Headlines Can Deceive
News headlines can be like movie trailers. Too short to give an accurate picture of the full story—designed to grab attention and often deceptive. This article will explain in layman’s terms the issues surrounding one of the hottest Bitcoin * topics of late, known as “the maximum block size debate”. There is a lot of confusion swirling around the media. For those lacking insight into the mechanics of Bitcoin, simply scanning headlines can leave a misinformed impression. Mike Hearn, a former core developer, recently created major commotion with his blog post titled, The resolution of the Bitcoin experiment . His post was followed by a prominent article in the New York Times, sounding a premature death knell for Bitcoin .
Relying on headlines can be misleading, no matter the topic. Many may avoid reading articles about Bitcoin because of the technical jargon and complexity involved to sort fact from fiction—or simply allow their opinions to be shaped by sensational claims. Fortunately, understanding the underlying technical issues in this debate is fairly simple from a high-level perspective. It’s the human element that’s more complex.
This blog post will first set the stage with a quick review of the supporting cast before introducing the main characters, the software developers. After setting the scene and reviewing a few basic concepts, it will be possible to sort out which issues are purely technical and which are political in nature. Forecasting the technical problems that may arise if the number of transactions exceed the present-day capacity of the Bitcoin network is straightforward. Determining how the various players will respond is a more complex consideration, especially given the open source nature of Bitcoin. A few key scenarios will be examined in light of this lively debate.
Players on the Bitcoin Stage
Here is a quick review of the cast of characters in the Bitcoin story, and the roles they play in this discussion:
- Users own bitcoins and save and/or transact with them using software and hardware digital wallets. The blockchain is a fully transparent ledger that maintains the history of all bitcoin transactions. No one knows the exact number of bitcoin users but millions of wallets exist.
- Nodes will refer to full nodes which each maintain a complete copy of the blockchain, verify and transmit transactions, and broadcast new valid blocks from miners to other nodes. The number of full nodes is about five thousand.
- Miners are an essential and special type of full node. They aggregate transactions into blocks. Valid blocks can only be created by solving a difficult computational puzzle using specialized high-tech hardware. Finding the next valid block is a random process, but the more computing power a miner can control, the greater his chances of success. Today, a miner who finds the next valid block is rewarded with 25 bitcoins and the fees associated with the transactions within the block.
- Startups include new crypto-currency ventures such as bitcoin exchanges, digital wallets, merchant payment services, remittance providers, ASIC designers, etc. There are hundreds of startups in the Bitcoin ecosystem and over a billion dollars of venture capital invested .
- Developers are the people doing the hard work of maintaining and improving the open source Bitcoin software. Much of the recent drama centers on disputes between developers and how to safely scale the Bitcoin network. Some of these characters are key players in the current scene, and will soon “enter, stage right”.
The “Bigger Isn’t Always Better” Scene
Bitcoin is a multi-facetted innovation, and other sides of it will be examined later in this essay. However, the apocryphal creator, Satoshi Nakamoto, clearly viewed Bitcoin as a system for making P2P electronic cash payments, and as such, the number of transactions that can be processed in a given timeframe is a key system parameter . The system is designed to create a new block every 10 minutes on average. A block contains recent transactions but is presently limited to 1 MB in size. Each new block is then appended to the blockchain, which then becomes the immutable public record of bitcoin transactions. The complete and ever-growing blockchain is stored and monitored by participants known as full nodes.
To help secure the system, the 1 MB upper limit was hard-coded into an early version of the Bitcoin software released in the second half of 2010 . This 1 MB block size effectively corresponds to a theoretical maximum of 7 transactions per second (TPS)—think of this as a speed limit on the Bitcoin system as a whole. Note that PayPal is able to process more than 100 TPS and major credit cards can handle orders of magnitude more than PayPal . Clearly, there are many obstacles for Bitcoin to overcome before it could compete with the incumbent payment systems.
Is a bigger block size really better? The 1 MB block size has helped secure the Bitcoin network, preventing it from being spammed with junk. This relatively small size limits the growth rate of the blockchain, thereby making it viable for standard computer hardware to run and maintain a full node. Today a computer needs about 50 GB to store the entire blockchain which is growing at about 3 GB per month . It is easy to see that if the block size were increased substantially, then fewer full nodes would exist because of greater hardware requirements. Also, larger blocks are going to take longer to propagate for fixed Internet connection speeds, possibly affecting transaction reliability. Increasing the block size could therefore threaten the censorship resistance of Bitcoin and damage its attractiveness as a platform for open innovation.
As adoption continues to increase, transaction volume is likely to overwhelm the network’s capacity in the near future. This is arguably a good problem, since it indicates Bitcoin usage is growing and the system is succeeding. However, it should be intuitively obvious that, at the very least, the performance of the Bitcoin network will degrade, perhaps substantially. What do the experts think, and what insight can they grant?
Lead Actors in the Bitcoin Drama
Let’s consult the opinions of four acknowledged experts: Gavin Andresen,
Mike Hearn, Pieter Wuille, and Jeff Garzik—all developers playing significant roles in the unfolding drama. As a quick characterization: Gavin is the doting father of the leaderless open source Bitcoin project; Mike is the impatient adolescent who just had to leave home because he couldn’t get his way; Pieter is the level-headed technologist seeking new workarounds; and Jeff Garzik is a big picture guy and conciliatory free-market advocate. (I warned you, dear reader, that the politics were more complex than the technology…hence all the drama!) Gavin Andresen, who took the reins from Satoshi Nakamoto as lead Bitcoin developer, raises the following concerns if the block size is not increased above 1 MB soon :
- More time to confirm transactions with a set fee, I.e., slower payments.
- Extremely low fee transactions completely excluded from blocks.
- Transaction fees will necessarily need to rise.
- Higher fees will deter some users and prohibit some applications.
- Adoption and growth of Bitcoin will be subdued.
Clearly, without an increase in the block size, users will need to pay higher fees on average to incentivize miners to include their transactions in the next block.
Will Our Story Include A Train Wreck?
The competition for fees could certainly leave some users behind. But could the outcome be even more severe? Mike Hearn thinks so. He predicts a positive feedback loop on fees that will increase out of control, a bloated backlog of transactions, increased risk of double-spending, and crashing nodes—a real train wreck! In Mike’s mind, not increasing the block size signals the end of Bitcoin . “Exit, stage left”.
On a more moderate note, developer Pieter Wuille acknowledges Gavin’s concerns cited above and agrees that, in the long term, Bitcoin will have to evolve or be left behind. The block size will likely have to increase, someday. Wuille, however, is fine with the use cases of Bitcoin changing to fit its capacities, and does not feel any urgency to scale the network by arbitrarily increasing the block size . Wuille, a master technologist, has an ace up his sleeve. He recently proposed a concept called Segregated Witness that is currently being tested. Among other things, SegWit allows more transactions to fit inside a 1 MB block—about twice as many on average. In addition to SegWit, there are a couple of other technical developments that could immediately reduce pressure on Bitcoin’s transaction speed limit and postpone the urgency of increasing the block size. These include something known as the Lightning Network and sidechains, both being developed by Blockstream, which just announced $55 million in Series A venture funding . Pieter Wuille is part of the Blockstream team. Both the Lightning Network and sidechains will benefit if SegWit is adopted, and both offload transaction traffic from the underlying Bitcoin network .
To Fork, Or Not To Fork?
Before examining the role of our fourth protagonist, Jeff Garzik, there is another essential element to our plot, Bitcoin hard and soft forks. Pieter’s SegWit is a perfect segue way to understand this issue. SegWit can be implemented using a soft fork in the Bitcoin code. Soft forks have been used in the past to successfully upgrade Bitcoin. There still needs to be a high level of consensus amongst mining nodes, but soft forks do not affect everyday users and merchants who fail to upgrade to the latest software.
The crux of the scalability debate revolves around the need for a hard fork to increase the block size above the original hard limit of 1 MB. A hard fork requires a massive coordinated effort amongst all constituents to succeed and has never been accomplished. A hard fork poses systemic risk, and there is a real chance of losing money if a full node doesn’t upgrade its software in step with the network consensus . A hard fork mishap could create massive confusion and undermine confidence in the network, quickly crashing the price of bitcoin. Mike Hearn and Gavin Andresen attempted a hard fork with their Bitcoin-XT code released this past summer, jumping the block size up to 8 MB and intending to double it every two years. Bitcoin-XT never reached a critical mass with miners, and the 1 MB status quo prevailed . Hence, the rancor in the great Bitcoin debate. Because Mike didn’t get his way, he left in a huff and will be applying his skills in the banking sector, working on private blockchains.
There are now three main camps of developers: Bitcoin Core, Bitcoin Classic, and Bitcoin Unlimited—each proposing solutions to scaling the network, each with differing approaches. Gavin, fatherly steward he is, continues to contribute to the various camps that are all working to scale Bitcoin as user adoption continues to grow .
The Existential Question: What Is Bitcoin?
Jeff Garzik, our big picture guy and free-market advocate, is yet another Bitcoin developer worth heeding in the great block size debate. Garzik agrees that fees will increase as users bid to get their transactions processed, while the overall Bitcoin system becomes erratic and unreliable. But more importantly, his views help one bridge the technical vs. political gap and grasp the bigger picture.
Different stakeholders in the Bitcoin ecosystem have divergent visions for its future. This is one of the main sources of discrepancy in the urgency (or not) of raising the block size. What is the ultimate purpose of Bitcoin and the blockchain? Views span a vast spectrum. Should it remain a niche market for a scarce digital commodity, and a limited use P2P payment system for techies? Or can Bitcoin eventually scale to become a global payment system capable of displacing major credit cards?
Jeff thinks some who purchased bitcoin early on may be content with the status quo, believing their investment is best secured with a proven system. However, this may be short-sighted in that bitcoin will likely only increase in value if greater adoption is nurtured. The hard limit of 1 MB blocks will almost certainly become an obstacle at some point, and thereby suppress the long-term value of bitcoin. Miners, in general, are likely to be cautious about raising the block size too quickly. They don’t want Bitcoin growth stifled, but fear the reduction of their fee revenue and additional hardware costs associated with a blockchain growing too rapidly. On the other hand, some startups, with visions of competing with the likes of VISA or PayPal, want to see Bitcoin’s scalability expanded as rapidly as possible .
How to resolve the various forces pushing on the block size limit? Jeff argues that a few software developers should no longer be the ones determining Bitcoin’s scalability. He proposes a hard fork that will remove the static 1 MB limit. The block size will thenceforth be subject to market forces determined by the input of miners. Dependent on miner consensus, the block size could be doubled (or halved) on a stepwise basis over a designated timeframe .
Tragedy, Comedy, or Hollywood Happy Ending?
Some insight can be gleaned from a similar drama that provides context for the unfolding Bitcoin story. The Internet has evolved remarkably from its first implementation. In the early days there were many naysayers who viewed scaling the Internet as an unsurmountable obstacle. However, its open nature encouraged massive innovation on many fronts. Consistent advances in computer technology, increases in telecom bandwidth, and improved protocols have all contributed to the Internet’s success. Many experts forecast that within the next decade the entire global population will have access to the World Wide Web. The UN even acknowledges Internet access as a basic human right .
Perhaps Bitcoin is best viewed as a new fundamental layer of the Internet that transcends mere information transfer. Bitcoin is a new protocol layer that stores and transmits value in digital form without the need of a trusted third party such as a bank. Other layers may be built atop of Bitcoin that will solve its current scaling issues, such as the Lightning Network. In addition, creative actors will build new innovations that leverage the security and value layer of Bitcoin. For example, Factom is designing a scalable protocol to secure data records, e.g., land titles, using Bitcoin as its anchor layer . Many ventures are looking towards using the blockchain for asset trading and settlement solutions. These are just a couple of a vast array of possibilities.
There are many paths forward—from a hard fork to increase the block size, and soft fork solutions that work around current limitations. Consensus of stakeholders will be difficult to come by, but the many and multi-facetted resources and approaches being brought to bear will almost certainly find apath forward that increases the utility and value of the Bitcoin ecosystem in the long run.
How will our players behave and our story end? It’s way too soon to tell. We are only viewing scenes from the first act of the Bitcoin play. The script is being written in real time. There is no single playwright. Key characters and performers get to influence the plot and help write the unfolding drama. Along the way, there will certainly be elements of tragedy and comedy, unavoidable as they are in the human condition. Some players may come and go, others rise and fall. However, participants all get to play their parts in an open, collaborative production—aiming to compose a new future for finance and digital money—free of the constraints and potential abuses of so-called “trusted” third parties. What more motivation do the players need?
The show must go on!
*Bitcoin with a capital B is used for the protocol and system as a whole, whereas
bitcoin with a small b refers to the actual digital currency unit.
|ASIC||Application Specific Integrated Circuit|
|CPU||Central Processing Unit|
|GB||Gigabyte (1 billion bytes)|
|MB||Megabyte (1 million bytes)|
|TPS||Transactions Per Second|
1. Hearn, M. (2016, January 14). The resolution of the Bitcoin experiment. Retrieved from https://medium.com/@octskyward/the-resolution-of-the-bitcoin-experiment- dabb30201f7#.a8e72b9fu
2. Popper, N. (2016, January 14). New York Times. A Bitcoin Believer’s Crisis of Faith. Retrieved from http://www.nytimes.com/2016/01/17/business/dealbook/the-bitcoin- believer-who-gave-up.html?_r=0
3. Venture Scanner. (2015, September 4). The State of Bitcoin in Six Visuals. Retrieved from http://insights.venturescanner.com/tag/bitcoin-startup-list/
4. Nakamoto, S. (2008, December). Bitcoin: A Peer-to-Peer Electronic Cash System. Retrieved from https://bitcoin.org/bitcoin.pdf
5. About Scalability FAQ. (n.d.) BitcoinWiki. Retrieved from https://en.bitcoin.it/wiki/Scalability_FAQ#What_is_the_short_history_of_the_block_size_limit.3F
6. About Scalability. (n.d.) BitcoinWiki. Retrieved from https://en.bitcoin.it/wiki/Scalability
7. Session 3 – Network Scalability, University of Nicosia, MIS-536 Digital Currency Information Systems and Resources
8. Andresen, G. (2015, June 25). GitHub. Implementation of BIP 101: maximum block size increase. Retrieved from https://github.com/bitcoin/bitcoin/pull/6341
9. Hearn, M. (2015, May 7). Crash landing, What not to do. Retrieved from https://medium.com/@octskyward/crash-landing-f5cc19908e32#.f5massluy
10. Wuille, P. (2015, June 26). The need for larger blocks. Retrieved from http://lists.linuxfoundation.org/pipermail/bitcoin-dev/2015-June/009098.html
11. Blockstream website. (n.d.) Retrieved from https://www.blockstream.com/
12. Levine, A. (2016, January 9). Let’s Talk Bitcoin! #277 Separating Signatures with Segregated Witness. Retrieved from https://prezi.com/lyghixkrguao/segregated- witness-and-deploying-it-for-bitcoin/
13. Bitcoin Core website. (2016, January 7). Statement from Bitcoin Core. Retrieved from https://bitcoincore.org/en/2016/01/07/statement/
14. Palmer, D. (2016, January 11). CoinDesk. Scalability Debate Continues As Bitcoin XT Proposal Stalls. Retrieved from http://www.coindesk.com/scalability-debate- bitcoin-xt-proposal-stalls/15. Andresen, G. (January 2016). Classic? Unlimited? XT? Core? Retrieved from http://gavinandresen.ninja/classic-unlimited-xt-core
16. Garzik, J. (n.d.). Making Decentralized Economic Policy, BIP 100 – Theory and
Discussion v0.8.1 – draft. Retrieved from http://gtf.org/garzik/bitcoin/BIP100- blocksizechangeproposal.pdf
17. About Right to Internet Access. (n.d.) Wikipedia. Retrieved from https://en.wikipedia.org/wiki/Right_to_Internet_access
18. Factom website. (n.d.) Retrieved from http://www.factom.com/#home_section