic and purposefully were designed to not make a single participant accountabile. Trying to merge those two worlds creates the worst of both: permissioned-on-permissionless.
On p. 183 they write:
If Ethereum gets big enough, there may eventually be those who call themselves Ethereum Maximalists!
Yes, they exist and largely self-selected themselves into the Ethereum Classic world… you can see that by their antics on social media.
On p. 183 they write about issuance models:
Next, consider if the distribution is fair. Remember that a premine (where the assets are mined before the network is made widely available, as was the case with bytecoin) or an instamine (where many of the assets are mined at the start, as was the case with dash) are both bad signs because assets and power will accrue to a few, as opposed to being widely distributed in line with the egalitarian ethos.
Let’s tone down the talk on egalitarianism in a market fueled by greed and a perpetually high Gini coefficient.
In practice as of July 2018, many ICOs are pre-mined or pre-allocated, most as ERC20 tokens that are controlled by a singular entity (usually an off-shore foundation).100
Is this a “bad sign”? It would be helpful to see what the explicit criteria around token distribution should be in the next edition.101
On p. 183 they write:
For example, Ethereum started with one planned issuance model, but is deciding to go with another a couple years into launch. Such changes in the issuance model may occur for other assets, or impact those assets that are significatnly tied to the Ethereum network.
Those decision are made by individuals. Perhaps by the next edition we will know what FinCEN and other regulatory positions on individuals creating monetary policy and running financial market infrastructure.
On p. 184 they write:
With Dogecoin we saw that it needed lots of units outstanding for it to function as a tipping service, which justifies it currently having over 100 billion units outstanding, a significantly larger amount than Bitcoin. With many people turning to bitcoin as gold 2.0, an issuance model like Dogecoin’s would be a terrible idea.
What? Why? This passage conflates many different things.
As Jackson Palmer has repeatedly said: Dogecoin was set up as a joke, based on a meme. The authors seem to be taking its existence a little too seriously.
Dogecoin was originally based on Luckycoin which had a random money supply, so its original hashrate charts were all over the map, bipolar.
Its money supply was changed in part because it ran into an exitential crisis that it later (mostly) solved by merge mining with Litecoin in 2014
How does any of this have to do with maximalist narrative of “gold 2.0”?
On p. 186 they write:
The only way attackers can process invald transactions is if they own over half of the computer power of the network, so it’s critical that no single entity ever exceeds 50 percent ownership.
Technically this is not quite right.
The actual figure to sucessfully censor and/or reorg the chain may be as low as 33% and perhaps even 25% (dubbed “selfish mining“).102 More than 50% would mean the participants could do so repeatedly until their hashrate declines and/or a permanent fork occurs.
Aside from pressure on social media, there is nothing to prevent such “ownership” from taking place. And there is no legal recourse or accountability in the event it happens. And such “attacks” have occured on many different cryptocurrencies.103
On p. 186 they write:
In other words, miners are purley economically rational individuals–mercenaries of computer power–and their profit is largely driven by the value of the crypotasset as well as by transaction fees.
This should be reworded from the next edition because it is not true. Miners and mining pools are operated by people and they have various incentives, including to attack networks or abandon them altogether.
On p. 186 they write:
A clearly positively reinforcing cycle sets in that ensures that the larger the asset grows, the more secure it becomes–as it should be.
This is not true for proof-of-work coins.
If anything, mining and development have both trended towards centralization. For instance, it is estimated that Bitmain-manufactured hashing equipment currently generates 60-80% of the network hashrate and Bitmain-affiliated mining pools comprise about 50%+ of the current Bitcoin network. Maybe that is just momentary but singular entities on the mining side dominate many other cryptocurrencies as well. Perhaps that changes later in the year so it is worth revisiting in the next edition.
At the risk of being repetitive, more hash rate signifies more computers are being added to support the network, which signifies greater security.
This is a non sequitur. A new hashing machine capable of generating 10 times the amount of hashes as the previous machine could — ceteris paribus — result in other machines being turned off. In practice, you often have the Red Queen Effect take place (see Chapter 3).
Either way, depending on the costs of more efficient ASIC design, there could actually be fewer (or more) hashing machines added to a network depending on the expected price of the coin minus operating costs.
And in some cases, the network may become more centralized and therefore arguably less secure. Worth revising in next edition.
On p. 188 they write:
While hash rate often follows price, sometimes price can follow hash rate. This happens in situations where miners expect good things of the asset in the future, and therefore proactively connect machines to help secure the network. This instills confidence, and perhaps the expected good news has also traveled to the market, so the price start going up.
This passage has entered Rube Goldberg territory, where a series of specific events turn into a virtuous cycle in which prices go up and up but not down? How can we ever know what caused certain price increases or decreases with this type of asymmetric information occurring in the background? Suggest scrapping it in the next edition.
On p. 188 they write:
Ethereum’s mining network, on the other hand, is less built out because it’s a younger ecosystem that stores less value. As of March 2017, a 230 megahash per second (MH/s) mining machine could be purchased for $4,195, and it would take 70,000 of these machines to recreate Ethereum’s hash rate, totaling $294 million in value. Also, because Ethereum is supported by GPUs and not ASICs, the machines can more easily be constructed piecemeal by a hobbyist on a budget.
There are a few issues with this:
How do the authors measure or quantify “less built out”? Is there a line that is crossed in which Ethereum or other coins are “more built out” or the right size?
About a year ago a coin reporter asked me to detail the hypothetical lower bound costs for recreating the hashrate of the Bitcoin network. I provided those numbers based on Bitmain’s latest device… but the article instead ignored any of that and instead quoted some random conspiracy theory from a Twitter personality. Rather than rehashing the full story here, keep in mind that the geographic distribution and control of mining equipment is arguably as important as the aggregate network hashrate.
Their last sentence does not make much sense. How to define a hobbyist? If a hobbyist is defined as an individual who can afford to spend $4,195… then they can probably also buy ASIC equipment as well for other cryptocurrencies, including Ethereum today.
On p. 188 they write:
This range is a good baseline for the innovative investor to use for other cryptoassets to ensure they are secured with a similar level of cpaital spend as Bitcoin and Ethereum, which are the two best secured assets in the blockchain ecosystem.
There is another appearance of the “innovative investor,” remove in next edition?
Also, if security is solely measured by hashrate then yes, Bitcoin (BTC) and Ethereum (ETH) might be the “best secured.” But that assumes a purely Maginot Line attack and not a BGP or wrench attack.
On p. 189 they write:
Overall, hash rate is important, but so too is decentralization. After all, if the hash rate is extremely high but 75 percent of it is controlled by a single entity, then that is not a decentralized system. It is actually a highly centralized system and therefore vulnerable to the whims of that one entity.
This probably should come at the beginning of the chapter, not in this location. Also recommend adding some citations to the Onename and BGP posts.
On p. 189 they write:
It’s apparent that Litecoin is the most centralized, while Bitcoin is the most decentralized. A way to quanitfy the decentralization is the Herfindahl Hirschman Index (HHI), which is a metric to measure competition and market concentration.
HHI is used with known, legally identifiable parties. With cryptocurrencies such as Bitcoin, Litecoin, and Ethereum — the mining entities were not originally supposed to be known at all — over time they self-doxxed themselves.104
Should the Department of Justice and similar organizations coordinate and carry out HHI analysis on mining pools to prevent monopolization, oligopolization, and/or coordination? What happens if participants refuse to comply?
On p. 191 they write:
Blockchain networks should never classify as a highly concentrated marketplace, and ideally, should always fall into the competitive market place category.
Okay, but what if they don’t and no one cares? Who should enforce this?
At times, Bitcoin has been a moderately concentrated marketplace, just as Litecoin mining is currently a moderately concentrated marketplace. Litecoin recognizes the impact that large mining pools can have on the health of its ecosystem and the quality of its coin. To that point, Litecoin developers have instituted an awareness campaign called “Spread the Hashes” for those mining litecoin to consider spreading out their mining activies. The campaign recommends that litecoin computers mine with a variety of mining pools rather than concentraing solely in one.
The anthropomorphism needs to be removed in the second edition. “Litecoin” does not recognize anything because Litecoin is not a singular autonomous entity.
There are individual people, developers who work on a certain implementation of Litecoin that may promote something — and if they coordinate (which they do) then perhaps they could be classified as administrators.
Either way, this “Spread the Hashes” campaign didn’t seem to work:
As the pie chart above illustrates, just 5 entities currently account for about 90% of the network hashrate. And the largest 3 effectively could coordinate to control the network if they wanted to.
Worth noting that similar marketing campaigns to “spread the hashes” have been done on other networks. Back in 2014 when GHash.io reached the 50% mark, reddit was filled with discussions imploring miners to switch to P2Pool.
Not all nodes are made equal. A single node could have a large number of mining computers behind it, hence capturing a large percentage of the overall network’s hash rate, while another node could have mining computer supporting it, amounting to a tiny fraction of Bitcoin’s hash rate.
Sort of. There are two different nodes: nodes that fully validate and attempt to append the blockchain by submitting a proof-of-work that meets the necessary difficulty threshold… and nodes that don’t. In practice, today we call the former “mining pools” and the latter, just nodes.
For instance, in Bitcoinland there was a vicious war of words from 2015-2017 waged by several parties who did not operate mining pools, or nodes that generated proofs-of-work.105 One subset of these parties used various means and channels to insist that miners did not ultimately matter, that it was “users” who truly controlled the network and they labeled themselves “UASF.” And some of the most vocal members of this “populism wing” insisted that the nodes run by mining pools were no more important than the nodes run by some hobbyist in an apartment.
The views were irreconcilable and the ultimate result is that one group involved in that battle, forked off and created a new chain called Bitcoin Cash (BCH), whereas many of the other parties coalesced with what is called Bitcoin (BTC). There is a lot more to the story, a messy emotional divorce that still continues today.
Technically the decision to fork or not fork is made by mining pools and the nodes they each manage, but there are more nuances and politics involved that go beyond the scope of this review.
On p. 194 they write:
William Mougayar, author of The Business Blockchain, has written extensively about how to identify and evaluate new blockchain ventures and sums up the importance of developers succinctly: “Before users can trust the protocol, they need to trust the people who created it.” As we touched upon in the prior chapter, investigate the prior qualifications of lead developers for a protocol as much as possible.
Two problems with this:
I wrote a lengthy book review of Mougayar’s book and found it disappointing and do not recommend because of statements like the one above.
What were Satoshi’s qualifications? No one knows, but no one really cares either. Similarly, what were Vitalik Buterin’s qualifications? He was 19 when he announced Ethereum at Bitcoin Miami and had recently dropped out of college. Similarly, Gavin Wood was a 34 year-old developer building music-related apps prior to co-founding Ethereum. Would these two key guys been deemed qualified? What are the qualifications necessary to be a blockchain wizard?
On p. 194 they write:
Developers have their own network effect: the more smart developers there are working on a project, the more useful and intriguing that project becomes to other developers. These developers are then drawn to the project, and a positively reinforcing flywheel is created. On the other hand, if developers are exiting a project, then it quickly becomes less and less interesting to other developers, ultimately leaving no one to captain the software ship.
A couple of thoughts:
This is a nice sounding theory, but that’s not really what happens with most of these projects. Generally developers are attracted due to the compensation they can receive… they do a risk-reward analysis. I’ve met and spoken to dozens, perhaps north of 100 cryptocurrency-related teams in the past 12 months across the globe. Attracting talented developers is not nearly as easy and clear cut as the authors make it sound above.
Also, having a single “captain of the ship” seems like a single point of failure and a centralization risk. Is that part of the undefined ethos?
On p. 195 they write:
Recall that this is how Litecoin, Dash, and Zcash were created from Bitcoin: developers forked Bitcoin’s code, modified it, and then re-released the software with different functionality. Subscribers refer to people wanting to stay actively involved with the code. In short, the more code repository points, the more developer activity has occured around the cryptoasset’s code.
That’s not necessarily true, and in fact, has been gamed by coin issuers who want to make it look like there is a lot of independent activity and traction with developers… by creating spam accounts and very small changes to simple documents (like grammar).
Readers may also be interested in CoinGecko to see how this acitivity is weighted.
On p. 198 they write:
A different approach is to monitor the number of companies supporting a cryptoasset, which can be done by tracking venture capital investments. CoinDesk provides some of this information as seen in Figure 13.13. Though as we will address in Chapter 16 on ICOs, the trend in this space is moving away from venture funding and toward crowdfunding.
Actually, as mentioned a couple time earlier, there has been a noticeable divergence the past 12 months: coin sales that are done as private placements versus coin sales that have a public facing sale.
In general, most of the coins that have raised capital through private placement deals typically have less than 100 investors, many of which are the aforementioned “crypto hedge funds” and coin-focused venture funds such as Andreessen Horowitz and Union Square Ventures.
The public facing sales are generally eschewed by venture funds. If venture funds are involved in a coin that does a public sale, they typically are involved in what is called a “pre-sale” where they receive preferential terms and conditions, such as discounted coins.
Upon the conclusion of the “pre-sale” the actual public sale begins with heavy marketing on social media towards retail investors. Sometimes these sales have hundreds or even thousands of individual participants. That could be called a “crowdsale” and these participants typicallyget worse terms than those who participated in the pre-sale.
On p. 199 they write:
Another good proxy for the increased acceptance of a cryptoasset and its growing offering by highly regulated exchanges is the amount of fiat currency used to purchase it.
Maybe consider revising because we have all been told that cryptocurrencies would not only displace “fiat currency” but also topple and replace the existing financial system… how does measuring these new internet coins with old money help achieve that?
For instance, at the time of this writing none of the US-based retail exchanges with domestic bank accounts have recently listed an ICO (with the exception of ETH and ETC). This includes: itBit, Bitflyer, Coinbase, and Gemini.106 Kraken’s retail exchange uses payment processors and banking partners outside of the US.107
On p. 199 they write:
in the one-year period from March 2016 to March 2017, ether went from being traded 12 percent of the time with fiat currency to 50 percent of the time. This is a good sign of the maturation of an asset, and shows it is gaining wider recognition and acceptance.
Why is that specific ratio or percentage deemed good? The next edition should include a table explaining this in further because it is unclear why it is good, neutral, or bad.
On p. 201 they write about wallets from Blockchain.info:
Clearly, having more users that can hold a cryptoasset is good for that asset: more users, more usage, more acceptance. While the chart shows an exponential trend, there are a few drawbacks for this metric. For one, it only shows the growth of Blockchain.info’s wallet users, but many other wallet providers exist. For example, as of March 2017, Coinbase had 14.2 million wallets, on par with Blockchain.info. Second, an individual can have more than one wallet, so some of these numbers could be due to users creating many wallets, a flaw which extends to other wallet providers and their metrics as well.
In the past I have written extensively on how these headline wallet numbers are basically gimmicks and don’t accurately measure users or user activity.
Why? Because it costs nothing to open one. And often there is no KYC or AML involved in creating one as well. As a result, bots can be used to create many each day to inflate the metric.
Coinbase has actually removed usage data in the past and they still don’t define what the difference between a user or wallet is. Nor do either company provide traditional DAU / MAU metrics. It’s not hard to do and it is unclear why they don’t. The only way we have some semblance of an idea of what Coinbase user numbers were between 2013-2015 is because of the IRS lawsuit mentioned above.
On p. 201 they write about a search trend, “BTC USD,” first described by Willy Woo:
If we assume this to be true, then Woo’s analysis indicating a doubling in bitcoin users every year and an order of magnitude growth every 3.375 years. He calls this Woo’s Law in honor of Moore’s Law […] It will be interesting to see how Woo’s Law holds up over time.
How has it done? “Woo’s Law” has thus far not held up.
For instance, below is a 5 year trend chart of the same search term promoted by Woo and others last year:
As we can see above, this term has some correlation between interest in coins specifically during price bubbles. But this has not translated into large quantities of new daily users.108
The next edition of this book should remove this faux eponym because it has not withstood the test of time and doesn’t measure actual users.
On p. 202 they write:
Figure 13.17 shows the hyper growth of Ethereum’s unique address count. With Ethereum, an address can either store a balance of either, like Bitcoin, or it can store a smart contract. Either denotes an increase in use.
The next edition should include a caveat because it is unclear from this chart alone what kind of use is taking place. Is it coin shuffling, miner payouts, gambling payouts, Crypokitty activity, etc.? Maybe it is just someone spamming the network?
For instance, according to DappRadar which tracks 650 ethereum Dapps, over the past 24 hours there have only been 9,926 users sending 43,652 transactions. That may sound intriguing but… nearly about 2/3rd of all these users are using decentralized exchanges (DEX). If trading and arbitraging are the “killer apps” of cryptocurrencies, then the next edition of this book could be a lot slimmer than it is now.
As described in “Slicing data,” not all transactions are the same and a deep dive needs to be done to fully describe the behavior taking place.
But this is just an estimate from Blockchain.info and is likely widely exaggerated because Blockchain.info — like most wallet providers — probably has no idea what the intent behind those transactions are. We need data from all of the exchanges, payment processors, and merchants that accept coins in order to conclusively know what activity was commercial versus non-commercial in nature.
For instance, a large portion of those transactions could simply be “change address.”
Not to get too technical, but with Bitcoin, in order to manually send X amount of bitcoin on-chain, users typically must enter a “change address” unless the whole amount of UTXO is consumed. It’s kind of like a bank teller moving money from one till to another between shifts. No new economic activity is actually taking place in the bank or in the real economy, but in this specific chart above, there is no way to differentiate “change address” activity with real commercial activity and so it all gets mixed and muddied.
On p. 204 they write:
If the network value has outpaced the transactional volume of that asset, then this ratio will grow larger, which could imply the price of the asset has outpaced its utility. We call this the crypto “PE ratio,” taking inspiration from the common ratio used for equities.
Except, without a thorough deep dive from an analytics provider who has mapped out activity into all of the exchanges, payment processors, and merchants — it is very difficult to actually differentiate the noise from the actual transactional utility.109
Here the authors take all on-chain transaction volume at face value. The next edition should scrap this section unless they get access to a thorough deep dive.
On p. 204 they write:
One would assume that an efficient price for an asset would indicate a steadiness of network value to the transaction volume of the asset. Increasing transactional volume of an asset should be met by a similar increase in the value of that asset. Upside swings in pricing without similar swings in transaction volume could indicate an overheating of the market and thus, overvaluation of an asset.
In Figure 13.22 the top line is called the resistance line, indicating a price that bitcoin is having trouble breaking through. Often these lines can be numbers of psychological weight, in this case the $300 mark.
I looked it up and couldn’t find a definition for what “psychological weight” is, so this should either be defined in the book or removed in the next edition.110
On p. 209 they write:
You’ll find many instances of newer cryptoassets experiencing wild price swings after their creation, but over time these younger assets begin to follow the rules of technical analysis. This is a sign that these assets are maturing, and as such, are being followed by a broader group of traders. This indicates they can be more fully analyzed and evaluated using technical analysis, allowing the innovative investor to better time the market and identify buy and sell opportunities.
Technical analysis may have its uses but by itself it is basically cargo cult science.
Since cryptoassets are digital bearer instruments, they are unlike many other investments that are held by a centralized custodian. For example, regardless of which platform an investor uses to buy stocks, there is a centralized custodian who is “housing” the assets and keeping track of the investor’s balance. With cryptoassets, the innovative investor can opt for a similar situation or can have full autonomy and control in storage. The avenue chosen depends on what the innovative investor most values, and as with much of life there are always trade-offs.
This is true: there are many choice. But in practice, as noted above by Jonathan Levin, a significant majority of transactions typically involves a 3rd party intermediary.
Why? Because Securing a bearer instrument can be a major hassle, as a result companies like Coinbase and Xapo offer custodial services. While re-introducing an intermediary helps with coin management that kind of defeats the purpose of having a pseudonymous bearer asset in the first place.111 But that’s a different discussion.112
On p. 212 they write:
Anyone with a computer can connect to Bitcoin’s network, download past blocks, keep track of new transactions, and crunch the necessary data in pursuit of the gold hash. Such open architecture is one of Bitcoin’s strongest points.
It may sound like a irrelevant nitpick but this is not unique to Bitcoin. Nearly every cryptocurrency listed on Coinmarketcap has the same set of “features.” Similarly, many enterprise vendors also are open source and anyone could set up their own network with the software. Future editions should include a more nuanced definition of “open.”
On p. 213 they write:
The first computer – or mining rig – with ASIC chips that were specifically manufactured for the process was connected in January 2013.
The citation the authors included was for Avalon. This is true insomuch as these systems were available for purchase to the general retail public. But the first known ASIC-mining system was launched in late 2012: ASICMiner privately run out of Hong Kong (from BitQuan and BitFountain). 113
On p. 214 they write:
For perspective, the combined compute power of Bitcoin’s network is over 100,000 times faster than the top 500 supercomputers in the world combined.
This type of stat is frequently repeated throughout the Bitcoin world but it is not an apples-to-apples comparison and should be removed in the next edition. The supercomputers are largely comprised of CPUs and GPUs which — as their names suggest — are flexible and capable of handling many different types of general purpose tasks.
ASICs on the other hand, are focused and specialized: capable of doing just one set of tasks over and over. ASICs found in a Bitcoin mining farm are not even capable of creating blocks to propagate on the network: they simply generate hashes. That is how limited they are in functionality.
On p. 214 they write:
Conceptually, mining networks are a perfect competition, and thus as margins increase, new participants will flood in until economic equilibrium is once again achieved. Thus the greater the value of the asset, the more money miners make, which draws new miners into the ecosystem, thereby increasing the security of the network. It’s a virtuous cycle that ensures the bigger the network value of a cryptoasset, the more security there is to support it.
I think this could be rewritten in the next edition to be closer with what happens in practice.114
For instance, as coin prices decrease, margins are squeezed and “marginal” operators exit, leaving fewer overall miners. In the past this has led to bankruptcies, such as KnC and HashFast.
Does this lead to a less secure network?
Maybe, maybe not. Depends on how we define secure and insecure. Pure hashrate is just one attribute… geographical location, amount of participants, and diversity of participants could be others as well. For example, see the discussion earlier on selfish-mining.
On p. 215 they write:
Before investing in a cloud-based mining pool, conduct research on the potential investment. If it sounds too good to be true, it probably is.
This is good advice.
Also worth mentioning that “cloud-based mining” kind of the defeats the purpose of pseudonymous mining. If you have to trust the infrastructure provider to manage and operate the hashing equipment, why not just buy the coins? Why take that risk and also have to divulge your identity?
Incidentally, NiceHash is one of the most well-known cloud mining services available today. It partly cemented its notoriety (this is not an endorsement) as its mining units have been rented and used to attack several different cryptocurrencies. A site called Crypto51.app categorizes the costs of doing a brute force attack on dozens of coins and even lists the amount of hashrate NiceHash has in order to perform a hypothetical attack.
On p. 216 they write:
However, Ethereum will potentially switch to proof-of-stake early in 2018, as it is more efficient from an energy perspective, and therefore many claim is more scalable.
Quick note: this transition has been delayed again until at least the end of 2018 and more likely sometime in 2019 (although it has been moved many times before as well).
On p. 217 they write:
To this end, today numerous quality exchange are available to investors looking to gain and transact the more than 800 cryptoassets that currently exist.
In the next edition it is worth clarifying and defining what “quality” means because just about every retail / consumer-facing exchange has had its share of problems, including hacks and thefts.115 This is one of the reasons the SEC has denied ETF proposals.
With that said, there are a number of OTC trading desks run by reputable financial organizations that enable investors to trade, however, typically the minimum order size (buy/sell) is $100,000.116
On p. 218 they write:
Cryptoasset transactions are irreversible; therefore chargebacks are impossible. While an irreversible transaction may sound scary, it actually benefits the efficiency of the overall system. With credit card chargebacks, everyone has to bear the cost, whereas with cryptoassets only those who are careless bear the cost.
Two comments worth considering for the next edition:
Transactions in cryptocurrencies are possible through block reversals, which can and do happen. Often times they are relatively expensive to do, but during a “51% attack” it can occur, thus it is not impossible. In fact, as part of the Nano class action lawsuit, one of the suggested remedies is a roll-back.
As far as credit card chargebacks: this is largely borne by the merchant (not everybody). In fact, charge backs are largely a consumer-friendly feature, a type of insurance.117
On p. 221 they discuss insurance at exchanges.
At this time, no retail cryptocurrency exchange actually insures a users coin deposit. As a result, most custodians and intermediaries have had to self-insure (e.g., create their own insurance entity). There are institutional products (vaults) which are attempting to get 3rd party insurance.
Prior to the hack, Bitfinex had settled with the CFTC for $75,000 primarily because its cold storage of bitcoin ran afoul of CFTC regulations. The move to place all clients’ assets into hot wallets is cited by many as due to the fine and CFTC regulations. Either way, this hack proved that no matter the security protocols put in place, hot wallets are always more insecure than properly executed cold storage because the hot wallet can be accesssed from afar by anyone with an Internet connection.
This passage should be revised in the next edition for a few reasons:
First, as mentioned earlier, Bitcoiners like to find a good boogeyman and in this hacking incident, they blamed the CFTC.
For instance, I reached out to Zane Tackett who — at the time — was head of communications for Bitfinex.
According to Tackett: “We migrated to the bitgo setup before any discussions or anything with the CFTC happened”
I then publicly pointed out, to Antonopoulos and others, that the CFTC blame game was false. But instead of deleting that tweet and focusing on who actually hacked Bitfinex, the ideological wing of the Bitcoin tribe continues to push this false narrative.
Tackett even explicitly answered this question in detail on reddit that same day.
So either Tackett is lying or Antonopoulos is wrong. In this case, it is likely the latter.
The second point worth adding to the passage above in the book is that after nearly two years we still haven’t been told exactly what happened with the hack and theft. This, despite the fact that Bitfinex has said on more than one occasion that it would provide an audit and public explanation.
An ETF is arguably the best investment vehicle to house bitcoin.
This is debatable. Last year Jack Bogle – founder of Vanguard, a firm that popularized broad market index ETFs – implored the public to avoid bitcoin like the plague for several reasons. Critics say he is out of touch, but even if that were true that doesn’t mean his expert views on structuring ETFs should be dismissed.
On p. 238 they write:
Regardless of what people expected going into the SEC decision most everyone was taken aback by the rigidity of the SEC’s rejection. Notably the SEC didn’t spend much time on the specifics of the Winklevoss ETF but focused more on the overarching nature of the bitcoin markets. Saying that these markets were unregulated was an extra slap to the Winklevosses, who had spent significant time and money on setting up the stringently regulated Gemini exchange. In focusing on the bitcoin markets at large, the rejection implied that an ETF will not happen in the United States for some time.
For the next edition, this paragraph should probably be removed.
The facts of the Bitcoin markets today are as follows:
Mining is the process of minting new coins as well as processing transactions and… is largely unregulated in any jurisdiction.
Many exchanges, in particular those outside the US, comply with a hodge podge of regulations, often without the same strict KYC / AML / sanctions checks required for US exchanges.
Gemini and the Winklevoss have no ability to police these unregulated trading venues and unregulated coin minters. That probably won’t change in the near future.
Perhaps the SEC will eventually approve an ETF, but they arguably were not being rigid — they were being practical. In their view: why allow an unregulated asset whose underlying genesis and trading market is still very opaque and frequently is used for illicit activity?
Lastly the next edition should include a citation for who “most everyone” includes, because in my own anecdotal experience, the majority of traders at US exchanges I interact with did not think it would be allowed at that time. Note: my deep dive on the COIN ETF and its ever changing history, can be found here.
On p. 238 they write:
On Monday, naysarers were faced with the reality that bitcoin was once again back over $1,200, and the network for all cryptoassets had increased $4 billion since the SEC decision. Yes, $4 billion in three days.
A couple of thoughts:
Typo: naysarers should be naysayers
Recommend removing this sentence in the next edition because the attitude comes off as a little smug and has an ad hominem. People are allowed to have different views on the adoption of technology which is separate from what the price of a coin will be. And justifying a trading position based on price movements which are based on the mood of retail investors should probably not be the takeaway message for a mainstream book.
On p. 240 they write:
By purchasing XBT Provider, GABI strengthened the reliability of the counterparty to the bitcoin ETNs and added a nice asset to its growing bitcoin investing platform for institutions.
For the next edition, recommend removing “nice” because that is a subjective word. There are other ways to describe this acquisition.
On p. 242 they write:
It also created an independent advisory committee, including bitcoin evangelist Andreas Antonopoulos to oversee its pricing model, which utilized prices from various exchanges throughout the world.
Why is this specific person considered an expert on futures? There are a lot of articulate developers involved in promoting cryptocurrencies, but their expertise is typically not in finance. If anything, this specific person has a vocal disdain for regulators, financial institutions, and regulated instruments… just see his tweet above in Chapter 14.119
Maybe in the next edition discuss the controversy of having a futures contract that is not physically deliverable. Could also include how the CFTC has subpoenaed the four partner exchanges working with the CME: Coinbase, Kraken, itBit, and Bitstamp. These four exchanges create the price used in bitcoin futures by the CME.
On p. 249 they write:
For first-time founders who want to approach venture capitalists for an investment, often they must know someone-who-knows-someone. Having such a connection allows for a warm introduction as opposed to being among the hundreds of cold calls that venture capitalists inevitably receive. To know someone-who-knows-someone requires already being in the know, which creates a catch-22.
This is a very good point. However, it would be worth adding in the next version how most ICOs and coin sales now require knowing someone because most private sales involve roughly the same insular, exclusive set of funds and investors as the “old method” did.
On p. 252 they write:
Before we dive into the specifics of how a cryptoasset offering is carried out, the innovative investor needs to understand that the model of crowdfunding cryptoassets is doubly disruptive. By leveraging crowdfunding, cryptoasset offering are creating room for the average investor to stand alongside venture capitalists, and the crowdfunding structure is potentially obviating the need for venture capitalists and the capital markets entirely.
In the next edition, worth mentioning that this was the general pitch for ICOs starting with Mastercoin (2013) all the way up through 2016. But over the past two years and certainly in the past 12 months it has dramatically shifted back towards the traditional venture route.
One of the reasons why is because of the filtering and diligence process. Those that don’t get selected and/or those ICOs that don’t meet the requirements of this small group of funds often decide to do a public sale. And many of these ideas were half-baked and sometimes fraudulent, according to one recent report: More Than Three-Quarters of ICOs Were Scams
On p. 253 they write:
Monegro’s thesis is as follows: The Web is supported by protocols like the transmission control protocol/Internet protocol (TCP/IP), the hypertext transfer protocol (HTTP), and simple mail transfer protocol (SMTP), all of which have become standards for routing information around the internet. However, these protocols are commotidized, in that while they form the backbone of our internet, they are poorly monetized.
It could be argued that Monegro’s thesis has failed to live up to its hype thus far. And counterfactually, if “tcpipcoin” existed, it may have actually stunted the growth of the internet as Vinton Cerf and Bob Kahn would have allocated more time promoting the coin rather than the technology. We can disagree about this alternative scenario, but I have mentioned it before in Section 8.
For example, we frequently see that dozens of nonsensical conferences and meetups conducted on a weekly basis globally try to promote a shiny new protocol coin of some kind. Trying to monetize a public good with a coin thus far has not removed the traditional incentive and sustainability issues around a public good. That would also be worth discussing in the next edition.120
On p. 253 they write:
All the applications like Coinbase, OpenBazaar, and Purse.io rely on Bitcoin, which drives up the value of bitcoin.
Worth updating this because Purse.io added support to Bitcoin Cash. And OpenBazaar switched over to Bitcoin Cash altogether.
Also, Coinbase has become less maximalist over time and now provides trading support for four different coins.121 Though it probably wouldn’t be technically correct to call Coinbase or Purse a Bitcoin application. In the case of Coinbase, users use an off-chain database to interact and Coinbase controls the private key as a custodian / deposit-taking institution.
On p. 254 they write:
Interestingly, once these blockchain protocols are released, they take on lives of their own. While some are supported by foundations, like the Ethereum Foundation or Zcash Foundation, the protocols themselves are not companies. They don’t have income statements, cash flows, or shareholders they report to. The creation of these foundations is intended to help the protocol by providing some level structure and organization, but the protocol’s value does not depend on the foundation.
This is another reason to heavily modify chapter 7 in future versions because it is not an apples-to-apples comparison: coins and coin foundations are not the same thing as for-profit companies that issue regulated instruments (stocks, bonds, etc.).
Also, the very last sentence is highly debatable because of how often foundation and foundation staff are integral to the longevity of a coin.
Recall that blockchains do not maintain or market themselves, people do. And is often the case: staff and contractors of these foundations frequently use social media to promote potential upgrades as well as publicize the coins attributes to a wider audience. In many cases it could be the case that the protocol’s value does depend on the work and efforts of others including specifically those at a coin foundation.122
On p. 254 they write:
Furthermore, as open-source software projects, anyone with the proper merits can join the protocol development team. These protocols have not need for the capital markets because they create self-reinforcing economic ecosystems. The more people use the protocol, the more valuable the native assets within it become, drawing more people to use the protocol, creating a self-reinforcing positive feedback loop. Often, core protocol developers will also work for a company that provides application(s) that use the protocol, and that is a way for the protocol developers to get paid over the long term. They can also benefit from holding the native asset since inception.
There are several points here that should be modified or removed in the next edition:
For instance, with Bitcoin, due to a variety of political fights and personality conflicts, multiple “core” developers have had their access rights removed including: Jeff Garzik, Mike Hearn, Gavin Andresen, and Alex Waters. Thus it is not true that anyone can join a team. It is also unclear what those merits may be as most of the projects don’t explicitly provide those in written format yet.
In addition, internet coins are often traded on secondary markets in order to provide liquidity to coin holders such as developers. They all need access to capital markets to stay afloat. No project is self-sustainable at this time because no coin is being used as a unit of account — miners and developers must liquidate coins in order to pay their bills which are denominated in foreign currency.
Lastly, in practice, there are many coins that have died or lost any developer support yet initially they may have had a small army of programmers and media attention. According to Coinopsy, more than 1,000 coins are dead. Thus in the next edition the “self-reinforcing” loop should probably be removed too.
On p. 256 they write:
ICOs have a fixed start and end date, and often there is a bonus structure involved with investing earlier. For instance, investing at an early stage may get an investor 10 to 20 percent more of a cryptoasset. The bonus structure is meant to incentivize people to buy in early, which helps to assure that the ICO will hit its target offering. There’s nothing like bonuses followed by scarcity to drive people to buy.
This should definitely be removed. In May, the SEC released a parody website called “HoweyCoins” which explicitly points to this precise FOMO behavior as a big no-no for both issuers and investors alike.
Also recommend the inclusion of the Munchee Order in this chapter as it would help illustrate what regulators such as the SEC perceive as improper fundraising techniques. Specifically, include this in the “announcing the ICO” section.
On p. 258 and 259 they discuss the Howey Test. It is strongly recommended that these two pages be reworded and modified based on the enforcement actions and guidance from the SEC and other securities regulators.
For instance, they write:
A joint effort by Coinbase, Coin Center, ConsenSys, and Union Square Ventures with the legal assistance of Debevoise & Plimpton LLP, produced a document called, “A Securities Law Framework for Blockchain Tokens.” It is especially important for the team behind an ICO to utilize this document in conjunction with a lawyer to determine if a cryptoasset sale falls under SEC jurisdiction. The SEC made it clear in July 2017 that some cryptoassets can be considered securities.
The first sentence should probably be moved into a footnote and the second sentence removed altogether because this document did not age well.
In fact, the current version of the document – as it exists on Coinbase – informs readers in bright red that:
Please note that since this document was originally published on December 7, 2016, the regulatory landscape has changed. The information contained in this document, including the Framework may no longer be accurate. You should not rely on this document as legal advice and you should seek advice from your own counsel, who is familiar with the particular facts and circumstances of what you intend and can give you tailored advice. This Framework is provided “as is” with no representations, warranties or obligations to update, although we reserve the right to modify or change this Framework from time to time. No attorney-client relationship or privilege is created, nor is this intended to be attorney advertising in any jurisdiction.
On p. 259 they write:
Does the token sale tout itself as an investment? It should instead be promoted for its functionality and use case and include appropriate disclaimers that identify it as a product, not an investment.
This is arguably not good advice and should be removed. Why? Courts in the US will likely see through this euphemism. For other things not to do, recommend reading the ICO Whitepaper Whitepaper from Stephen Palley.
On p. 260 they write:
One of the oldest groups of angel investors in the blockchain and bitcoin space is called BitAngels. Michael Terpin of BitAngels has been active in angel investing in blockchain companies for as long as the opportunities have existed. Terpin’s annual conference, CoinAgenda, is one of the best opportunities for investors to see and hear management from blockchain startups present their ideas and business models.
For the next edition, I’d reconsider including this type of endorsement.123 There are some interesting stories that involving these specific entities worthy of a different post.
On p. 263 they write:
For instance, if Bitcoin influences how remittances are handled, what impact may that have on stocks like Western Union, a remittances kingpin? If Ethereum takes off as a decentralized world computer, will that have any effect on companies with cloud computing offerings, such as Amazon, Microsoft, and Google? If companies can get paid more quickly with lower transaction fees using the latest cryptocurrency, will that have an impact on credit card providers like Visa and American Express.
For the next edition, this paragraph — or at least argument — should come earlier, perhaps even in Chapter 7 (since there is a discussion of specific publicly traded companies).
Another thing that should have been added to this section is actual stock prices for say, the past five years of the companies mentioned: Western Union, Visa, and American Express.
I have included those three below:
If the narrative is that Bitcoin or the “latest cryptocurrency” will erode the margins and even business models of existing payment providers, then at some point that should be reflected in their share prices.
As shown above, that does not seem to be the case (yet).
Perhaps that will change in the future, but consider this: all three of the companies above have either directly invested in and/or are collaborating in blockchain-related platforms — most of which do not involve any coin. Perhaps these firms never use a blockchain. In fact, maybe they find blockchains to be unhelpful as infrastructure altogether.
That is possible, hence the need to update this chapter to reflect the actual realities.
In addition, the other three companies listed by the authors have publicly discussed various blockchain-related efforts beyond just pilot offerings.
For instance, both Amazon and Microsoft have supported blockchain-as-a-service (BaaS) offerings in production for over a year. Google has been a laggard but has internal projects attempting to leverage some of these ideas as well.
On p. 266 they write:
In 2016, the father-son team of Don and Alex Tapscott published the book Blockchain Revolution: How the Technology behind Bitcoin Is Changing Money, Business, and the World, and William Mougayar published the book, The Business Blockchain: Promise, Practice, and Application of the Next Internet Technology.
I wrote lengthy reviews of both. The short summary is that both were fairly superficial in their dive into use cases and vendors. The Mougayar book felt like it could use a lot more detailed meat. The Tapscott book was riddled with errors and unproven assertions. Would reconsider citing them in the next edition (unless they each dramatically update their content).
On p. 266 they write:
For companies pursuing a DLT strategy, they will utilize many of the innovations put forth by the developers of public blockchains, but they don’t have to associate themselves with those groups or share their networks. They pick and choose the parts of the software they want to use and run it on their own hardware in their own networks, similar to intranets (earlier referred to as private, permissioned blockchains).
These are pretty broad sweeping comments that should be modified in the next edition. Not every vendor or platform provider uses the same type of chain or ledger. These are not commoditized (yet).
There are many nuances and trade-offs for each platform. For the next edition, it would be helpful worth doing a comparison of: Fabric, Pantheon, Quorum, Corda, and other enterprise-focused platforms. In some cases, they may have an on-premise requirement and in others, nodes can run in a public cloud.
We see many DLT solutions as band-aids to the coming disruption. While DLT will help streamline existing processes–which will help profit margins in the short term–for the most part these solutions operate within what will become increasingly outdated business models.
Perhaps that it is true, but again, this language is very broad sweeping and definitive. It needs citations and references in the next edition.
On p. 267 they write:
The incumbents protect themselves by dismissing cryptoassets, a popular example being JPMorgan’s Jamie Dimon, who famously claimed bitcoin was “going to be stopped.” Mr. Dimon and other financial incumbents who dismiss cryptoassets are playing exactly to the precarious mold that Christensen outlines:
Disruptive technologies like cryptoassets initially gain traction because they’re “cheaper, simpler, smaller.” This early traction occurs on the fringe, not in the mainstream, which allows incumbents like Mr. Dimon to dismiss them. But cheaper, simpler, smaller things rarely stay on the fringe, and the shift to mainstream can be swift, catching the incumbents off guard.
For the next edition it would be good to remove the misconceptions repeated in the statement above. Jamie Dimon was specifically dismissing the exuberance of coin mania, not the idea of enhancing IT operations with something like a blockchain.
Worth adding to future versions: JPMorgan has financial sponsored Quorum, an open-source fork of Ethereum modified for enterprise-related uses. The bank has also invested in Digital Asset. It is also a member of three industry organizations: EEA, Hyperledger, and IC3. In addition, JP Morgan has filed blockchain-related patents, has launched a blockchain-based payment network with several banking partners, and also partnered with the parent company of Zcash to integrate ZSL into Quorum.
While Jamie Dimon may not share the same bullish views about coins as the authors do, the firm he is the CEO seems to be taking “blockchains” seriously.
On p. 267 they write:
One area long discussed as ripe for disruption is the personal remittances market, where individuals who work outside of their home countries send money back home to provide for their families.
This specific use case is a bit repetitive as it has been mentioned 5-6 times before in other chapters. Should probably remove this in future editions unless there is something different to add that wasn’t already explained before.
On p. 268 they write:
It’s no stretch then to recognize that bitcoin, with its low cost, high speed, and a network that operates 24/7, could be the preferred currency for these types of international transactions. Of course, there are requirements to make this happen. The recipient needs to have a bitcoin wallet, or a business needs to serve as an intermediary, to ultimately get the funds to the recipient. While the latter option creates a new-age middleman–which potentially has its own set of problems–thus far these middlemen have provided to be much less costly than Western Union. The middleman can be a pawnshop owner with a cell phone, who receives the bitcoin and pays out local currency to the intended recipient.
This should be modified in the next versions because it is a stretch to make those claims. That is the reason why multiple Bitcoin-focused remittance companies have pivoted or branched out because “moving” bitcoins across borders is the only easy part of the entire process. For instance, the KYC / AML checks during the on- and off-ramps are costly and are required in most countries. This should be included in any analysis.
Also, there are no citations in this paragraph. And the last sentence is describing the pawnshop owner as a money transmitter / money service business which is a regulated operation. Maybe the laws change, which is possible. But for the next version, the authors should include specific corridors and the costs and margins for MSBs operating in those corridors.
Lastly, any future analysis on this topic should also include the online and app-based product offerings from traditional remittance players such as Western Union. In nearly all cases, these products and services are faster and cheaper in the same corridors relative to traditional in-person visits.
The impact of this major disruption in teh remittance market should be recognized by the innovative investor not only because of the threat it creates to a publicly traded company like Western Union (WU) but for the opportunities it provides as well.
It is strange to hear this repeated multiple times without providing quantifiable specifics on how to measure this threat.
As mentioned a few pages earlier, if competitors (including, hypothetically cryptocurrencies) were to erode the margins of publicly traded companies, we should be able to see that eventually reflected in the share price. But Western Union has been doing more or less the same as it has the past couple of years.
What about others?
Above is the five year performance of Moneygram, another remittance service provider.
What happened the past two years? Did Bitcoin or another cryptocurrency pound its share value into the ground? Nope.
What happened is that one of Alibaba’s affiliates – Ant Financial – attempted to acquire Moneygram. First announced in early January 2017, Ant Financial wanted to acquire it for $880 million. Despite approval from the Moneygram board, the deal faced scrutiny from US regulators. Then in January 2018, the deal was axed as the US government blocked the transaction on national security grounds.
This hasn’t stopped Alibaba and its affiliates with finding other areas to grow. For instance, last month Alipay (part of Ant Financial) announced it had partnered with G Cash to in the Hong Kong – Philippines corridor, using a blockchain platform for remittances. No coin was needed in this process so far.
There may be some success stories of new and old MSBs that utilize cryptocurrencies in ways that make them more competitive, those should be included in the next edition along with more metrics readers can compare.124
On p. 270 they write:
For the long term investor, careful analysis should be undertaken to understand if insurance companies are pursing DLT use cases that will provide a lasting and meaningful solution. Lastly, some of the major consulting firms may be so entrenched in incumbent ideology that they too may be blind to the coming distruption.
A few comments that should be finnesed in the next version:
Similarly, every major consulting company and systems integrator has a team or two dedicated to helping clients build and integrate applications with specific enterprise-related “blockchain” platforms. Many of them have joined related consortia too. There are too many to even list here so it is unlikely they will get collectively blind-sighted as alluded to in the passage above.
On pgs. 272 and 273 they write about consortia:
Another consortium, The Hyperledger Project, offers more open membership than R3. Remember, one of the strengths and defining aspects of an effective blockchain project is its open source ethos.
While the [EEA] consoritum will work on software outside of Ethereum’s public blockchain, the intent is for all software to remain interoperable in case companies want to utilize Ethereum’s open network in the future.
Based on the passages above the next edition should incorporate a few changes.
The Hyperledger Project (HLP) is a non-profit group that does not itself aim to commercialize or deploy or operate any technology.125 The membership dues are largely used to maintain code repositories and sponsor events which educate attendees on projects incubated within HLP. It currently has around 200 members, including R3 which was a founding member. There are more than 5 codebases that are officially incubated, the most well-known is Fabric. However, HLP seeks to maintain a neutral position on which platform its members should use. Other notable platforms incubated within HLP include Iroha and Sawtooth (Lake).
In contrast, R3 is a for-profit company that set up a consortium in order to commercialize and deploy technology within the regulated financial industry.126 Its membership model has changed over time and it is the main sponsor for Corda, an open source platform. The consortium composition initially started with 42 banks and now includes about 200 entities including insurance companies, central banks, financial market infrastructure operators, and others.
The third most known consortium is the Enterprise Ethereum Alliance (EEA). It is kind of like the combination of the two above. It is a non-profit organization and itself does not aim to commercialize or deploy or operate any technology. It seeks to be a neutral entity within the greater Ethereum ecosystem and has many different working groups that span topics similar as the other two consortia above. It has hundreds of members and the main efforts have been around formalizing an enterprise-focused specification (EEA 1.0) that other vendors can create implementations of (such as Pantheon).
Like the members of the other two consortia above, nothing prevents an EEA member from using any other platform. Thus the authors usage of “open network” is superfluous because all of the codebases in each of these three consortia is open, anyone can download and use. The key differences are: what are the trade-offs with using each platform versus what are the benefits of membership for joining the consortia. These are two separate points that could be discussed further in the next edition.
On p. 276 they write:
The CFTC Director of Enforcement, Aitan Goelman, tried to clarify his opinion with this satement, “While there is a lot of excitement surrounding bitcoin and other virtual currencies, innovation does not excuse those acting in this space from following the same rules applicable to all participants in the commodity derivatives markets.” It is clearly confusing that the Direct of Enforcement of the agency that ruled bitcoin a commodity also called it a “virtual currency.”
For the next edition the authors should remove the unnecessary attitude in the last sentence.
Up through 2017, most US and even foreign regulators used the term “virtual currency” — not as a slight against Bitcoin or cryptocurrencies, but because that was the catchall term of art used for many years.
For instance, in March 2013, FinCEN released its guidance and it was entitled: “Application of FinCEN’s Regulations to Persons Administering, Exchanging, or Using Virtual Currencies”
Throughout the guidance, the term “virtual currency” is used more than 30 times.
And one relevant passage – especially for this book review – involves the definition of an administrator. According to FinCEN’s guidance:
“An administrator is a person engaged as a business in issuing (putting into circulation) a virtual currency, and who has the authority to redeem (to withdraw from circulation) such virtual currency.”
As it relates to the CFTC, earlier this year a federal judge in New York ruled that: “virtual currencies can be regulated by CFTC as a commodity.”
The ruling (pdf) specifically uses the phrase “virtual currency” not as a slight, but as a term of art. Perhaps other terms are used over time. For instance, in its new customer advisory issued this week, the CFTC mentioned potential scams that describe themselves as “utility coins” or “consumption coins.” Worth revisiting in the next edition.
On p. 280 they write:
Here’s another Burniske-Tatar Rule: Don’t invest in bitcoin, ether, or any other cryptoasset just because it’s doubled or tripled in the last week. Before investing, be able to explain the basics of the asset to a friend and ascertain if it fits well given the risk profile and goals of your investment portfolio.
This is good advice. And while the eponymous rule was coined several chapters ago, future editions should probably drop the name of that rule… because similar advice with slightly different wording has existed for decades (e.g., don’t invest more than you can afford to lose, do your own research, etc.).
On p. 282 they write:
Are millenials turning to bitcoin and cryptoassets for their investment? Is a Vanguard fund or a small investment in Apple any better? Whereas the Vanguard fund has a minimum investment amount and buying an equity will require commission, millennials see cryptoasset markets as a way to begin investing with a modest amount of money and in small increments, which is is often not possible with stocks or funds.
They also include a footnote that reads:
Each bitcoin can be divided into 100 million units, making it easy to buy 1/2, 1/10, 1/100 or 1/1000 of a bitcoin
Would recommend removing this passage altogether because there really aren’t many good surveys that indicate who actually bought coins versus who was just interested in them.
For instance, a flawed Finder.com survey that is still being cited, says that 8% of Americans have invested in cryptocurrencies.127 While it says the majority of investors are “millenials,” the survey doesn’t ask the most important question: does the investor control the private key. If you do not control the private key then you do not control the coin, someone else does.
In addition, there are online brokerages that do allow investors to invest with modest amounts, the most notable being Robinhood (which coincidentally also allows users to purchase several different cryptocurrencies). There are also a variety of spare change investment apps and robo-advisor products that allow users to have some exposure to regulated capital market too.
Lastly, regarding the footnote they provide: due to the fees required by Bitcoin miners, in practice over the past several months 1/1000 of a bitcoin is typically the minimum transaction fee. This is one reason why many investors simply leave coins on cryptocurrency exchanges: so they don’t have to pay fees to move them to other wallets.128
On p. 282 they write:
The important point is that at least they’re doing something to invest their funds and build the groundwork for a healthy financial future. We have seen firsthand millenials who have learned about investing from buying cryptoassets and have implemented investing approaches, such as taking profits at certain price points, seeking diversification into multiple assets, and so on.
This should probably be removed too because the same thing can be said to a new cohort of investors twenty years ago, such as the ones that invested in dotcom-related companies. Who remembers Beenz?
I fully expect some reaction towards this review along the lines that it was too picky or too pedantic. Perhaps this a little true but consider: what is the right size for a thorough book review in the age of so-so fact-checking?129 Also, most of my previous reviews were about the same length, or at least used the same page-by-page model.
There is obvious room for disagreement in areas involving opinions, but there are many technical and non-technical mistakes that the authors made, not just a small handful. By highlighting these, not only could the next edition be significantly improved but it helps readers new to this space get a better understanding of what the prevalent themes versus realities are.
The goal of this review was not to be overbearing but to be dispassionate about supposed common wisdom promoted in the cryptocurrency world.
For example, just the other day I noticed in a chatroom the following statement from a maximalist:
HODLer = DAU. Bitcoin has the most DAUs on any protocol.
Several people in the room agreed with those this statement and they are not alone. If the reader is interested in learning about the sociology and subculture of many Bitcoin enthusiasts, its worth skimming reddit and twitter occasionally to see how passionate coin investors think.130
But for businesspeople who are not part of the inner sanctum of Bitcoinland, the statement above from the chatroom may make you shrug.
After all, HODLing a dollar doesn’t make you a dollar user. HODLing a barrel of oil doesn’t make you a oil user. HODLing a brick of gold doesn’t make you a gold user. HODLing a digitized Pokemon card doesn’t make you a Pokemon user. HODLing a Stradivarius violin doesn’t make you a violin player. HODLing an Olympic medal doesn’t make you an Olympic athlete. And so forth. The valuation of an auction house isn’t measured by the amount of rare collectibles it sells in a day, why should internet coins and their platforms be an exception to that rule?131
Inactivity isn’t how activity is measured. Or to look at this argument from another angle: HODLing is not ‘active’ anything. If all an investor did was buy bitcoin and then lose their keys, they would accomplish the same thing described in the chatroom.132
Sure it is possible to redefine what Bitcoin or cryptocurrencies are supposed to do, but that’s after the fact. For example, if Satoshi had wanted to explicitly build “digital gold” he/she would likely have mentioned it in the original paper at least once and even architected Bitcoin to be something different than what it looked like in 2009.133 As mentioned above, the first app he looked at building was for poker.
This is definitely a topic worth including in the next edition, but I digress.134
Other general areas for improvement:
Add a glossary.
Add financial disclosures of coins owned by each author.
Provide specific definitions for vague terms like “the community,” “administrator,” and the attributes of a target investor; ditch the “innovative” investor nomenclature.
Chapter 7 probably should be removed until more accurate comparisons can be found and Chapter 17 seemed a bit unfocused and covered a wide array of topics instead of just one or two… even dropping in thoughts about regulators. Future versions likely need an entire set of chapters focused on regulations, not just mentioned in passing.
Based on the incorrect view of financing mentioned in Chapter 5, interview Vitalik Buterin and other co-founders regarding how Ethereum was bootstrapped.
In one of the future regulatory chapters, would be good to have a discussion around PFMI, CBDCs, and settlement finality.
Provide a lot more references and citations regarding cryptocurrency-focused use cases, especially remittance providers. This seemed to be the most repeated use case but nary a mention of a specific Bitcoin remittance company, its valuation, or volume corresponding to the use case.
Have a book or paper you’d like me to look at? Feel free to send it across. Also, it just came out but this one sounds like a doozy already. See my other book reviews.
Send to Kindle
To be fair, Burniske is not the only analyst-turned-VC who has not publicly disclosed his trading positions of coins, but that’s a separate topic. [↩]
One reviewer mentioned: “Likely it was partially intentional to release in late 2008 / early 2009, but did in fact coincide mainly with internal constraints. We could also argue that the GFC commenced in mid-2007 when BNP Paribas froze two mortgage-backed security funds which became the catalyst of the summer 2007 credit crunch, but that is neither here nor there. I also debate the argument that it was ‘intended’ as anything other than a solution to the double-spend problem, be it a payments system or an investment.” [↩]
As an aside, Brian Kelly, frequently promotes various coins on CNBC. Unclear what his trading positions are on each coin at the time of recording. While that may not be illegal, it’s arguably not classy. [↩]
One reviewer mentioned: “This was literally the ethos that led to the GFC. Securitization and Mark-to-model were heralded as “innovation” and championed for their ability to move faster than the academic foundation and until 2007 seen as a way to ‘completely engineer risk out of from the system.'” [↩]
One reviewer mentioned: “The authors also miss that “value” is still a function of ‘the market’, i.e. supply and demand. Simply by fixing supply does not equalize demand. I also take massive issue with the governance in “a [de]centralized and democratic manner.” Are the authors able to write C++ or GOLang protocol code for Bitcoin Core or GETH? Likely not. So if anything this walks us towards a new form of governance, except where we elect leaders in the US who ultimately appoint Fed governors in cryptocurrencies there are generally no elections. Long story short, in all cases, it ain’t democratic and it probably remained at least partially centralised at a given point in time.” [↩]
See Central bank digital currencies from the BIS. I know, I’ll get spammed by all the “sound money” promoters out there who insist that Bitcoin will replace central banks — it’s a religious zeal to many. [↩]
For example, about a month ago, Jonathan Levin from Chainalysis did an interview and mentioned that: “So we can identify, it is quite hard to know how many people. I would say that 80% of transactions that occur on these cryptocurrency ledgers have a counterparty that is a 3rd party service. More than 80%.” [↩]
For instance, on p. xxvi they list “the top 50” coins at the end of 2016 and don’t disclose if they own any specific ones at all, but talk about many of them in positive ways. Adding a disclosure would be helpful. [↩]
The Economist wrote a nice short article on this behavior — the greater fool – last year. [↩]
For example, on p. 9 they write: “Shortly thereafter, Satoshi vanished. Some speculate it was for the good of Bitcoin. After all, being the creator of a technology that has the potential to replace much of the current financial system is bound to eventually invoke the wrath of powerful government and private sector forces.” This seems like a strawman. Bitcoin was designed for just one simple thing: payments. The financial system is an interwoven network of hundreds of regulated and unregulated goods and services, not just payments. Also, this paragraph, like a few others later, has elements of conspiratorial boogeymanism. Just around the corner, the government is preparing to shut down Bitcoin! Nothing like that has happened in the past 9+ years. In fact, the opposite has been true as most jurisdictions have been pretty accommodating, arguably even too lenient on the issuance and usage of cryptocurrencies, but that is a topic for a different post. [↩]
One reviewer mentioned: “Are the authors aware that CMOs first appeared in 1983, and that in many countries where they were heavily utilised including in the late 2000s they worked as advertised? In fact many CMOs in the US performed as modelled. The issue was, and is, always liquidity, over-leverage and most of all deteriorating lending standards. Cryptocurrencies will most likely be looked at as catalysts of these risks should their notional rise substantially, not their saviour.” [↩]
One reviewer commented: “Are they arguing that people would have been more able to pay their mortgages or that home values wouldn’t have fallen if CMOs were on a blockchain?” [↩]
One reviewer explained: “When someone claims that blockchain would have prevented the mortgage crisis, they are revealing their ignorance of their ignorance. I worked with some of that CMO data. One former colleague works for one of the large consulting firms ‘blockchain’ practices. He posted something about how blockchain would address the problems with mortgage servicing . When I privately asked him how it would do so,and that the problems with mortgage servicing that I was aware of were either failure to do certain required activities or their failure to record that they did them, as opposed to someone changing the record after it was entered, he did not respond.” [↩]
For example, at the time of this writing, Coinmarketcap tracks 1641 different types of coins and tokens. Many of these are likely ERC20 tokens and thus rely on Ethereum itself and are not independent blockchains. [↩]
Worth re-reading the recent DoJ indictment of GRU officers as the DoJ provides a reason for why Bitcoin was used versus other transmission methods. [↩]
Someone should create a website that tracks all of the gigantic bullish claims from Bitcoin promoters on how it will topple banks and destroy governments. There are at least more than 100 such public predictions each month. [↩]
But “be your own payment processor” isn’t a catchy phrase. [↩]
Some literature describes the proof-of-work process used in Bitcoin as a “scratch-off puzzle.” [↩]
One reviewer mentioned: “A model that I like to describe this with is how the main professional soccer leagues are selected in Europe and other regions. For example, France specifically has an annual selection of the “League 1” after the Coupe de French. Basically any team can enter, but practically there is minimal turnover because a team from a town of 5,000 people is unlikely to reasonably beat a team like Paris or Lyon which has multi-million euro budgets. There are few upsets, but these can generally be modeled by statistical chance.” [↩]
For example, Coin Center circulated a borderline defamatory note to ESMA with regards to Corda – even before the Corda introductory whitepaper was released – likely because its author was unfamiliar with how the platform actually worked. [↩]
It seems to be a euphemism and code word for “someone with money who should buy coins.” [↩]
Based on public information, over the past four years pretty much the only cryptocurrency-related companies that probably were profitable equity investments were: exchanges and handful of mining companies operating outside of the US (e.g., some service providers have also generated steady income including several law firms and conference organizers). [↩]
In both cases, consensus is achieved by the longest chain rule. [↩]
May not be a Freudian slip here, but keep in mind all blockchains have operators and maintainers. See “arewedecentralizedyet” for more. [↩]
It arguably could have been a self-fulfilling prophecy: investors outside of Cyprus hear news about the Cyprus bailout and bitcoin… thereby marketing bitcoin to new retail investors who then go out and buy bitcoins to try it out. [↩]
It is common to see Bitcoin promoters regularly demonize these companies who are trying to improve and automate infrastructure, vilified as a bourgeoisie activity that must be shunned. Worth revisiting to see if this changes over time. [↩]
Furthermore, in September 2014 I gave a presentation (video) (slides) that similarly tried to bucket different types of proposed coins as “commodities” and the like. And I know I wasn’t the first to try and do so. Recommend readers do a bit more digging on this topic if they’d like to see a more thorough origin story. [↩]
One reviewer mentioned: “The native tokens / coins / assets inside a ledger are “cryptocurrencies”, they are currency in the single sense that they the only form of compensation accepted by the miner / staker in a network. This cryptoasset business really only makes sense in the context of units which are not used to pay for the security of a blockchain.” [↩]
But that doesn’t necessarily excite speculators and coin holders. [↩]
There are few religious undertones here that could be removed in the next edition. [↩]
As mentioned above, The Economist wrote a nice short article on this behavior — the greater fool – last year. [↩]
The authors of this book are likely unintentionally promoting coin buying with a security-like mentality, the wording could be modified in the next edition. [↩]
One reviewer mentioned: “Unless the authors explain how ETH is worth precisely zero based on the same logic then their statement seems disingenuous. Not that I believe that is the case, but I am not the one stating that scarcity in the future is the reason for the value.” [↩]
Since the authors are making this claim, would they be willing to disclose or be transparent about their own coin holdings for the date when they published this book? [↩]
The most likely answer is: speculators bought these coins because they knew others would buy it too thus driving the price higher. [↩]
Or conversely, you are considered “one of us” if you promote the policies and antics of said coin promoters. [↩]
Note: it should be apparent at this stage that “Bitcoin developers” should be in quotes because it is certain key individuals — and centralized organizations such as “Core” — who have the power to sway decisions such as BIP approval. These are arguably administrators of financial market infrastructure. See also: In Code(rs) We Trust: Software Developers as Fiduciaries in Public Blockchains [↩]
It is these types of passages that make a reader scratch their head as to whether or not the lessons for why equity ownership — and the rights afforded to equity holders — evolved to where they have in developed countries. [↩]
This narrative needs to be buried but probably won’t. [↩]
This is a common refrain that needs to stop being repeated. [↩]
A few months before Cryptoassets was published, the SEC published a report that said they found The DAO to have all the hallmarks of a security but they never enforced any specific legal action on its creators. [↩]
On p. 63 they write: “For example, a fully functional decentralized insurance company, Airbnb, or Uber all hold great promise, and developer teams are working on similar use cases.” Why do these hold great promise? Because everyone else says that on stage? [↩]
One takeaway is that other speculators may buy your coins at a later date when the prices go up, so you should get in before they do. [↩]
One of the biggest flaws in Chapter 7 is that all of the pricing information for the coins are based on markets that are opaque and unregulated… some of whom may be considered bucket shops of yesteryear. Lack of transparency is one of the reasons why all of the Bitcoin-related ETFs have been (so far) axed by the SEC. See: Comments on the COIN ETF. [↩]
For example, later on p. 104 they write: “More surprisingly, the portfolio with bitcoin would have had lower volatility.” Because of the time period? We could probably find other things with the same or lower volatility. That seems like cherry picking. [↩]
Maybe they are both, but that still doesn’t mean that the coins, say that Placeholder Capital invested in, shouldn’t be classified as securities. [↩]
Also, these are all arguably poor stores of value because of their relatively high volatility. For instance, “number goes up” or rapid price increases is not the definition for a store of value. Claiming bitcoin is a good store of value because it sees swift increases in price appreciation as measured by actual money is a contortionist view which ignores the empirical reality of how money is used. [↩]
For example, later on p. 110 they write: “While many cryptoassets are priced by the dynamics of supply and demand in markets, similar to more traditional C/T assets, for some holder of bitcoin — like holder of gold bars — it is solely a store of value. Other investors use cryptoassets beyond bitcoin in a similar way, holding the asset in the hope that it appreciated over time.” Spoiler alert: everyone that owns internet coins hope they appreciate over time. [↩]
And there are specific projects — such as Bitcoin — in which one clique of developers waged an effective propaganda campaign against miners. For more on this, look into the actors and organizations behind the Segwit / Segwit2x / UASF online debates. [↩]
Not to rekindle the flames of the Bitcoin blocksize debate but in retrospect, several Blockstream employees and contractors were arguably more effective at swaying public opinion than Coinbase was, even though the latter generates significantly more revenue and has actual customers whereas the former is largely just a R&D dev shop. This discussion deserves its own post but neither company is very forthcoming about client or partnerbase… although Coinbase has published a bit more information over the years relative to Blockstream. [↩]
The book downplays illicit activity as if it is not a valid, reliable use case when it is. For instance, the GRU allegedly used bitcoin to finance some of its operations focused on the 2016 US elections and they did so to obfuscate their tracks. [↩]
A fundamental problem with this book is that it wants to have it both ways, with no clear goal posts for what a good or bad platform is and how to measure it. How can an investor know if a coin is any good? A table of attributes is recommended for the next edition. [↩]
Simply multiplying the amount of mined / pre-mined / pre-allocated coins by the market price to arrive at a “market cap” is a disservice to how market capitalization is actually determined. See Section 6. [↩]
As an aside, even though there is no law preventing consumers and merchants from using or accepting gold (or silver) as a means of payment in the US, basically no one does because they’d rather hold it with the expectation of future price appreciation. I am sure lots of angry trolls will point out that legal tender laws in the US do not currently include precious metals and neither are cryptocurrencies. Yet there are other economic reasons why people would rather hold onto an internet coin or a gold bar versus use it as money, and simply blaming legal tender laws is missing those. [↩]
And as mentioned in the section above, both Zelle and Swift (gpi) will likely make a lot of inroads in the same national and international areas that cryptocurrency advocates were touting… but without needing a coin. The struggle is real. [↩]
Note: both have since left those jobs. Bogart became a partner at Blockchain Capital (a venture fund focused on coins) and Luria joined D.A. Davidson [↩]
In the next edition if possible, try to include Placeholder’s research so we can have an idea of the firm’s internal thinking on these issues. [↩]
Note that selfish mining has some odd game theoretic properties which may not hold up in the real world. But if the selfish mining pool manages to stay a block ahead on average, they can reveal a longer chain whenever they see transactions they want to censor. It comes with the caveats that it’s not completely reliable in that they aren’t guaranteed to be a block ahead of the rest of the network 100% of the time (due to the inhomogenous Poisson process mentioned earlier). However, if they manage to effect a cohort of self-interested selfish miniers, they could… and that’s the equivalent of a “51% attack.” [↩]
A user can be defined as a person who controls their private keys without relying on a 3rd party intermediary. [↩]
Several analytics providers include: Chainalysis, Blockseer, Elliptic, Scorechain, and CipherTrace. [↩]
This is reminiscent of the BearWhale nonsense a few years ago. [↩]
Recall that historically, humanity went from only having to bearer assets up through the 19th century. And that for a variety of reasons these became registered and immobilized and then later dematerialized altogether. Cryptocurrencies recreates a financial order that had already existed. [↩]
Butterfly Labs began accepting pre-orders in the summer of 2012 but delivered them late in 2013… and got sued by the FTC. [↩]
Regarding ‘perfect competition,’ four years ago Jonathan Levin opined that: “Another simple thing about this is that it is unsurprising that the bitcoin network got into this mess as it is economically rational to join the biggest pool. Minimises variance and ceteris paribus reduce orphans increasing expected return per hash. The other point is that there is still hardware bottlenecks so designing the theoretically most robust system may fail due to market imperfections. Implicitly in many arguments I hear about mining people assume perfect competition. Do we need to remind people what are the necessary conditions for perfect competition? Perfect information, equal access to markets, zero transportation costs, many players ……. this is clearly not going to be a perfectly competitive decentralised market but it certainly should not favour inherently the big players.” See p. 114 of The Anatomy [↩]
Needs a larger sample size conducted in a public venue, and/or with the help of an experienced sampling organization. [↩]
This then leads to incentives to attack and hack exchanges, because they end up acting as deposit-taking institutions, aka banks. [↩]
There were probably 50% more hand-written notes or comments that I could have added that I skipped over. [↩]
The HODLing “digital gold” meme which was only passingly mentioned in this book ultimately degenerates into goldbugism but that’s a topic for a different post. HODLing arguably became a thing once the ideologues realized Bitcoin itself wasn’t a competitive payment system. An enormous amount of revisionism has taken place since 2014 regarding what Bitcoin was and is and should be. [↩]
One reviewer mentioned: “By hoarding then actively purchasing more coins to hoard, they might temporarily create an effect whereby each marginal contribution to Bitcoin through mining rewards in expanding the effective monetary base is partially neutralized. In addition to marketing campaigns, this can lead to higher USD values and may incentivize additional mining power, which in turn creates higher hashrate. However, you cannot make the same argument for gold because simply driving the price of gold up doesn’t make gold harder to find or more secure, and in fact we see the opposite.” [↩]
For instance, the supply of gold is actually elastic whereas many cryptocurrencies including Bitcoin have an inelastic money supply. Where in the whitepaper does it talk about a store of value? If that was the goal, surely it would’ve been mentioned in the whitepaper or the first few emails upon Bitcoin’s initial release. [↩]
SEC Director Hinman gave a public speech about three weeks ago which was subsequently affirmed by testimony from Chairman Clayton the following week. A key point in his speech for many was his observation that ether (ETH) was no longer a security because it had become sufficiently decentralized.
In Hinman’s speech, he also provided an ad hoc checklist for issuers to go through to make sure they do not
fall afoul of securities registration requirements.
But in doing so, the guidance does not really seem helpful
as it raises more questions than answers which are discussed below. Note: the discussion for how – if at all –
the rules are updated or changed is not within scope of this short article.
The first is technical and pedantic: ETH as we know it, is
actually the first major fork of Ethereum Classic (ETC). Recall that following The DAO attack (hack)
in June 2016, key participants in the Ethereum ecosystem coordinated a hard
fork which resulted in two separate chains – what we now call ETH and ETC – and
that ETC represents the original chain.
So transitively, does Hinman actually mean ETC is sufficiently decentralized too?
Or is it just a property for the fork, ETH?
While most people are aware that it was the exchanges —
specifically Poloniex — which decided to recognize
and associate specific chains with specific ticker symbols, exchanges were also
key – existentially critical —
during the first moments after The DAO attack (hack) was discovered.
How integral were they?
And how did they coordinate?
A public chat log from June 17, 2016 details the coordination between core Ethereum developers and exchanges. One such dialogue is the following:
The specific passage highlighted above shows how key
developers requested that exchange operators stop trading ETH and after some
discussion, the major exchange operators – such as Poloniex – temporarily
acceded to the request. It bears
mentioning that the chat above is missing some additional context around locked
DAO tokens (which were still locked up for several more weeks) and that this
provisional trading freeze was being done to protect all coin holders.
This wasn’t the first time that core developers attempted to
coordinate with exchanges after a mistake has occurred. In March 2013, a group of Bitcoin developers,
miners, and exchanges did something perhaps more glaring – coordinated
off-chain to stop a hard fork — in an IRC chat room during an unintended hard fork of Bitcoin.
If you follow the dialogue in either chat room, it is clear
that a relatively small set of participants has influence and especially in the
Bitcoin fork instance, arguably administrates
the chain and its governance
during these critical time periods.
And at least one question arises from this: is the ongoing
success of the system (and the token’s value) reliant on the ongoing efforts of
others? If not, why not.
Arguably the answer is yes for most of these public
blockchains. And based on the handy “arewedecentralizedyet” chart we
can see that similar – and probably more – types of centralization exists for
many other cryptocurrencies too.
But recall that Hinman’s speech seems to assume that ether
was a security at one point and then through some process that is still not
explained, is no longer a security. What
day did that transition take place? Was
it before or after the ICO in 2014?
Before or after the coordinated hard fork in July 2016 (as a solution to
The DAO attack (hack))? Before or after
<insert other milestones or forks>.
For those counting at home, following the July 2016 hard
fork, this means that the ETC chain was actually created twice and in both cases was the work of a small group of known
people, some of whom continue to maintain it.
After all, blockchains don’t automagically fix themselves.
Is it possible for a coin or token to become
un-decentralized? And if so, do the
maintainers get something like a 90 day grace period to make it
re-decentralized otherwise the coin is sent to some kind of securities
While we wait for more clarity and specific answers to these
questions, another potential issue is with HoweyCoins.
HoweyCoins is a parody ICO website published by the SEC on
May 16, 2018 – right smack in the middle of “Blockchain Week” in New York
City. The website is supposed to serve
as a lampoonish illustration to coin issuers of what not to do when fundraising
– and also serve as a warning to investors for what to look out for as red
flags such as early participation discounts. It’s definitely good fun – my friends and I
frequently refer to it in jest.
And while Hinman’s speech explicitly punted on how Ethereum
was initially funded, the Ethereum public
sale back in July 2014 also relied on discounts
(or bonuses) to early investors during its six week sale. The first two weeks, participants received
2,000 ETH per BTC, which linearly declined until the final epoch in which
investors received 1,337 ETH per BTC.
Were the designers of the HoweyCoins website aware of this discounting?
In looking at the actual HoweyCoins whitepaper
there is no technical meat that issuers or investors alike can count on for
guidance. That is to say: there is no
technical attributes describing the functionality of how HoweyCoin mechanically
works. In its 8 pages it describes a
couple use-cases but – like most ICO whitepapers – is very vague at how it will
accomplish or achieve them.
But ignoring the Ethereum initial fundraising period and its
and Conditions, a problem that is not resolved in either Hinman or
Clayton’s recent speech and testimony is that HoweyCoins, for all of its vague
promises of high yield returns is a strawman that does not really help provide
guidance as it relates to the facts and circumstances around how Ethereum – or
any crowdfunded coin – can become sufficiently
After all, what is to stop someone from spinning up their
own blockchain called “HoweyCoins,” raise ~$17 million through a public
sale and 12 months later, formally
launch the mainnet (as Ethereum did)?
The wording and justification for why Ethereum is not still
a security – that it somehow at some point became sufficiently decentralized – seems ripe for debate and will surely
be gamed by future coin and token sales.
Without explicit parameters, if Ethereum is sufficiently decentralized, then so to – at some point in the
future – could HoweyCoins. And then it’s
no longer a parody.
A third and final point is that while Hinman alluded to them,
even if these chains are finger quote “decentralized,” and thus not falling
strictly under the “common enterprise” in the literal sense (i.e.
where there is literally a single legal entity determining the success), how
does a “community enterprise” present less risk to investors? Tangentially, isn’t this the same type of
argument that mob bosses frequently used and as a result RICO Acts were created
to pierce through?
If we take the view that the spirit of the regulations
(1933, 1934, 1940 Acts) which led to the Howey ruling was actually to protect
small investors, does a non-singular, and quasi-independent, but systemically
important influential organization actually reduce risk?
If that is the view that they are taking, it would be
helpful to see how the commission has come to that conclusion.
After all, it is plain as day to see that most coin
foundations are heavily influential in the maintenance and success or failure of
a coin. And representatives from investment
groups like DCG are routinely making
statements which have the single effect of moving the price and/or
direction of coins such as ETC. Even if
these groups are superficially independent of the mining and operation, on the
day-to-day they are directly traceable to the volatility in the market and to a
large extent the success or failure of projects.
Most, if not all of the coin foundations, market and
advertise milestones which depend on the coordinated effort and work of developers
that are paid with investors’ money. Coin foundations typically register and own
trademarks and other IP so (theoretically) they could force exchanges to
associate a specific tickersymbol with a specific
chain. And often there is a hierarchy within
a coin foundation with respect to the “community” it manages and oversees: it
owns IP, controls investor funds, manages the verified social media accounts,
and empirically calls the shots.
Look no further than Nano
and dozens of other coin projects that have been hacked or “exit scammed”
because of how centralized the command and control structures typically are. Another instance just last week, EOS block
producers got on a conference call and paused
their network (and later proposed scrapping
their constitution). Ignoring their
billion ICO, is that series of actions sufficiently
To be fair, the SEC has an unenvious role to try and
regulate something (a network-based coin) in just one jurisdiction whereas
these coins are also trading and custodied in other jurisdictions. For instance, the FCA doesn’t currently
regulate tokens (e.g., that are not equity or debt instruments). And the Howey test is not applicable in the
Perhaps opening a public comment period to provide
suggestions could be helpful in conceptualizing objective measurements and quantifying decentralization
(assuming it is not an oxymoron).
Though, that inbox would likely just get spammed so maybe just start
opinions such as those of the BIS.
In closing, a hypothetical HoweyCoins and its benevolent
overseers and thought leaders at the HoweyCoins Foundation could mimic other sufficiently decentralized projects and
host an annual HoweyCon; simultaneously emceed by none other than Howie Mandel
and Howie Long. If and when this occurs,
is the only thing that HoweyCoins did “wrong” was promise to provide discounts
to early investors?
Maybe not, at least if they donate some of their proceeds to
coin lobbying groups to help explain to policy makers and regulators that HoweyCoins
is not a security, because it is sufficiently
decentralized (e.g., more than one Howie exists). Either way, I bet there will be some amazing
schwag at HoweyCon, really looking forward to it. There may also be an announcement about the
forthcoming HoweyCoins Classic fork.