[Note: below is chapter 3 to Great Chain of Numbers]

As innovative and groundbreaking as Bitcoin has been, it has several known technical limitations.1 Simultaneously, the current development team is hard at work on priorities revolving around improving the security of the protocol from vulnerabilities and exploits.2 This is not a criticism of their activities and actions, especially in light of the transaction malleability issue that caused frenetic activity within the ecosystem during the middle of February.3 Other developers in the community have tried to assume the mantle of responsibility for improving the functionality and capabilities of this space.  Some projects involve fusing exoskeleton systems built around the Bitcoin protocol; others create their own independent ledgers; still others have even created bridges between Bitcoin and other ledgers.

Below, I introduce eight projects that are currently developing a mechanism to design and transport smart contracts or smart contract functionality.4 For each, I attempted to interview the main developers.

Colored Coins

As noted above, one way to utilize a crypto blockchain to verify wares is through a process being developed called Colored Coins.5 In a nutshell, this endeavor allows users to “color” a token to represent a specific asset such as a car, home, boat, commodity, a share, a bond – virtually any type of asset (e.g., 0.5 BTC colored green to represent your home).  These tokens can then be exchanged, just like bitcoin tokens, by anyone anywhere.  This enables a decentralized, trustless form of asset management that uses a blockchain as both a ledger and transportation mechanism.

Alex Mizrahi, who is leading the development of the Chroma Wallet used by the Colored Coins project says that “it is going to be very easy for the asset management industry as a whole to use Colored Coins.6 For example, some of the first places we are going to have adoption will likely be real-estate and portfolio management.  In fact, for any type of asset management it’s going to be simple to issue his own color that represents his goods.  A portfolio manager can issue one color that represents a portfolio of stocks backed by the real holding and sell it globally.  If he is savvy and his products are good, his colors are going to have demand.  So transferring ownership is very easy, quick and safe — just like bitcoins.  In the real estate industry someone can issue their apartments using colored coins and have them float on the blockchain, or manage time-sharing based on color.”7

Meni Rosefeld, another member of the development team, described several of the advantages of using a secondary attribute (color) within the asset management industry. “The greatest advantage is the removal of barriers of entry.  Currently, new businesses wishing to raise capital use cumbersome and inefficient private deals; and those aspiring to be listed in order to allow for the market to valuate them with an efficient mechanism, can only do so with a great expenditure. With colored coins, anyone can easily raise funds in exchange for equity, removing barriers of entry, encouraging innovation and allowing society as a whole to better allocate its resources between ventures.”

One area of confusion within the Bitcoin community is the misplaced understanding – that centralized servers are needed to issue and track a secondary attribute (the “color”).  According to Rosefeld, this is incorrect. “No centralized servers are needed for tracking – this is done in the decentralized network of the host currency (such as Bitcoin).  There does need to be an entity issuing each particular colored coin – however, an entity raising funds for a generic purpose is not usually in the business of running an exchange. Without colored coins, they would have to resort to a large 3rd party exchange with all the usual problems of barrier of entry (for both issuers and exchanges) and vendor lock-in.  With colored coins, they can outsource the tracking and exchange to the efficient decentralized network.  The issuer is only involved when issuing or recalling the coins; investors can then trade the coins between themselves without involving any 3rd party, which has implications for privacy, efficiency, and the kind of advanced transactions one can do.”

I also spoke with Amos Meiri, head of dealing at eToro, another member of the development team for the Colored Coins project.8 I asked: would it be easier to simply conduct all trade privately at the centralized exchange where it will be more scalable and private.  In his view, “Centralized exchanges definitely have their advantages, but colored coins can be useful for following reasons.  First, users do not need to trust their bitcoins to a centralized exchange.  Companies cannot manipulate ownership records (to commit fraud, for example).  So basically, if somebody gives you an IOU, it isn’t a good idea to leave it with the person who issued it or to affiliated parties.  Another reason is that companies cannot control how its shares are being traded, thus it cannot block trade.   And lastly, there is no need to maintain servers or manage security due to its integration with the blockchain.”

While this is obviously easier said than done, as noted above, this idea of using cryptoledgers to manage smart property has inspired and motivated numerous other groups to put forth similar efforts.  For example, Counterparty was launched in January.9 Its mysterious, relatively anonymous development team has released similar open-source applications, documents, binaries and tools that allow users and entrepreneurs to build smart property functionality such as derivatives and dividends in a decentralized manner.  Also in January, reporter Jon Southurst discussed several other groups including Reality Keys that can utilize a crypto protocol to build a predictions market or a way to hedge against currency fluctuations.10

Mastercoin

At the beginning of January 2014 I spoke with Taariq Lewis, the founder and CEO of BitcoinBusiness, a Bitcoin advisory firm and he is also the Smart Property and Business Development Lead of the Mastercoin Project.11 Mastercoin is a crowdfunded, non-profit endeavor to create an open-source decentralized exchange protocol for Bitcoin.  As noted above, the Mastercoin project has received 4,700 bitcoins ($5 million at the time) in crowdfunding which has been used to pay for bounties, building tools and write documentation all of which is ultimately released on an open-source basis.1213

According to Lewis, “we are on the tip of the iceberg of the democratization of upper level finance and investment management.  One apt analogy is that the current system involves a highly siloed, highly centralized organization reminiscent to the music industry prior to P2P innovations.  We are now approaching the first wave of people being able to distribute financial products to each other on a peer-to-peer basis.  While this obviously has regulatory repercussions such as the SEC and CFTC oversight in the US, there is no “Wolf of Wall Street” in crypto.  In fact, projects like Colored Coin, Counterparty and Mastercoin will create applications that will decentralize stock and bond exchanges allowing individuals and entrepreneurs to build dividend products and distribute the assets without middlemen.”

I also spoke with Ron Gross, co-founder of Bitblu and executive director at the Mastercoin Foundation, who also pushes the open-source nature of the project. “With Mastercoin, we are all developing open source software and tools that eventually will enable anyone to build their own applications on the platform.  We are still hiring people for the core development team yet ultimately we want to move into a decentralized structure where we as team do not actually own anything or manually hire and fire but rather a Decentralized Autonomous Application (DAA) does.  In addition we have put together a series of external bounties, where we give away $100,000 each month to developers outside the organization either working on specific milestones or just doing general innovation around the ecosystem. Thus new programmers to this space could immediately be financially rewarded for looking through a list of bounties and submitting solutions to them, or for being creative and building around the infrastructure.”

Gross sees this ecosystem eventually mapping the real world in a digital space: as self-reinforcing entrepreneurial activity – continuously builds the ecosystem a new financial system will emerge that serves as a bridge between cryptoledgers and the existing world.  As part of this vision, a natural outgrowth encompasses decentralized applications, bonds, asset backed coins, commodities, real estate, betting and prediction markets that correspond to a smart property token will emerge.  One on-going project he highlighted in particular was an open-source omniwallet, which will eventually be capable of handling and tracking the cornucopia of altcoins, metacoins, and even colored coins.

Yet getting there will obviously involve hurdles.  According to Gross, “just getting the protocol developed and robust will be a rewarding challenge.  The infrastructure is not quite ready for large more complicated projects and is undergoing massive development yet Mastercoin and all the other protocols in the same space are still accessible due to the open-source nature.  Any developer, anyone can come – look at the spec, go into the debates, send in your pull requests, look at the code – and contribute immediately.  There is no need for a central brick-and-mortar building because if you contribute anything that is positive, you will get rewarded for it.  BitAngels is launching a fund soon that is going to invest in protocols, development of DAOs and other “2.0” initiatives through hackathons where the top winners will receive a $500,000 investment.14 And through these efforts we will build a better financial system, one that is decentralized and creates complete financial freedom. The impact of creating such tools is obviously a matter of speculation but even a fraction of the pie is going to be really large.”

I also spoke with David Johnston, managing director of BitAngels, the first angel investment network focused on digitial-currency startups, and a board member at the Mastercoin Foundation.15 In his view, “cryptocurrencies are more than a payment network, it is more than a new type currency or store of wealth.  It is a whole new platform and is a way for people to now make programmable money and that gives rise to smart contracts.  Now that this money is programmable I can put it into applications, I can create other digital tokens.  That’s what really gets me excited where anyone can build anything.  In the long-run we also plan to turn the entire project into a DApp, to maximize resources and improve efficiencies.”

A DApp is short for decentralized application.  The Mastercoin platform, like arguably every other one, is still a work in progress and has gone through several iterations based on community feedback.  It also faces market competition from several others in this space such as Open-Transactions, Invictus (formerly BitShares).  As a consequence, it looks like a promising area for Christensen-style innovation.

NXT

Launched in late November 2013, NXT is a new cryptoplatform written entirely from scratch in Java.16 The platform has the ability to natively track “colored coins” – tokens that represent a specific asset based on their “color” (e.g., using a fraction of NXT to represent a car or house).  It also includes a decentralized asset exchange, which means you can buy and sell assets without going through a 3^rd party.  For instance, one of the problems that impacts centralized exchanges and online stores today is that both your fiat and tokens are vulnerable to theft, hacking and other abuse.  In one notable instance, in December 2013, an online commerce site called Sheep Marketplace was hacked and 96,000 bitcoins were removed from its web-based wallet making it the largest known cryptoheist.17 This type of abuse is nearly impossible in a decentralized peer-to-peer exchange because there is no single centralized point of attack.18

In February 2014, I exchanged messages with “Uniqueorn,” contributor to the NXT development team.19  In his view, “the best way to compare NXT to the other cryptocurrencies is basically to not do it. NXT is not an altcoin at all.  While most of the cryptocoins being circulated are typically clones of the Bitcoin codebase with a few slight variations, very few of them bring anything new or substantial to cryptocurrency functionality.  On top of this is a built-in encrypted messaging system (like BitMessage) and anonymous payments (similar to Zerocoin) which adds an additional layer of privacy to protect confidential information and trade secrets.  Yet a lot of work still needs to be done both with our platform and the rest of the industry.  You cannot expect that your mother and father are going to sit down and understand this. For them it is supposed to be a tool to make their lives easier, not harder.”

Another key difference is that unlike Bitcoin and Litecoin which utilize proof-of-work mechanisms that scale in difficulty with network hashrate (i.e., additional hashrate added to a cryptoledger proportionally increases the block difficulty level); NXT instead utilizes something called ‘Forging,’ which is basically recirculation of NXT (Proof-of-Stake).20 “Uniqueorn” noted that, “proof-of-stake allows ‘miners’ to generate NXT without requiring the use of relatively large sums of electricity that other cryptocoin proof-of-work systems currently do.”  In other words, the barriers to entry are significantly lower as user does not need to utilize a top-of-the line ASIC machine which is discussed later in Chapter 7.  Therefore, a user can “forge” tokens on a smart phone, a solar powered Raspberry Pi, or a laptop computer.  In practice, an algorithm randomly picks one node to process all of the transactions and all other machines know this system is the sole transaction ‘forger’ – thus all other erroneous transactions can be discarded.  All machines participating in this ‘forging’ effort are rewarded according to the proportional amount of NXT they have; thus if you have 1% of the tokens you have a 1% chance of being selected to forge the next block.  Because the transactions nodes are known, this provides increased security, an estimated 90% of the NXT tokens must be controlled by one agent in order to compromise the network via a double-spend (e.g., 51% attack).21

I also corresponded with ‘Graviton’ who is the Nextcoin.org community founder.22 According to him, one of the motivations for why the core team decided to move beyond Bitcoin was, “there certainly seemed to be demand for a technically advanced cryptocurrency with a completely new codebase that puts away the requirement for energy expensive Proof-of-Work once and for all.  The environmentally green and attack resistant Proof-of-Stake algorithm, plus the important fact that NXT is not only a payment instrument but a new generation platform natively supporting a suite of services such as decentralized trading and encrypted messaging, seems to have filled gaps that were shining open wide with the existing old school cryptocurrencies.”

He is also looking forward to the deployment of a decentralized asset exchange as well as colored coin functionality on the NXT platform and believes that these will “become a popular standard for quite a bit of trading applications, for both – cryptocurrencies and assets denominated in them.  The rest of the industry will integrate seamlessly to that, so the distinctions between various crypto brands will start to dissipate.”  And like several other developers interviewed, “the killer app would be to have available the simplest possible means to pay for merchandise & services in fiat nomination but from one’s cryptocurrency wallet, to be able enjoy the fiat price appreciation with the same wallet, and to flip your wallet contents to another crypto with a push of a button.  Preferably on mobile.”

Ethereum

Another “2.0” project that is gaining traction is Ethereum, announced in January 2014 which brings together both a cryptoledger and a Turing-complete programming language.  In short, a Turing-complete programming language means that the language can be used to simulate any other computer language (not just its own).  The original Bitcoin protocol and software implementation released in 2009 included a language called Script that had many limitations (it was intentionally not Turing-complete) and as a consequence has largely been underutilized.  As a consequence, developers have had to try and use these duct-taped exoskeleton wrappers to build on top of the protocol to enable new functionality.  Many developers, including those with the Ethereum project, recognized this limitation and, rather than building and providing a specific feature set, will instead use a Turing-complete C-like language (CLL) that software developers can then use to build a cornucopia of tools, including any type of smart contract, asset management instrument or even a decentralized autonomous organization (DAO) that can then be automatically executed, controlled, and audited by the Ethereum ledger.23 While its approach is one of the most holistic thus far, its long-term success still requires a critical mass, mind-share and the network effect.

To find out more about Ethereum, I corresponded with Vitalik Buterin, head writer at Bitcoin Magazine and a lead developer on the Ethereum project.24 Because of the all-encompassing abilities “2.0” projects are slated to have, it could be confusing for developers to determine on which platform to initially build their apps, but that may not be the only hurdle.  In his view, “I would say the main challenge in the 2.0 space is going to be (1) building contracts, and (2) building interfaces. These have always been problems, of course, but up until now they have been eclipsed by other, larger, problems, like maintaining server infrastructure and scalability, ensuring security of funds, regulatory compliance and having banking relationships. With decentralized apps, most of those problems are gone, so the only two issues that still remain – contract design and interface design – are now at the forefront. The two problems can easily be handled separately; someone should be able to write a derivatives trading GUI and have that port over automatically to various systems inside of Ethereum, BitsharesX and whatever else people want to trade on.”25

Several other developers and investors I spoke with had similar sentiments: creating easy-to-use, intuitive interfaces for end-users would quickly set your product apart from the pack.  While there have been many advances, especially for merchant plugins, backing up and securing wallets can be quite cumbersome and even a chore to handle at times, stunting wider-spread adoption.26

Buterin had previously worked on both the Colored Coins and Mastercoin project.  While portable, both of these currently utilize the Bitcoin protocol, which has a couple of limitations.  In Buterin’s view, “one of the key features of Bitcoin is that it has a concept of “simplified payment verification” (SPV), where a Bitcoin node can verify the validity of a transaction in the blockchain by only downloading the very small subset of data in the blockchain that is relevant to that particular transaction.  Given that a “full” Bitcoin node now takes 14 GB of space to run, beyond the reach of many users, this mechanism has become an essential part of Bitcoin security. The problem with on-blockchain meta-protocols, however, is that they do not benefit from this protocol. The underlying Bitcoin layer has no way of knowing whether or not a given transaction is valid in the context of the meta-protocol, so the Bitcoin blockchain will include transactions that are both valid and invalid, and so the validity of a given meta-protocol transaction can only be calculated by recalculating the entire state of the protocol up until that point – requiring the full blockchain.  Ethereum solves these issues by not being a meta-protocol, instead relying on an independent blockchain.”

SPV is a type of thin client that provides Bitcoin users a lightweight method for sending and confirming transactions without having to carry around the entire database.27 It does this by downloading only the headers for all the blocks (i.e., the Merkle tree) and not the entire blockchain itself.  As a consequence, this flexibility enables Bitcoin clients to be used by point-of-sale registers that may not have enough space or bandwidth to continuously download the entire blockchain.  And at this time, as Buterin notes, the only way to completely confirm that a transaction based on Colored Coins or meta-coins like Mastercoin is valid is to re-check the entire blockchain.  This presents a significant obstacle to scalability.

When describing and defining what a “smart contract” and “DAO” are, it can be confusing at times because a robust smart contract is sometimes used synonymously with a DAO.  According to Buterin, “I would say there is no clear-line distinction between the two, but there are some general differences in connotation. To me, a smart contract is something that is single-purpose and ephemeral, so they are created for a specific task and can disappear at the end. A financial contract is a good example there. An autonomous agent is something that is more long-term focused, and includes an internal AI to make decisions. And finally, a decentralized autonomous organization is a long-term contract between many people, perhaps even with the ability for people to join in as signatories or trade their positions away, whose main role is to hold on to assets and use some kind of voting system to manage their distribution. There can be many different types of DAOs; the more basic ones live entirely on the blockchain, but more advanced ones might have some of their data stored on other decentralized networks or across a number of servers.”28

Throughout this manuscript, several of Mike Hearn’s presentations are referenced, including the Turing 2013 conference.29 While both Hearn and Vitalik Buterin use the same name, DAO, the definitions for what the term implies, varies.  In an email exchange, according to Hearn, “what Vitalik calls DAO’s are not quite the same as what I discussed in the Turing talk. I used to think they were the same, but on closer inspection he called Bitcoin itself a DAO so it’s obviously different.  Assuming you mean agents, there are so many challenges I doubt it will happen any time soon. Really you need trusted computing for it to work well and that won’t work well at least until Intel release CPU’s supporting their SGX extensions, which they didn’t even announce a date for.”

Trusted computing is a term for computers that can be controlled a certain way via encryption.  Many governmental agencies such as the US Department of Defense require that computers acquired by vendors have such functionality.  In September 2013, Intel released its programming reference manual for Software Guard Extensions (SGX) which could potentially create similar functionality in consumer-based systems.30

Throughout this guide I describe simple “smart contracts” with the assumption they do not have any sophisticated internal AI components.  Similarly I refer to relatively simple DAOs that wholly reside on the blockchain.  As programmers become more acquainted with decentralized software and the technology evolves and begins to be used in practical applications, it is likely the specific meaning of each term will be subject to change.

BitShares

This last point is viewed as a critical issue to other 2.0 project managers as well.  I had an email exchange with Daniel Larimer, the creator of BitShares, and the first person to describe Bitcoin as a Decentralized Autonomous Company (DAC).31 BitShares is a new way to view cryptocurrencies where you view your wallet balance as shares rather than coins.   According to Larimer, Bitcoin can be viewed as a DAC where each bitcoin represents one share in the Bitcoin ecosystem.  The transaction fees that Bitcoin charges can be viewed as revenue to Bitcoin and the mining rewards can be viewed as expenses paid by Bitcoin to secure the network.

Larimer decided to change the analogy from coins to shares so that the underlying economics could be considered when designing next generation crypto systems.   Based on this analogy, he sees several ways to improve Bitcoin when viewed as a company.  In his view, the driving principle is that all companies should generate profits by minimizing expenses while maximizing revenue from product sales.

In the case Bitcoin, the primary expense is security which is provided by an expensive proof-of-work (PoW) process described in chapter 2.  In BitShares systems all security is provided by proof-of-stake (PoS).  In his view, a PoS (which is also used in NXT) can be thought of as having the shareholders vote on the valid transaction ledger.  In this way those who own the system secure the system without having to spend increasingly larger sums of capital to do more work than any attacker can.  This last point was recently described by Nicolas Houy, a researcher at CNRS, stating, “Bitcoin miners have engaged in an arm race to computational power and in the end, much hardware, engineering and energy are used to solve mathematical problems that are artificially made extremely complex.”32 A PoS system is supposed to remove this artificially complexity and lower the capital costs for entry.

The other thing BitShares systems do, according to Larimer, is focus on increasing the value of the transactions that can be performed and thereby generating additional transaction fees.  Because there are no miners to pay, transaction fees can be viewed as profits for the system and these profits are used to buy back and retire shares.  This has the effect of increasing the value of the shares still in circulation.  It is economically similar to earning a dividend.  The value from the fees is transferred to the shareholders proportional to their stake.

The first BitShares system being developed by his team is called BitShares X which continues with the company analogy to implement the business model of a bank and exchange simply be defining a new set of transactions supported by the blockchain.  According to Larimer, one unique attribute about BitShares X is that there are no counter-parties, employees, vaults, or contracts and yet according to him, BitShares X facilitates the creation of BitUSD purportedly the same way that the Federal Reserve creates FedUSD: it lends it into existence backed by collateral.

BitShares X uses shares in the system as collateral to back BitUSD.  BitUSD can be thought of as an asset that you can sell for a dollar’s worth of shares in BitShares X.  Depending upon when you buy or sell your BitUSD you will get a different number of shares, but based on their initial model the purchasing power should be approximately a dollar.  And according to him, like Bitcoin where there are no issuers backing the value of a bitcoin, there are no issuers of BitShares X shares or BitUSD.  The entire system operates on nothing but a chain of numbers following a predefined set of rules enforced by the consensus of the network.

Larimer also believes that BitShares X is just one of many potential business models that could be defined entirely in software.  And while one of the challenges is finding developers with an understanding of both economics and consensus, yet other business models his team sees opportunities in include insurance, domain names, gaming, auctions, and voting.  Voting is another issue that other entrepreneurs in this space touched on, which is described in greater detail in the NGO segment in chapter 8.

I also spoke with Charles Evans, economic advisor with the Invictus-run BitShares project.  The way he looks at BitShares is that

“a share can be issued for agricultural commodities, like coffee, tea, cardamom, etc. If someone who grows a commodity that has a corresponding BitShare sees that the BitShare can be sold for more than it would cost to deliver the commodity, then the grower can offer, e.g., 100 kg of cardamom in exchange for 100 kg of BitCardamom, sell the BitCardamom on the open market, and ship the cardamom to the buyer.  Note that the BitCardamom is not “backed” by cardamom. It trades on a prediction market, in which players worldwide try to discover a single, global price for a fungible commodity. When someone with specialized local knowledge sees an arbitrage opportunity—here, simultaneously buying BitCardamom with a promise to deliver cardamom and selling the BitCardamom on the open market—that party can exploit the opportunity.  Instead of negotiating with local wholesalers, who might have information advantages over local growers, and relying on one’s own ability to haggle well, the grower can use a global information market as a guide.  Likewise, if the price of BitCardamom rose over time, prospective growers worldwide would be able to see the price and respond to the price signal.”

Counterparty

For perspective I had an email exchange with Ryan Orr, who is a professor at Stanford University (teaching Global Project Finance and Infrastructure Investment) and chairman at Zanbato.33 Orr has also been closely following Counterparty, which is the first functioning protocol layer fully integrated with the Bitcoin blockchain that supports peer-to-peer transfers of a coin called XCP.34 At the beginning of January the Counterparty development team announced that they had successfully released a working protocol including asset-backed issuance, betting, dividends, callable assets and the world’s first decentralized exchange.35

As the next few months will involve a race between Colored Coins, Mastercoin, and Counterparty as well as other non-blockchain equivalents such as Ripple and Open-Transactions, with each system bringing its new innovations, many outside commentators have expressed interest over Counterparty’s integration with the Bitcoin blockchain and execution to date.  “The fact that we have six serious competitors is a huge development for the entire segment,” says Orr, “The early days of this race will be about tech execution whereas the later days will involve regulatory finesse. The ‘value web’ (as opposed to the ‘information web’) is finally here.  The significance of these developments for the future of the field finance are gargantuan – what we are witnessing could be the equivalent of the invention of http on top of TCP/IP, and these are the protocols that are likely underpin the evolution of the value-web over the coming decades.”

In February 2014 I exchanged messages with one of the lead developers, who used the pseudonym “PhantomPhreak.”36 According to him, “Counterparty is a protocol, and a piece of software, that takes the technology underlying Bitcoin and extends it beyond simple payments, implementing a wide range of financial instruments.  It may be used to trade cryptocurrencies, create assets, make bets, and more, with all other Counterparty users, safely and anonymously, with no middleman at all.  It is built on top of the Bitcoin blockchain, so it can be very simple and reliable. It is being developed very quickly, and it has a large feature set already. Counterparty inherits all of Bitcoin’s security and reliability.  It is open-source, and its launch was entirely decentralised, as is the protocol itself.  And as its name suggests, implements a completely distributed, automatic and deterministic clearing house, so there is no counterparty risk to speak of in most transactions. Of course, if someone were to issue an IOU using Counterparty that he did not make good on, then the anonymous nature of the protocol would leave the slighted party with little legal recourse.”

This last sentence is of particular interest as it still shows a problem that is currently not solved in a decentralized manner, as Preston Byrne identifies in Chapter 2.  As this space matures, developers will need to learn how to structure smart contracts so they are legally and commercially useful.  How to enforce these clauses without an escrow-based DAO, without an independent mediator or without a reputation system (e.g., credit score) can and will be tricky but could be a business opportunity for experienced professionals in those segments who are looking to get exposure to the cryptocurrency sector.  One competing developer explained to me that, “Counterparty is way ahead of the game because their distributed financial system is deployed today.  In many ways, the team is reminiscent of Satoshi: they are people in our community who saw a problem with prior attempts and are fixing it.  All others are still spinning their wheels and really need to deliver functionality on which we can all explore further.  What’s more, proof-of-burn is a big commitment and raises the stakes for everyone. That’s why there’s so much development activity going on with Counterparty. The investors have to pull to make the coin work and they’re pulling hard.   They released alpha software and folks are losing money, but they’re shipping code updates daily which means the software is getting better and the markets more active. This is an exciting space and this level of competition motivates all of us to take it up a notch.”

Proof-of-burn (POB) is a unique turn on allocating “scarce resources” (tokens).  Whereas cryptocurrencies such as Bitcoin, Litecoin and Dogecoin use proof-of-work to allocate resources (e.g., a token), proof-of-burn requires that the miners (or any user actually) send their tokens such as a bitcoin to a provably unspendable address (a terminator address) where they are untouchable forever by any party.37 The first and only “burn” took place beginning on January 2, 2014 and lasted for thirty days – now all of the XCP that will ever exist have been created.  During that time, 2,130 BTC were effectively destroyed amounting to roughly $2 million in market prices (the actual repercussion was that all other holders of bitcoin saw a net gain in value by roughly 0.01%).38 Counterparty then automatically converted the “burned” token into its own unit, called an XCP resulting in no premine or foundershares.  It currently takes five XCP to create your own asset, the five are destroyed in the process as a spam control function.  While it is a controversial method, proof-of-burn does remove the human element from the equation.  That is to say, while other ‘”2.0” projects are typically funded by IPOs whose assets are then (usually) managed by a non-profit organization, because there is still a trusted 3^rd party involved, abuse can occur.  That is not to suggest that any abuse is happening, but rather that Counterparty is re-solving the Byzantine General’s problem in a different yet mathematically similar, manner than what Satoshi did in 2008.39

“PhantomPhreak” also sees potential in other decentralized platforms, “I think that there’s a very good chance that so-called second-generation cryptocurrencies will “take off” in the next year or so. Bitcoin was a revolution, in a number of ways, and now it’s time for an evolution of the core concepts and paradigms that it introduced.  Computer science has to catch up with it, so to speak.  A secure, distributed blockchain can be used for so much more than simple payments: advanced financial instruments (a la Counterparty), messaging protocols (c.f. Bitmessage, Twister), etc.  Certainly the future of finance is more decentralised than the present, and the economy as a whole will have to change accordingly.”40

Bitmessage is a peer-to-peer protocol that allows users to send encrypted messages to anyone in a decentralized trustless manner (i.e., Bitcoin for messaging).41 Twister is an encrypted decentralized peer-to-peer microblogging application that uses both the Bitcoin and BitTorrent protocols to enable users to tweet and communicate anonymously.42 Other projects in this space are Bitcloud (decentralized cloud services), Maidsafe (decentralized dropbox and API platform) and SyncNet (decentralized web browser).43

In addition, he believes there are many applications that financial instrument designers could contribute to this space and in particular Counterparty, stating: “the most obvious possible contributions are simply new features.  Right now, for instance, Counterparty only has two different types of ‘bets’, namely simple ‘Equal/NotEqual’ bets and contracts for difference.  Counterparty, however, has the potential to implement very nearly the entire range of tools commonly available to professionals in the financial industry.  Of course, pretty much any developer could contribute a lot to the Counterparty project, which still has a relatively small codebase and an underdeveloped software ecosystem, simply by writing user-friendly interfaces, or algorithmic trading engines, for example, on top of the reference client.”

I also exchanged messages with “cityglut” who is another member of the development team.  In terms of business opportunities, it is his view that, “what cryptocurrencies in general and Counterparty in particular allow for that is arguably most significant is further decentralization.  I believe that businesses which capitalize on this aspect of Counterparty will have opportunities they have not had until now.”  As noted above, this project does have code that is shipped and is currently being used by the community at large.

He also sees that there are a number of areas of low-hanging fruit.  According to him, “in my mind the most obvious financial instrument that Counterparty is currently lacking is a real options function. Counterparty allows for binary (Equal/NotEqual) bets and the creation and (distributed) sale of assets, and I believe that a combination of these functions could create a full-blown options function, but it may well be that in Counterparty’s current implementation this is infeasible.  Even if an options function can’t be built from Counterparty’s extant functions, it seems to me both possible and desirable to implement options in Counterparty in some way.”

Yet there are challenges too, “It is precisely Counterparty’s brand new functionality that entails greater necessary due diligence on the part of users. Since anyone can make an asset, and anyone can publish a broadcast upon which to bet, users must do what they can to make sure the asset they are purchasing is legitimate, and that the broadcast upon which they are betting has not been “tampered” with.  In an effort to facilitate the former, we have recently implemented a description space for every asset: issuers of assets can include up to 42 bytes (in UTF-8) with each issuance, describing the asset being issued. Regarding broadcasts, aside from the financial incentive feed-operators have to stay honest (namely, collecting betting fees), we imagine that an – albeit informal – reputation system will naturally evolve, helping users to decided which addresses’ broadcasts to bet on and which to avoid.”

This secondary attribute, a type of descriptive space is a feature that many of the other platforms are trying to enable in order to organize and manage different types of assets.  The issue involving reputation is also a theme repeated by many other investors, developers and experts and one that a DAO escrow could potentially resolve.

Open-Transactions

Chris Odom is a cofounder and CTO of Monetas and the lead developer of Open-Transactions (OT).44 Open-Transactions is an open-sourced digital software suite that utilizes current technology to enable trustless financial cryptographic interactions through privacy features such as blind signatures.  It is also portable and ledger agnostic allowing developers to bridge its applications to other cryptoledgers.

Many outside investors and businesses frequently ask Odom a theme on the same question, what business solutions can be developed for this segment?  Yet according to Odom, “asking what profitable business opportunities there are for crypto currency is the same as asking that question for the Internet in general. It is extremely broad in scope.  I think we are talking about a transformative invention, comparable to electricity, computers or the Internet.  It’s going to create all new spaces, and it’s also going to transform all existing sectors.  While Open-Transactions currently is integrated with Bitcoin, it is ledger agnostic because it is a financial crypto library, similar to how OpenSSL is a communications crypto library.  In terms of immediate opportunities, we have some bounties posted on CIYAM.org/open.  However, people should definitely be aware of risks.  Cryptocurrency can be used in legal and illegal ways, so it’s not the currency itself, so much as how you use it.  You just have to watch out for regulatory compliance issues, and if those get too onerous, you have to look at moving your company to another country.  Some countries are less free than others.  For an investor I might also point out some of the unique propositions of OT, one being its ability to operate in a low-trust way, that it is federated.  And that it’s also able to fill the gap and do the things that all the other servers do in the Bitcoin world like the MtGox server, or the BitStamp server, or any of these Bitcoin services that use a server.  Any of them could be replaced in a lower-trust fashion, using OT at least, using OT as the financial engine, not necessarily the web GUI pieces.”

One advantage that Open-Transactions has over conventional blockchains which have algorithmic delays, is that because it uses known servers, Bob can trade near instantaneously.  Whereas confirmation of bitcoins, bitshares and other blockchain based instruments are measured in minutes, users can only execute trades in those intervals as well.  And if you can put OT on a distributed database, in theory that means you can have cryptocurrencies that confirm instantly without centralized control as well.

Ripple

Ripple, commercialized by Ripple Labs, is a payments protocol that acts as a payment platform, decentralized currency exchange, and smart contract network that can be used with any digital currency, including Bitcoin.  Ripple provides a solution for implementing an asset cloud via “trusted” gateways.4546 At scale, Ripple or Ripple like systems provide instant liquidity and exchange between counter parties, where there can be a trustless exchange between 3^rd parties, and those 3^rd parties can decide where their exchanged assets will settle within the network, such as any gateway who provides redemption for the represented asset.  In addition, unlike other payment platforms that use variants of proof-of-work, it uses a consensus ledger which is distributed to a global network of servers.47 These servers continually receive transactions and proposals from other servers on the network and these are compiled into a “Unique Node List” (UNL).  Proposals from servers not in the network are discarded while those remaining are vetted and algorithmically “voted” on by the servers.  Once a consensus (defined at 80% agreement on what transactions are legitimate) is reached, the server validates the proposals and closes the ledger, creating a “last closed ledger” (similar to a block).  The process then repeats itself.  This process takes roughly five to fifteen seconds allowing quicker transactions than nearly any proof-of-work system today.  Altogether its network processes roughly $20 million each month from approximately 68,000 user accounts.48

Beginning last year, Ripple Labs created an initial money supply of 100 billion XRP which was predetermined to be enough to last for hundreds of years.49 The designers of Ripple realized they had a problem if someone wanted to flood the network with useless transactions, which is the currency equivalent to spam.   To this the network charges a transaction fee which permanently deletes 3 “drops” of XRP.  Each drop is equivalent to the smallest possible amount of XRP, thus .000003 per network transaction cost.  A drop is equivalent to the Bitcoin “satoshi” the smallest possible unit of BTC which is .00000001.  As of this writing more than 3,500 XRP have been permanently removed from the network.

According to Jon Holmquist, an early Bitcoin adopter and Community Liaison at Ripple Labs, “beginning 15 years ago a merchant could create a webshop in 10 minutes and attract visitors from around the globe.  Yet, they could not easily pay for the merchandise until 5 years ago.  With the development of cryptocurrencies such as Bitcoin, consumers can now use money without borders.  However one of the biggest issues today is obtaining bitcoins especially when you reside in an economically depressed region.  Ripple lets you exchange and obtain whatever currencies you want to use.  As Ripple continues to build partnerships, the network creates a self-reinforcing positive feedback loop that takes care of itself.  As a consequence, because Bitcoin has gotten a lot of push behind it, it is possible to have a fiat exchange in every country which then allows customers to finally purchase from any country, with their own currency.”

A math-based currency is a term often used by members of the Ripple Labs team to describe the concept of “programmable money” – that is to say, virtual tokens that are mathematically constrained by algorithms and difficult if not impossible to forge.5051 The Ripple payment system works alongside Bitcoin by enabling users to use XRP, a token, to represent certain financial instruments (like currencies) which can then be instantly transferred globally and exchanged for Bitcoin and then a fiat currency.

Both XRP and the Ripple protocol can be leveraged in other ways as well.  Steve Bennet, a finance professor at San Jose State University and an angel investor with CrossCoin Ventures which is a new business incubator partnered with Ripple Labs, points out that the project has “built a new incubator which will later become an accelerator focused on building out the Ripple ecosystem.52 Currently we are focused on attracting Bitcoin-related companies which can leverage the Ripple platform to provide new value to customers globally.”  Bennet’s team (including Ryan Orr mentioned above) plans to work with both new startups and existing companies, provide them access to Ripple’s management and even exchange Ripple for a percent of equity much like several other “2.0” projects have done (e.g., Mastercoin, NXT).  His vision is to leverage the incubators’ resources (e.g. networking, mentoring, legal) and help the incubated teams focus their energy on providing value-added services to a broad array of consumers who are unfamiliar with cryptocurrencies.

I also spoke with Stefan Thomas, co-founder of WeUseCoins, creator of bitcoinJS and CTO of Ripple Labs.  In his mind, “the easiest way to describe Ripple right now is that it is a FOREX platform that removes most intermediaries and does so in a matter of seconds.53 And because you reduce the total amount of fees that are charged from the remittance process, this provides thinner spreads for users who traditionally have had to worry about currency fluctuations.  That is to say, in the past, it could take hours or days for funds to move across borders whereupon the value of the currency could decrease.”  Yet as noted above, Ripple’s platform is nearly instantaneous and powered by a distributed network of ledger “processors” (slightly akin to “miners” but requiring very little infrastructure) and gateways.

And according to Thomas, “while processor nodes do vote and verify the ledger integrity to prevent forgery and double-spending like a blockchain, these processors are unlike the “miners” used in blockchains because of the way the light-weight method consensus is determined which requires substantially less infrastructure (e.g., no need for ASICs or GPUs).  Consensus of the ledger is done through a peering method; similar to how peering with trusted nodes works with internet providers.  The ledger itself is a bundle of digitally signed data transactions which is sent through the network and voted upon by client peers (nodes).  These nodes poll one another to see which transactions came first, the ones that are determined to be false or illegitimate are discarded and all others are included in a verified ledger state that is then considered closed.  This entire process takes between every 5 to 15 seconds and nodes that become unreliable with spam are then ignored by peers.  The reason the timing is not a fixed rate because transaction bundles not only vary by size (e.g., consumption by consumers does not happen at a unison, flat rate) – and it also illustrates how data itself is processed globally through current public infrastructure.  In contrast, the reason that Visa is slightly faster is that they use private centralized nodes which requires significantly more overhead and capital expenditures.”

“Gateways are the actual organizations that move assets in and out of the Ripple network.  They can range from single-individuals to large banks.  Users establish trust lines with gateways which can be located in any part of the world, providing liquidity into nearly any local currency.  A unique feature about gateways is, that while they may be a single-point-of-failure in the traditional sense, users can still route around censored nodes.  Furthermore, gateways cannot appropriate the assets of one specific user: either they default for everyone or none at all.  So for example, Bob can create a debt line with Alice who is trusted party, a gateway.  Gateways live by reputation, so they have an incentive to fulfill their obligations.  Bob can then exchange a local currency with Alice for IOUs (XRP) with which Bob can then send to any other gateway and convert XRP into the local currency.  This can be done in a matter of seconds, which is significantly faster than any blockchain-based system, yet is actually more secure (51% versus 80%).”

Ripple Labs refer to their technology as a ‘value web’, an ‘Internet for money’, and ‘http for money.’ Existing financial institutions could serve as gateways today by establishing ‘trust lines’.  The gateway system enables financial institutions to exchange value in the form of digitized assets (e.g., commodities, fiat currency).   For instance, Bob’s Bank of Buffalo could set up a gateway and trustline with Alice’s Agriculture Bank in China.  Bob could provide USD liquidity to Alice and Alice could provide RMB liquidity to him in a cheaper and quicker manner (between 5-15 seconds) than existing wire services which could take days and charge relatively high fees.  Ripple acts as trusted ledger for all participants, yet cash balances must be settled outside of the Ripple protocol.  XRP is the only currency native to the network.

Thomas continued with, “another competitive advantage that the Ripple protocol has over others in this space is that our code uses the smallest amount of trusted code base, basic OP codes which provide the most secure assembly code to which to build from (e.g., interacts directly with the iron, with the metal of the semiconductors).  Thus the native software client is less vulnerable to exploits that occur from building above with other higher-language layers.  And over the past two years we have open-sourced a significant amount of codebase including the protocol to the public.  This in turn has led to further refinements and security fixes.  In addition, we are continually looking at ways to expand the protocol’s use, making the ledger essentially a database that will allow for the transaction of smart contracts.54 And because this network is slightly more efficient than most other platforms, this allows for new innovations to take place down the road.”  This contract-based system will be Turing complete and include two-stages, the first of which is non-deterministic which enables contracts to interact with real-world protocols such as DNS and HTTP and also allows users to include language interpreters and reference libraries.

“This space is rapidly evolving; for instance, the original Bitcoin client was much more cumbersome than it is today.  For beginners it used to take 24 hours to download the blockchain and confirm transactions.  Now there are numerous projects each of which trying to provide value-added services and this competition is pushing us to look at new ways to innovate, such as peer-assisted key derivation function (PAKDF) – a mathematical way of utilizing blind signatures.55 One of the user-adoption problems in this space is that it is hard to memorize long secure passwords and frustrating for new users to learn how to securely save passwords on disk drives.  In contrast, PAKDF will allow Alice to use relatively weak passwords that can be sent to Bob who will sign something (e.g., a contract) without knowing and therefore unable to break Alice’s password.  This is called a blind signature which adopts a form of homomorphic encryption and we are integrating into Ripple.”56

Whereas a user would need to memorize a long passphrase, this specific application of securely signing a password could lead to ease-of-use for end-users.  In a nutshell, a blind signature scheme “allows a person to get a message by another party without revealing any information about the message to the other party.”57 The analogy typically used to describe how this worked is, Alice places a message inside a carbon lined envelope.  This envelope is sent to Bob, who cannot read or see any of the information, but can sign on the outside of the envelope, which imprints the signature on the carbon inside the envelope.

Current Cryptoprotocol Infrastructure

This Euler diagram shows two main systems, those currently part of a cryptoledger and those that are not, which in this case is solely Open-Transactions (OT).58 As noted earlier in the chapter, OT works by connecting its OTX protocol to other services (much like SSL does with other databases) such as Bitcoin and is therefore ledger agnostic.

Within the cryptoledger diagram are essentially two other distinctions, those that use a blockchain and those that use a consensus ledger.  At the time of this writing only the Ripple protocol uses a consensus ledger.   When it was first created, Namecoin was also originally its own independent blockchain but the mining process has since merged with the Bitcoin ledger.  The other independent blockchains above are Litecoin, Dogecoin, NXT, BitShares and Ethereum.   At the time of this writing, the Ethereum team has not settled on which system it will use – it may use a hybrid approach similar to what Peercoin has done (proof-of-work and proof-of stake).

Proof-of-work (PoW) involves a network of mining machines as originally employed by Bitcoin in 2009.  Computers are given a series of increasingly difficult benign math problems which they complete as a way to stave off rogue attackers.  In this example above, Litecoin, Dogecoin, Namecoin, Bitcoin and potentially Ethereum use a proof-of-work method.

Proof-of-stake (PoS) is different in that the transaction node for a block is randomly assigned and all network participants communicate directly with it.  One advantage to this approach as it reduces the amount of hashing power needed to secure the network.  At the time of this writing, only NXT in the above diagram uses a pure PoS method; Peercoin uses a hybrid and Ethereum may also use a hybrid as well.

Proof-of-burn (PoB) is a unique method that has only been used thus far with Counterparty; a user sends a token (a bitcoin) to a provably unspendable address (a terminator address).  The largest benefit of using this approach is that it removes the need to have a trusted party or a custodian to look after “IPO” assets.

The inner red diagram illustrates the smart contract features described in this chapter.  While the Bitcoin protocol could conceivably utilize such contracts, the functionality has not been ‘turned-on’ by the development team (version 0.9 will allow for 80-byte hashes that could include a hash of a distributed contract).  While there are multiple different platforms that will offer such functionality, a stop-gap solution based on bitcoinJS (a Java-implementation) is being developed by Bitpay called bitcore and is described in chapter 7.  Other platforms that can or will shortly allow smart functionality include Colored Coins, Mastercoin, Counterparty, NXT, BitShares, Ethereum and Ripple.

Projects that are being built on top of a blockchain include Colored Coins, Mastercoin and Counterparty.  Both Colored Coins and Mastercoin work exclusively with Bitcoin’s blockchain, and while Counterparty does as well, other projects such as Peercover (discussed in chapter 5) have enabled Counterparty’s currency to bridge with Ripple’s network.

While Ethereum and Ripple are categorized as being the only Turing-complete platforms above, it should be noted that Ethereum has not yet shipped but is expected to in the next six months.  In addition, developers with NXT and BitShares expect to include similar robustness if not full Turing complete functionality at some point in the future.59

One large category that is not distinguished in the above diagram is that of “altcoins.”60 Strictly speaking, anything that is not Bitcoin is considered by early adopters as an altcoin.  Thus everything but Open-Transactions in the diagram is considered by some, as a type of altcoin.  However, this devolves into individual preferences and politics, so it is best ignored.

1. Numerous proposals have been submitted by core developers to improve the functionality; one common analogy used is that Bitcoin core development right now is trying to upgrade the original Wright Flyer to a Boeing 787 without landing.  While many advocates want Bitcoin to be an answer to all payment problems, these limitations likely impair it beyond the role of store of value and remittances.  See chapter 6 for more on remittances and chapter 8 for payment processing details.  See also Hardfork Wishlist [↩]
2. Bitcoin Core Development Falling Behind, Warns BitcoinJ’s Mike Hearn from CoinDesk [↩]
3. See The Bitcoin malleability attack graphed hour by hour by Ken Shirriff and Ripple Labs Chief Cryptographer David Schwartz Talks About Malleability In Bitcoin from Newfination [↩]
4. There are other projects currently under development such as eMunie or even released such as Freicoin. [↩]
5. Colored Coins [↩]
6. Chroma Wallet [↩]
7. Losing the private key to a smart contract (or Colored Coin in this example) could be problematic.  Currently bitcoins are still being lost and stolen despite awareness of web-based wallet vulnerabilities.   If security does not improve, growth might be difficult for smart assets. [↩]
8. eToro [↩]
9. Counterparty.co [↩]
10. See Reality Keys: Bitcoin’s Third-Party Guarantor for Contracts and Deals from CoinDesk and a slightly different idea but in a similar segment, RealityShares [↩]
11. BitcoinBusiness [↩]
12. The Mastercoin protocol supports the OP_Return function.  One way the user-defined assets are tracked in the Bitcoin blockchain is by sending a certain amount of satoshis (5430), just above the dust limit.  Note however, that the dust limit was originally announced at 5430 but was subsequently discovered to be 5460 which may impact some mastercoin transactions. See Dust limit defined as 5460 satoshi instead of 5430 in Bitcoin core at github [↩]
13. How is this functionality achieved?  There is not any ‘syncing’.  Nothing is every ‘synced’ with any blockchain.  Mastercoin does not use ‘OP_RETURN,’ though it plans to add support for it eventually. Counterparty supports `OP_RETURN` now, but it cannot really be used until Bitcoind 0.9 comes out.  Both Counterparty and Mastercoin support using multi-signature transactions to store data in the Bitcoin blockchain.  Just to clarify one misconception, there is no such thing as the ‘0.9 protocol’ — there is a 0.9 Bitcoind.  Also, Bitcoind uses LevelDB and Counterpartyd uses SQLite3. [↩]
14. $1 Million-Plus in Prizes, Contracts at Texas Bitcoin Conference Hackathon from MarketWired [↩]
15. BitAngels [↩]
16. See NXT :: descendant of Bitcoin from Bitcointalk and What is NXT? [↩]
17. There’s a £60m Bitcoin heist going down right now, and you can watch in real-time from NewStateman [↩]
18. There are ways that peers could be compromised vis-à-vis Sybil attacks.  See Establishing the Trustworthiness of Nodes without External Tokens (eg Passports) and Selfish Mining: A 25% Attack Against the Bitcoin Network by Vitalik Buterin [↩]
19. Personal correspondence, Nextcoin.org [↩]
20. NXT – Proof of Stake and the New Alternative Altcoin by Adam Hofman [↩]
21. The potential for such an occurrence is being argued in academic literature; see It Will Cost You Nothing to ‘Kill’ a Proof-of-Stake Crypto-Currency by Houy Nicolas.   A state agent, under the direction of a central bank and simultaneously uninterested in seeing their assets appreciate in value could conduct such an attack.  Otherwise it would likely be cost prohibitive for nearly any other value investor.  In addition, Nicolas’ argument is problematic in that it requires sufficient liquidity, that is to say even if the state actor would be willing and able to spend any amount of funds to acquire the tokens, he or she would still need to induce liquidity to participants holding 90% of the tokens. [↩]
22. Personal correspondence, February 25, 2014.  See also Interview with Graviton, Nextcoin.org Community Founder from Cryptocoinsnews [↩]
23. Satoshi Nakamoto recognized this shortcoming but deliberately chose to use Script to mitigate potential abuses (e.g., infinite loops freezing the blockchain).  One reviewer of this manuscript mentioned that developers should also realize that hypothetical constructs like a DAO essentially involve coding organizational law into programs.  While this may sound easy, law was built to enable release valves of forgiving judgment.  Code is not forgiving.  Thus if something happens in the real world, even the simplest unforeseen effect could derail an otherwise streamlined exchange process. [↩]
24. Bitcoin Magazine [↩]
25. For an example of Ethereum sub-currency contracts see this video from Joel Dietz and Joris Bontje.  See also Writing a Contract in LLL by Gav Wood [↩]
26. I have a friend who used the following method to generate bitcoin addresses and store the keys: 1) in offline mode store the private/public key pairs on USB sticks with Truecrypt partitions, with paper as backup (encrypted and printed out).  To a certain extent this mirrors what Coinbase does.  2) To reduce the chance of vendor back-doors, each of these drives should be different brands bought from different locations.  3) To generate the actual keys you have to deal with the issue of true randomness, plus not leaving any reproducible trace (e.g., logic stored in cache or writing on carbon-copies) thus an individual could buy a dozen non-loaded dice and use this to generate private keys.  4) For users who might be suspicious of the entropy coming from the Linux random number generator (RNG) you could randomly mash the keyboard, turn on the webcam and simultaneously run commands and programs from the start menu to generate some additional entropy.  5) Then use an air-gapped laptop with a freshly boot distribution of Linux. Here in particular you have to be careful as you would need to only use an in-memory distribution (e.g., boot from thumbdrive), because a user does not want the private keys cached anywhere at all on disk.  6) In addition a user would also want a distribution which will work with a standard USB printer for printing purposes because you never want the private keys to go over the wire. [↩]
27. See Scalability and Thin Client [↩]
28. Vitalik Buterin has recently written several more article detailing what he thinks DAOs can and cannot do, see: DAOs Are Not Scary, Part 1: Self-Enforcing Contracts And Factum Law and DAOs Are Not Scary, Part 2: Reducing Barriers [↩]
29. See Mike Hearn, Bitcoin Developer – Turing Festival 2013 video [↩]
30. Intel SGX for Dummies (Intel® SGX Design Objectives) from Intel [↩]
31. Invictus Innovations is leading the development of BitShares; whitepaper [↩]
32. CNRS is part of Groupe d’Analyse et de Théorie Economique.  See The economics of Bitcoin transaction fees by Nicolas Houy [↩]
33. Zanbato [↩]
34. While a completed technical white-paper has not been released, the development team has published The Counterparty Protocol [↩]
35. They had successfully released ‘callable assets,’ stating that, “Assets are now callable, if they are set to be so upon first issuance. An asset may be able to be ‘called back’ by its issuer at a fixed price from a particular date.” See Counterparty Protocol, Client and Coin (built on Bitcoin) – Official from Bitcointalk [↩]
36. Personal correspondence, February 4, 2014 via Bitcointalk [↩]
37. This terminator address is based on Vanitygen.  Based on known computational technology it would purportedly take 93,215,140,000,000,000,000,000,000,000,000 years to generate the private key to 1CounterpartyXXXXXXXXXXXXXXXUWLpVr with an i5 processor.  For critics who claim that the Bitcoin network is insecure, they could prove their skepticism by trying to generate the private key to that address.  See also Wallet security: why only 128 bit for secret seed? from Ripple [↩]
38. See I burned BTC through blockchain.info, how do I access my XCP? from Counterparty.co and the exact address was 1CounterpartyXXXXXXXXXXXXXXXUWLpVr.  On the first day a user would receive 1500 XCP for 1 BTC.  By the end of the fundraiser, it was 1000 XCP for 1 BTC.  Ultimately 2,648,756 XCP were created in total. [↩]
39. See Paul Bohm’s detailed explanation of this mathematical problem. [↩]
40. Personal correspondence, January 29, 2014 [↩]
41. See Bitmessage and Bitmessage Sends Secure, Encrypted, P2P Instant Messages from Lifehacker [↩]
42. See Twister and Out in the Open: An NSA-Proof Twitter, Built With Code From Bitcoin and BitTorrent from Wired [↩]
43. Maidsafe, SyncNet, Bitcloud and Bitcloud developers plan to decentralise internet from BBC [↩]
44. Monetas and Open-Transactions [↩]
45. While the XRP are centrally issued, the gateways are distributed.  The process for being a gateway for a ‘coin’ generally works as follows: 1) announce you are issuing a coin, 2) anyone can “trust” you for the coin, 3) accept the real coin, 4) make a Ripple payment for the coin.  Thus you can create fully backed precious metals on the Ripple network.  Ripple itself does not send USD, EUR, CAD or other currencies. It actually sends IOU’s for these currencies which must be redeemed by specific issuers who are acting as “gateways” into and out of the legacy banking system. [↩]
46. One way to audit and verify if a 3^rd party gateway (and exchanges in general) is not running a fractional scheme is to implement a ‘proof of reserves’ process Greg Maxwell recently proposed.  Another option that could happen is that exchanges may hire independent auditors in order to become covered by insurance; these audits could then be posted.  An unnamed insurance company purportedly provides services to one Bitcoin vault called Elliptic, which protects against a failure in a business’ storage methods, with customers opting for a “liability limit” for how much they want covered.  Another idea being discussed is some sort of FDIC-like insurance.  A company in beta called Inscrypto, which is located in Boston, claims it will be a “privately funded, decentralized version of the FDIC,” to help you “reduce or completely eliminate the risks of owning bitcoin.”   It is likely that following the Mt. Gox bankruptcy, many exchanges will seek such independent measures and likely have an incentive to do so (e.g., satiate consumer demand, provide transparency as a precursor to being acquired in the future).   See also Proving Your Bitcoin Reserves by Zak Wilcox, Bitcoiners Demand Greater Transparency in Exchanges from Cryptocoinsnews, Audit Report: Transparency and Accountability from Coinkite, After the Mt. Gox fiasco, calls for regulating bitcoin from Pandodaily and Will Bitcoin’s Libertarians Pay for Private Deposit Insurance? from BloombergBusinessweek [↩]
47. Introducing Ripple by Vitalik Buterin [↩]
48. See Making Money from Technology Review and Ripple Charts [↩]
49. Ripple credits [↩]
50. Chris Dixon of Andreessen Horowitz was one of the first persons to use that term.  Naval Ravikant founder of AngelList popularized the term “programmable money” which has a similar meaning.  See Real money starts to pour into math-based currencies like bitcoin from Quartz and Inside Bitcoin, The Programmable Currency For Our Digital Future from TechCrunch [↩]
51. A successful double-spend attack could be conducted against a proof-of-work-based algorithm if 51% of the hashrate is controlled by a malicious agent; and a similar attack is theoretically able to successfully take over a proof-of-stake if 90% of the token is controlled by one agent.   But there are many ways to recover from it (e.g., hardforks) and this topic has filled countless volumes already.  Yet, for an objective view on this matter of network attacks, I asked Nick Szabo (Personal correspondence, January 25, 2014), who had some original insights about how to prevent and mitigate this issue:

    “One contingency is to have a bunch of different cryptocurrencies around […] and if one gets successfully attacked users switch to another. We already have enough cryptocurrencies around for this purpose, but this doesn’t help the people holding Bitcoin or who’ve made other Bitcoin-specific investments. And there are substantial costs in switching to a new cryptocurrency, and such a crisis might persuade many merchants to give up on cryptocurrencies generally rather than switch.

    A practical means of disaster preparedness is for a number of independent engineers and auditors to keep copies of the block chain, as up-to-date as possible, even if they aren’t participating as a miner or mining pool. Just the fact that a few good engineers have up-to-date copies of the block chain should be enough to dissuade most 51% attacks. 51% is enough to persuade the cryptocurrency algorithms to believe a lie, but it’s not enough to persuade engineers (or auditors with suitable tools) who manually inspect the block chain, if the payor or payee who’ve been blocked or defrauded resend the original payment instructions directly to those engineers. Of course we don’t want to rely on such a manual process in the normal course of business, just for dire contingencies.

    In the event of a 51% attack there is a fork in the block chain, and the job of these engineers or auditors would then be to persuade users to use the minority but correct block chain and exclude the incorrect majority. It would be expensive but doable. Not something you want to normally see happen.

    Another way to put it is if there is a 51% attack we have to fall back on methods of ensuring integrity that, like the traditional financial system, are manual and expensive, and the big cost savings from the automated security are temporarily lost.  You might call this [ad hoc solution] “proof by engineer,” which would be replacing proof-of-work in the temporary emergency for the purposes of the transactions being disputed in the block chain fork.” [↩]

52. CrossCoin Ventures [↩]
53. The way the current system is setup, remittances and funds sent abroad go through multiple institutions via ‘correspondent accounts’ or ‘correspondent banking.’ [↩]
54. Ripple Developer Conference 2013: Future Focus of Our Engineering Team presentation by Stefan Thomas [↩]
55. Peer-Assisted Key Derivation Function (PAKDF) by Stefan Thomas [↩]
56. See What is Homomorphic Encryption, and Why Should I Care? by Craig Stuntz, Blind Signatures for Untraceable Payments by David Chaum and Untraceable electronic mail, return addresses, and digital pseudonyms by David Chaum [↩]
57. Blind Signature Scheme by Asanka Balasooriya and Kelum Senanayake [↩]
58. I designed it with Creately; the image is released under Creative Commons 4.0 Attribution license. [↩]
59. A Turing-complete solution proposed by NXT is to use the Automated Transaction Specification [↩]
60. Another category that was not highlighted is the proof-of-work algorithms: both Bitcoin and Namecoin are SHA256d based and Litecoin and Dogecoin use Scrypt. [↩]