Blockchain is like the internet before the browser

WARNING WARNING WARNING

This is a semi-technical blog update and it is rather long (20 minutes read). If you don’t like dealing with bits and bytes, you may want to turn away now.

I’ve been meaning to also write a blog update on all things about blockchain for a while, but just haven’t had time.  As it’s summer, I now have more time, so here goes.

Blockchain is a confusion because there are so many variations.  The bitcoin community claim the only blockchain that matters is theirs, because they have the most market value invested in their currency.  The problem is that the bitcoin blockchain isn’t very useful in the context of finance and government, due to its restrictions.

Its restrictions include that it is decentralised but the more the blockchain grows, the larger the requirements become for storage, bandwidth and computational power.  In fact, it demands more full nodes (see endnote) in the network, leading to a risk of much higher centralization if the blockchain becomes large enough that only a few nodes are able to process a block.

Secondly, the bitcoin-specific issue is that the blockchain has a built-in hard limit of 1 megabyte per block, which is the reason why validating transactions takes at least ten minutes and sometimes over an hour.  As I write this, the bitcoin network is processing just under 10,000 transactions an hour, which is around 160 transactions per minute or less than three transactions second. Removing this limit requires a “hard fork” to the bitcoin protocol, which is unacceptable to most as it means it invalidates the existing network and all of its history.

There is also due to the above two reasons, a high processing fee charge incurred for bitcoin transactions, and the potential for those fees to increase as the network grows.  When looking at this, we can see that each block verification rewards a miner with 12.5 bitcoins (this halved recently from 25). At current exchange rates, that equates to $8,150. The network currently processes around 1,500 transactions per block for an average cost of $5.43 per transaction (it was $23 two years ago, so it’s come down a lot, and is more accurate than the figures shown on blockchain.info).

In other words, the core issues of bitcoin are that it has to move to centralisation to scale and that would defeat its purpose.  As a result, it is a high cost, low volume processing system. Some believe this is a non-issue, such as well-known bitcoin expert Andreas Antonopoulos .  In a talk at CoinJar last fall, he made the following comment: “I have no worries that bitcoin can scale, and the simple reason for that is that I know that IPv4 can’t, and yet I use it every day.”  However, bankers will not commit to such a network.  There are other reasons why banks could not use such a network including transparency.  The bitcoin blockchain allows anyone to see a transaction and its history.  That won’t work for a secure market where only counterparties should see transactions.

This is why Ethereum has taken off as the bitcoin alternative.  Ethereum has its own currency, the Ether, and is the second most capitalised cryptocurrency out there.  It’s the backbone of many developments from Microsoft’s blockchain-as-a-service to the Bank of England’s next generation real-time gross settlement (RTGS) developments.

Now this could be slightly concerning if you’ve been following the recent events of the DAO, the Decentralized Autonomous Organisation.  The DAO is a computer program built by Simon and Christoph Jentzsch that doubles as an investment fund. It lets users invest their money into the investment vehicle, which will pick companies to invest in by gathering votes from investors who hold DAO tokens. Built on the Ethereum blockchain—a platform for secure transactions—it quickly became the biggest crowdfunding project in history raising $150 million in a month.  Then a whole pack of that investment fund was hacked, with over $50 million in Ether stolen.  The result is that, unlike the bitcoin approach, Ethereum determined a hard fork in their currency was needed which went live yesterday.

A fork can be a soft or hard rollback to reset the network, like a reboot.  The initial proposal was a soft fork. This would entail a majority of the Ethereum miners (those who verify transactions on the network) voting on the roll back. Unfortunately, a security flaw was found in the voting process, which eliminated this option.

That leaves a hard fork, where the core developers of Ethereum unilaterally make the decision to essentially create a new version of the network with different rules than the original. Then, miners, exchanges, and other major apps that are built on it need to decide if they want to a part of the new version of Ethereum or the original.   Luckily for Ethereum about 80% of their community have so far upgraded to support the process, so it should succeed.  But it does raise questions and hit confidence in the process.  Equally, Ethereum is also scalability challenged, processing just 100 transactions per second today.

This is why there are other blockchains out there that might be more robust to better serve the banking community, such as BigchainDB, Ripple and R3CEV Corda.  That's not an exhaustive list, as there are others out there like SafeCash and HyperLedger, but I choose these three as being the ones of most relevance to financial institutions and their use cases at this time.

BigchainDB is a new blockchain development that overcomes the scalability issues of bitcoin blockchain.  Released in February 2016, it generated a lot of hype due to impressive scalability benchmarks: it was able to process a million transactions per second.  BigchainDB also got a lot of praise from private blockchain enthusiasts with many partnerships including Eris Industries, Everledger, Capgemini and more.

This may well be the blockchain platform that could work for banks and their payment networks, such as Visa and SWIFT, but it also has challenges.  In a Reddit stream it is claimed that the BigchainDB structure is fundamentally flawed because all of the network’s nodes, by design, connect to a single RethinkDB cluster.  RethinkDB is an open-source, scalable database that makes building real-time apps dramatically easier.  The issue is that if something bad happens to that database cluster behind BigchainDB, then the whole blockchain goes down.

Ripple is a real-time gross settlement system (RTGS), currency exchange and remittance network based upon the Ripple Labs version of blockchain.  I’ve interviewed their CEO Chris Larsen for the ValueWeb book – you can read that interview in full for more background – but the key paragraphs explaining what they’re up to include:

“Ripple was borne out of Bitcoin’s incredible revolution. Bitcoin solved the double spend problem, creating, for the first time in history, a digital asset and decentralized ledger with no central operator. This breakthrough means people can send this new currency (bitcoins) to anyone else in the world in minutes and at no cost, instead of days and at the high expense in today’s traditional systems.

“Ripple is optimized to serve as improved, IP-based infrastructure for today’s payment systems. Amongst distributed payments technologies, Ripple uniquely works with any currency (dollars, euros, yen, etc.), and it settles transactions, including cross-currency transactions, in five seconds.

“So, whereas bitcoin depends on consumers and merchants adopting the currency and exclusively using the blockchain for payments, Ripple is meant for financial institutions to integrate into their core systems to dramatically increase the speed and lower the cost of payments. Financial institutions and networks using Ripple never have to touch digital currency; they continue to deal in the fiat currencies they deal in today.”

Royal Bank of Canada, which is leveraging the technology for remittances, whilst Standard Chartered and DBS have begun using Ripple's tech for trade finance applications. Santander recently launched an app for international payments using Ripple Labs’ blockchain technology, and more announcements are on the way.  So this is a credible attempt to build something but still has questions, as you will see at the end of this blog update.

Finally, we have R3 Corda.  Corda takes all the great things that the bitcoin blockchain offers, and then removed all the bits that would not work in the banking community.  For example:

• Corda has no unnecessary global sharing of data: only those parties with a legitimate need to know can see the data within an agreement
• Corda achieves consensus between firms at the level of individual deals, not the level of the system
• Corda’s design directly enables regulatory and supervisory observer nodes
• Corda transactions are validated by parties to the transaction rather than a broader pool of unrelated validators
• Corda supports a variety of consensus mechanisms
• Corda records an explicit link between human-language legal prose documents and smart contract code
• Corda is built on industry-standard tools
• Corda has no native cryptocurrency

Launched in April 2016, Corda is in early days developments but the firm make a comment about the key difference between this and bitcoin and Ethereum:

“The applications institutional customers want to use blockchains for are far more complex than anything done on top of Ethereum/Bitcoin so far, and so when we started designing Corda I knew we’d need to find ways to radically improve developer productivity … if making a simple app for Corda took as much work as making a simple app for Bitcoin does, we’d have a serious problem.” Mike Hearn, R3’s lead program engineer and former Bitcoin Core developer

OK, that makes sense … or does it.  During my thinking about all of these developments – Ethereum, BigchainDB, Ripple and R3 – I kept coming back to an interview with Andreas Antonopoulos, the leading Bitcoin luminary and author of the book Mastering Bitcoin and co-founder of the Blockchain Futures Lab.

In a 50 minute chat with journalist Mark Frauenfelder, there’s a sequence at the start that captures the essence of why some of our discussions may be on the wrong track:

Mark Frauenfelder: How are enterprise/private blockchains different from bitcoin’s blockchain?

Andreas Antonopoulos: First of all, I think we’ve seen this before. It’s very similar to the earlier stages of the consumer Internet, when the Internet was breaking out of the educational environment after the 1993 National Science Foundation decision to open it up to commercial use. At the time, you had people doing things like writing manifestos for the Internet — John Perry Barlow writing “A Declaration of the Independence of Cyberspace.”

The corporations were really excited about this technology. They could see this was going to be disruptive and they certainly wanted to at least defensively prepare themselves for it. The same narrative is playing out with bitcoin.

What they did then was try to create other networks with TCP/IP or intranets or various things like MSN and Compuserve — closed, curated wall ed garden versions of the Internet, where everything would be tame and PG-13 and have all the promise of the Internet, with none of the smut. None of the offensive anything, just a clean Disneyfied version of it.

Of course, those places failed because they were boring and because the real premise of the Internet wasn’t TCP/IP, it wasn’t interconnectivity. It was openness, it was permissionless innovation, it was content creators being individuals rather than top down. It was peer-to-peer. All of those things exploded with innovation until eventually everything else just seemed too boring and became obsolete.

The same thing is happening now with bitcoin. Bitcoin is finance without borders, finance without identity, finance for everyone, finance without banks and governments, finance without authority, without central points of control, etc. We basically present bitcoin to the world as open, decentralized, permissionless innovation, peer-to-peer finance without borders. The banks and the corporations say, “Oh, that’s awesome. We want that. Only without the open, decentralized, peer-to-peer, borderless, permissionless part. Could we instead have a closed, controlled, tame, identity-laden permission version of that please?”

We promoted the blockchain as, “Hey, listen. It’s not just about bitcoin.” This was two years ago. A lot of people who are in bitcoins started saying, “Yeah, it’s not about bitcoin alone.” You have to understand because people thought bitcoin was currency. The argument was it’s not about the currency. It’s about this trust network as a platform, as a global platform and what it can do.

We started saying, “Pay attention to the blockchain, not just bitcoin.” The banks took this all the way to the end and now they’re saying “blockchain without bitcoin.” Through 2015 you had this fad transformation where it became very cool and popular to say, “You know, after all, it’s not about bitcoin. It’s about the blockchain instead of bitcoin.”

“Bitcoin is going to fail but blockchain is going to change the world.” That’s like the great whitewashing. It’s hilarious to watch because, for one thing, it exposes their fundamental misunderstanding of what this is, where the value is, and how it works.

Will enterprise blockchains be successful?

They’re going to be successful in replacing even more insidious forms of centralization like centralized clearinghouses, like Swift, Depository Trust & Clearing Corporation, and other quasi-private, monopoly companies that control and clear equities and bonds and commodities.

They’re going to revolutionize Wall Street trading floors. Revolutionize, meaning they’ll drag it out of the 70s and maybe bring it in into the late 90s. That may be exciting to bankers. It gives me a major yawn.

Let’s start with the technical. The really cool thing is the open, decentralized blockchain, that is, a blockchain that has no central authority and is open for anyone to participate without asking permission because the trust mechanism doesn’t depend on access control like everything that came before. It depends on proof-of-work consensus, which is a decentralized approach to achieving consensus and trust over the state of the system.

You can only have that if you have a consensus algorithm that is decentralized and anonymous, such as proof-of-work mining and bitcoin. We haven’t found another good enough one yet. That only works if you have a reward associated with mining to create this game-theoretical competition and balance between the incentives for private profit and trust in the system. We know that works. It works very well and it’s scaled so far to double-digit billion dollars, which had never happened before and that in itself is astonishing.

The idea that you can make a blockchain without bitcoin or without a currency — you can really, but the only way to do that is to replace the decentralized consensus mechanism with a committee, which is what they call “signing instead of mining.” You get ten banks together and, instead of mining blocks competitively in an open way, they sign the blocks, they take turns.

There’s an algorithm for that called Byzantine Paxos, which is a distributed consensus algorithm. It’s basically what we used to call three-phase commit. It’s like a round-robin signing party where you elect a leader each round and they sign for all of the transactions. As long as you trust that nine banks are better than one central clearinghouse and won’t cheat as much, then okay.

But that system is not immutable. The banks can get together and roll back the blockchain and write something else. With bitcoin’s proof-of-work, you can’t do that. No one can do that because you can’t get enough miners to put enormous amounts of hashing power behind that kind of effort. You lose immutability. You lose decentralization.

Because the model is based on this round-robin of trusted partners, you lose the permissionless open access so it becomes something where you have to have permission to join and almost certainly isn’t open in public. It will almost certainly have to have ID added back into it, which brings all of the disadvantages of identity theft if you’re a consumer and will give us an absolute treasure trove if you’re Wikileaks. Imagine what happens when Lockheed Martin’s blockchain leaks. You’re going to have a time stamp for every financial transaction they’ve ever done. I can’t wait.

They’re adding identity back to this, which is absolutely idiotic. You can have a blockchain without the currency. You just can’t have an open, decentralized blockchain where the trust lies in this game-theoretical competition that is neutral and based on rules. Instead, you have to put trust in third parties, counterparties who are the signers.

That then becomes business as usual, which may be revolutionary for banks. But, immediately it cannot be borderless because in order to maintain control over a system like that, you have to control identity. That means you have the old Bank Secrecy Act, Know Your Customer rules, and Anti-Money Laundering rules which, in themselves, are responsible for this problem of global economic exclusion, the reason people can’t participate in the banking system. So you re-erect the borders.

Essentially what it is, is an intranet and an intranet is boring. It’s stale content. It’s the less secure version of everything out there. You can’t run the apps you want because corporate IT didn’t allow them. It’s stagnant innovation. In the end, it also lags behind in terms of security because it’s not exposed to the kind of robust peer review that an open Internet-based system has to have to survive. While bitcoin is getting stronger and stronger with security, these things will actually wither.

I did a presentation called Bubble Boy and the Sewer Rat, about that very topic.

The idea being that intranets end up being these really insecure places where you’re running Outlook and FrontPage and old versions of Apache that haven’t been patched. Whereas on the Internet, if you’re Facebook, if you’re Google, if you’re Apple and you’re operating Internet applications, you have to be robust. You have to respond to vulnerability. You have to make systems that are antifragile, resilient to attack and they’re constantly evolving and they’ve become very robust.

Whereas, the things that you put in a bubble are like Bubble Boy. If you remember from the 70s, a kid who lived in a bubble because he had no immune system is like raising a child in a bath of Purell. You’re going to turn them into an allergic weakling.

In meantime, bitcoin isn’t living in a bubble. Bitcoin is a sewer rat. It’s missing a leg. Its snout was badly mangled in an accident in last year. It’s not allergic to anything. In fact, it’s probably got a couple of strains of bubonic plague on it which it treats like a common cold. You have a system that is antifragile and dynamic and robust.

This is the big argument of 2015. It’s the “let’s take bitcoin, cut off its beard, take away its piercings, put it in a suit, call it blockchain, and present it to the board.” It’s safe. It’s got borders. We can apply the same regulations. We can put barriers to entry and create anti-competitive environment to control who has access. It will be more efficient than our existing banking system.

If what they want to do is the Outlook of currencies, go ahead, by all means. That’s what intranets turned into. It’s stale wiki content hosted on FrontPage and Outlook. People hate that. Eventually, successful, vibrant, innovative companies are the ones that turn their IT infrastructure inside out. In the future of finance, successful, vibrant, and innovative banks are the ones that turn their infrastructure inside out and make it part of an open, borderless financial system that serves the other 6 billion, that serves all the people who have been excluded from finance.

Because here is the thing: we’ve had the Internet now for 25 years. If you look at the statistics, financial inclusion is getting worse. That doesn’t make any sense unless you consider that, in the meantime, the traditional currencies and digital finance systems have been getting more and more prone to surveillance and tied to identity where everything is tracked. They’re closing in on themselves in order to maintain this ironclad control over who sends money to who, all in the name of terrorist prevention and anti-money laundering, which is bullshit because HSBC could money-launder billions. Always have since their birth in the Opium Wars and nobody ever goes to jail. God forbid, a kid buys a joint with bitcoin and everybody’s up in arms.

What is this doing in the meantime? The real question is how many billions of people are you going to sacrifice on the altar of money laundering, because the current answer is just a bit over four billion people. Four billion people who are in the deepest parts of poverty are excluded from banking and financial systems and it is because of Know Your Customer and Anti-Money Laundering and borders and control of networks.

You cannot create global finance and economic inclusion on the back of a carefully controlled show-us-your-papers identity-based system where everything is tracked. What you create is a global surveillance dystopia. Our entire financial system is heading into this thing where everything is surveilled. Bitcoin is heading in exact opposite direction. No identity by default, from weak pseudonymity to a stronger and stronger anonymity as time goes by.

As a result, it doesn’t do borders. It doesn’t care about borders. It doesn’t do Know Your Customer. It doesn’t do Anti-Money Laundering. It doesn’t do those things because those things are bourgeois concepts of the privileged financial elite. Those bourgeois concepts have a four-billion-people-in-poverty price tag.

Are banks and financial institutions trying to centralize the blockchain?

Absolutely. It’s not so much centralize the blockchain. It’s that they’re starting with a premise that finance must be controlled in order to be secure. In order to be controlled and secure, it has to have identity. In order to have identity, you have to control all the endpoints and it has to be a closed system. In order for it to be a closed system, you have to control access and centralize everything as much as possible.

You want end-to-end control. You can’t have leaks because leaks are the places where money laundering happens (money laundering for brown people, not money laundering for rich white people. “We don’t care about that,” [say the powers that be]). The end result is that they’re making the system more and more centralized in this desperate attempt to fulfill a dream that started in the 70s that you could create this perfect Disneyfied world economy where everybody’s finances are known.

When they talk about Know Your Customer and Anti-Money Laundering rules, they assume the information all flows in one direction — to them. They don’t assume that the CIA’s money laundering for weapons purchases are seen by the Chinese. They assume what that means is that the Syrian dissidents’ money is known by the CIA. It’s only one-way flow.

Anyway, not to get too political, but the point is that money has been heading away from cash and towards digital and, with that, digital came an attempt to turn it all into a giant mainframe, highly centralized. But the problem is that within the centralized systems, of course, corruptible centralization of power corrupts absolutely. It creates fragility, and so in this march to centralization for reasons of control, which is taken for granted among these circles, the end results have been fragile markets, fragile economies, fragile currencies, and increasingly, corruption.

The companies that have control of a clearing of equities front-run the transactions of their friends and delay the transactions of the ones they’re competing with. You have all of these rigged markets where privileged access is the name of the game.

What the banks want is they want it a hell of a lot more centralized and controlled than bitcoins and they certainly don’t want this ragtag army of misfits to suddenly create competition where they had this nicely parceled-up monopolistic market. But they don’t want it as controlled as the one clearinghouse. For them, a blockchain is this great liberalization. It’s decentralizing their clearinghouses. That’s because their definition of the scale of centralization is very narrow.

 

You can read and listen to the full interview with Andreas over here, and I think that what this demonstrates all-in-all is the very nascent nature of blockchain technology.  There are so many blockchains out there, and the challenge is which one to focus upon and which one will win.  Or does one win at all, as many are saying there will be multiple blockchains and interoperability and standards will be the key.  As can be seen it is still early days but, for those who get on board early, the rewards will be immense.

 

Endnote: A bitcoin full node 

A bitcoin full node is a program that fully validates transactions and blocks. Almost all full nodes help the network by accepting transactions and blocks from other full nodes, validating those transactions and blocks, and then relaying them to further full nodes.  Most full nodes serve lightweight clients by allowing them to transmit their transactions to the network and by notifying them when a transaction affects their wallet. If not enough nodes perform this function, clients won’t be able to connect through the peer-to-peer network—they have to use centralized services instead.  Many people and organizations volunteer to run full nodes using spare computing and bandwidth resources—but more volunteers are needed to allow bitcoin to continue to grow.