Blockchain - Revolution, or just another database?
Things continue apace at Cloudy Towers, and as a result I'm somewhat slower at blog writing this year. Today's topic was inspired a little by my colleague Mike Mandeville, who presented on all things blockchain at our annual kick off event in Lisbon earlier this month. Sadly I missed his presentation, but the buzz surrounding it (plus Bitcoin trucking along at nearly £3K/coin at time of writing) reminded me how excited people are about the topic, but how poorly understood it is. So, what follows is somewhat of a blockchain 101. Let's do it.
Why the excitement?
As mentioned above, blockchain gets people excited(!) Why? Well, probably a mixture of both "interesting tech/maths" and "money". As documented here, there's a ton of investment into this area from financial services organisations, both small and large. Beyond that, there's a real passion at the user level (mainly for the Bitcoin currency sitting atop a lot of public blockchain usage). This links to people using a product that they also have an ownership stake in – and are directly benefiting financially from its success as they and others use it. Whilst Bitcoin is very useful, we wouldn't be talking about it much if it was just a decentralized global ledger….
What is it?
As a respected person of IT, where else would one start trying to understand tech, than at Wikipedia? Well, alright maybe not, but their definition of a "sequential distributed database, found in cryptocurrencies derived from bitcoin" is actually sort of useful for this discussion since it brings a word you're likely more familiar with into the discussion: 'database'.
Taking it more high level for a moment, blockchain allows people who do not know or trust each other to build a dependable ledger ( a "ledger" is simply a record (list) of transactions of some kind). Basically it's some software, run on a network of computers, that creates a tamper-proof ledger.
Bitcoin is the most famous use of blockchain tech so far. One can buy and sell bitcoin using computers anywhere in the world to ensure robust and reliable transactions. It's fully decentralised, i.e. it has no bank "middleman" in the transaction, and those transactions are public and visible to all, on all computers at the same time. It's immediately verifiable and it's permanent. It even self-audits every 10 minutes to ensure it's consistent. Nice.
You mentioned databases?
A blockchain is basically a distributed database that maintains a shared list of records. These records are called blocks, and each encrypted block contains the history of every block that came before it with timestamped transaction data down to the second. Thereby…..chaining those blocks together. Hence the name… 'blockchain'.
Wow. Blockchain replaces my database?
No. Please don't fall prey to the marketing hype. In the majority of use cases in the real world, a database is a better fit than a blockchain solution. Always use that as a test when discussing blockchain tech: Could I do this in a database?
There's a superb article here on this very topic. No point in me re-inventing that particular wheel. If you don't meet those criteria listed, you almost certainly don't need a blockchain based solution.
Sorry for any lost sales as a result of this…..
What's this 'mining' thing….?
It's big business! Entire companies exist to try and generate Bitcoin by performing this service – at scale! All of the nodes in this most commonly used blockchain join voluntarily, but are given an incentive for taking part – the chance to "mine" Bitcoin. Basically it's a competition, rather than something using shovels, which is won by solving hard maths (note the "s", America). As the competition heated up some years back, dedicated hardware, plus an awful lot of electricity is now required to win the remaining Bitcoin.
Why do we care, beyond the tech?
Well, transactions, contracts and the records of them define how our countries and societies operate – not just at the financial level, but at the legal, political and identity level too.
You can begin to conjure up in your mind a future utopian world whereby contracts are code. Stored in open databases, protected from abuse, editing and adjustment. Every agreement, every payment could be stored and recorded digitally in perpetuity. Some argue that lawyers and banks wouldn't be needed (which is either good or bad, depending on your job!). I'm a little less dramatic on the topic in the medium term, and view brokers as the first to need to radically change their business model – as fundamentally people and companies can freely transact *safely* without their involvement. Longer term, do banks need to rethink their game plan? Yes. Will they collaborate on a sensible open standard in the near future? Almost certainly not.
Public or private blockchain?
This reminds me a bit of a debate about public vs. private cloud….and of course the answer there is that public cloud wins. Is the same true of blockchain? Maybe….
Public blockchains are the mostly widely known – and as the name suggests, they're open to all to take part. They need a token or currency to allow people to take part (to limit people from using the network for free), e.g. Bitcoin.
Private blockchains are much less well known – as you can only access them via an invitation. Typically this could be for n companies who want to work together in trust. One can argue of course that they must already trust each other somewhat to collaborate on this….thereby negating some of the value. However, they do still provide a reliable source of truth – without a middleman being involved.
There's a ton of examples and thought on this topic here. My favourite quote being this one:
"Private blockchains will not revolutionize the financial system. Public blockchains, however, hold the potential to replace most functions of traditional financial institutions with software, fundamentally reshaping the way the financial system works." Ryan Charles, founder of Yours.Network
As is the way with such things (see the Internet, public cloud and open source as examples), fully open solutions build network effect and momentum which is hard to stop. You don't see the same with closed and private systems – although the past is not always a guide to the future of course. Will the banks really collaborate effectively on a private solution (r3 are trying and doing well!)? Or will an open network like Bitcoin build enough momentum to really disrupt things? Come back in 10 years for the definitive answer. As argued well here, the answer probably depends on the use case.
For those looking to experiment and explore this area, huge credit to Microsoft for the work they've done to offer blockchain as a service atop Azure – this would be a great place to start. The rest of the market appears to be sleeping on this topic.
Opportunities for blockchain?
Ignoring the "DISRUPTION!!!!!" alarms for financial services in general, there are major opportunities for smaller banking institutions to save significant time and money on costly compliance processes (due to the clearly auditable nature of blockchain). This could allow a challenger to scale up, less inhibited by regulatory barriers.
Other cool examples? Well, I quite like real world example from DocuSign & Visa. Stoners of the world might appreciate this 'potchain' article. The registration of land ownership is also an interesting area in many developing countries – and perhaps it's in the less developed world, that many of the primary benefits of shared trust can be seen most vividly. Or back in the western world, how about Axa automating insurance payments? (this latter example uses Ethereum, a platform designed specifically for contract based workflows).
All gravy from here then?
Well…..no. There remains a ton of hype, which confuses the real use cases. There are scaling challenges as Bitcoin is widely adopted, there's a major government (China) shutting down Bitcoin exchanges due to regulatory concerns. The ever interesting Simon Wardley noted some time back that Bitcoin could actually be an economic threat to the state in a well argued and slightly worrying note here (primarily due to lost tax revenues). Due to the current ease of anonymity, there's also a fair amount of hacking, aka theft, going on which doesn't help the image of blockchain in general. Bitcoin is also brutally volatile, which doesn't lend itself to being seen as a credible alternative in some quarters.
My god there's a lot of hype. In many cases a good old fashioned database driven solution will solve the problem at hand more effectively. I do think in time we will see challenger banks emerging using blockchain based solutions to scale quickly. I also do think there are problems that can be solved with private blockchains, related to specific industry use cases – they just need to be well understood. Microsoft appear well positioned to capitalise on this.
Am I bullish on blockchain? Yes. On cryptocurrency? Yes. Bitcoin, Ethereum and Ripple – in that order.
For more posts by Chris Bunch, click here.