Hands of a guy on laptop keyboard

Bitcoin is at a crossroads. Should it scale up for the future or stay as it is?

24 August 2015

Bitcoin, as a leading cryptocurrency, is at a point that everyone knew it would come to. Everyone hoped this wouldn’t be potentially threaten the stability of the whole blockchain invention. Nevertheless.

Bitcoin is at a crossroads. Should it scale up for the future or stay as it is?

The Bitcoin system records all transactions made globally in a unique ledger of transactions called a blockchain. Two years ago, major Bitcoin engineers and developers raised concerns about the technical limitations of the original Bitcoin blockchain, imposed by its creator Satoshi Nakamoto. The size of a block in a blockchain is limited to one megabyte (1MB) of data per block, so this ledger can be run on any computer. In future, this could undermine the possibility of scaling up this electronic payment system to a global level. For example, the payment network Visa conducts about 45,000 transactions per second (yes, you read that correctly). Currently, Bitcoin supports around seven transactions per second (35% of the 1 MB per block limit).  In order to grow and accept its popularity at the level of global payment networks, Bitcoin will need a major technical overhaul. Clearly there is a serious problem here. Who will decide what kind of change to impose? Does this action undermine one of the core characteristics of cryptocurrencies: decentralisation?

Mike Hearn and Gavin Andersen, two (of five) most senior Bitcoin developers from the Bitcoin Foundation (a foundation which maintains and supports the original open source program blockchain), will start their own ‘version’ of Bitcoin core software. This new version (or fork) named BitcoinXT will allow blocks to be larger (8MB and more if needed).  They have reasoned this move by Satoshi’s original vision of Bitcoin as a system that would scale up to a Visa-level transaction volume.

How is this change possible? Since blockchain is open source software, full code is available for anyone to alter. Any alteration is accepted or denied through the consensus of users (nods) that run the original core software (blockchain) on their computers. Many new versions and improvements are introduced using this approach. Core developers would announce the change, discuss it with the community, and start implementing. Gradually the community would accept this through a decentralised mechanism of emergent consensus. Emergent because it happens in an unclear moment; there is no predestined date as with the election, voting, or polls. Instead, consensus emerges from the asynchronous interaction of thousands of independent users, all following simple rules.

Even though two major developers are rooting for this change (BitcoinXT core client is already run by 7.7% of users) other chief engineers and part of the wider community oppose this idea. Two years later, a possible consensus on the Bitcoin leadership team doesn’t look likely.  Communication has broken down; both sides feel they are vigorously defending decentralisation and the True Bitcoin vision. The community is divided.

To add to even more drama, a letter was sent to the core developer mailing list, allegedly signed by Satoshi Nakamoto, the creator of blockchain, writing that ‘he’ opposes the proposal of block size change. Of course whenever that name appears in public, media attention is drawn as the search for the ‘mysterious Satoshi identity’ continues. No-one can claim with any certainty that the letter was sent by Satoshi but Mike Hern has an answer in his blog: ‘Let’s imagine he turned up and said he’d rethought, and Bitcoin was a bad idea. Should everyone just give up and walk away? Or accept that people’s ideas do change. My article quotes Satoshi a lot, but that’s not to imply he’s some sort of god or absent dictator.’

The main argument from opponents is that any change will open up a Pandora’s Box and destroy the confidence people have in the stability of Bitcoin. If the 1MB maximum block size limit can be changed, why not the total number of Bitcoins issued? And why is this so urgent when there is more time before the limit is reached?

Chief Bitcoin engineer Gavin Andersen addresses this concern in his blog  arguing that acting in the faith of further development is always better that discussing it to the point where the system is gradually cluttered and unsecure.

The parallel existence of two core Bitcoin clients in the long run can hurt Bitcoin the most. After the introduction of BitcoinXT in January 2016, and even if only 75% of miners (users that run blockchain) switch on this new extension, these two versions will be incompatible.  Transactions made on one would not be acknowledged in the other. There is going to be a losing side in this event. Regular Bitcoin users will not be affected by this change, though, since the wallets/clients they use will just continue to follow the version of blockhain that ‘wins’.

Mr Joi Ito, Director of the Massachusetts Institute for Technology (MIT) Media Lab, and former ICANN board member, has raised the possibility that something like ICANN’s Governmental Advisory Committee (GAC) may be necessary to ensure that governments feel heard by the community as future decisions are made.

‘I think the key is that the regulators shouldn’t be running the show and should take a supporting stance. If you look at ICANN, for example, there is a “Government Advisory Committee” (GAC) as one of the groups that advise the board, but the board is independent and the government doesn’t control ICANN, but is part of the conversation that informs ICANN and through that process, the governments voice their concerns, but in the end, generally feel like they were part of a consensus process and allow ICANN to make the rules.

I’m not sure that we need an ICANN for Bitcoin, but I think that what WOULD be good is some sort of way for all of the governments to voice their concerns and then become comfortable with how the community interprets their concerns. What would be bad is if each state and government came up with lots of regulations that contradicted each other or were slightly different so that every Bitcoin company or idea had to get permission in each state and country and had to be slightly different in each region.’ (Reddit link)

Many from the Bitcoin community drawn to this idea by the promise of ultimate decentralisation, strongly oppose this opinion stating that GAC involvement in the Internet governance debate at ICANN has been far from ‘not influential’. They especially point out the governmental objections to the top-level domains like: ‘.amazon’ or ‘.patagonia’ arguing that GAC aggravated ICANN’s evolution from a technical agency into a policy organisation.

There is something interesting in this proposal from Mr Joi Ito. We can remember the days when the whole IANA function was managed by one man, Jon Postel, and kept in his private notebook (started with a Proposed Standard Socket Numbers). Nevertheless we trusted those data and let the innovation unfold and spread without the fear of possible future ways of misuse. Standardisation moved the game forward. 
We can be on the verge of scaling up and further developing a cryptography-based electronic currency by more active stakeholder engagement, or we could see a sharp downturn in Bitcoin’s prospects pushed by user societies paralysed by analysis, tests, and assumptions of possible ways for future development.
That is the thing with the ‘emergent consensus’. If a decision-making body decides one thing, the community can/will participate in that. But if the community instead decides to take another decision, there is nothing the formal structure can do to stop it. 
Whatever happens, the idea of user-based consensus and verification of the data ledger is here to stay. Its potential is beginning to be harvested by major economic institutions such as stock markets. At the end, as Mike Hearn states in his blog: ‘Bitcoin’s current design is unlikely to be the last word in making payments. It’s reasonable to think that one day it will be outcompeted or augmented by something else.’


Pic. https://bitcoinxt.software/

Why is Bitcoin forking? (Mike Hearn blog)
July 2015 chain forks (BitcoinWiki)


Subscribe to Diplo's Blog