Centralized vs Decentralized Networks
By Vlad Costea — External Contributor to Trezor Blog
Decentralized networks are your protection against Big Brother. Unlike their centralized and easily corruptible counterparts, they champion individual freedom through censorship resistance. Because they don’t route communications through centralized servers, there is no way for one individual to control what you send over the network.
In the cryptocurrency space, decentralization is perhaps the most fundamental prerequisite. No project can justify its existence without at least enabling universal participation and eliminating single points of failure. After all, why bother trying to provide alternatives to established companies and their services if your network is just as easy to control and shut down by authorities?
This article seeks to provide some nuance in the whole debate about cryptocurrency network decentralization. Today, there are thousands of projects that claim to decentralize money or deliver uncensorable smart contracts and applications. Some of them are also focused on privacy, but their utility is complementary to other use cases. Yet addressing the staggering amount of networks now in existence requires a series of instrumental criteria that can help anyone spot a decentralized project in a crowd of pseudo-decentralized (or arguably centralized) ones.
Our benchmark for decentralization is Bitcoin, whose network has grown exponentially and organically over the years. By all metrics, it is the most persistent, resilient, and globally-distributed project ever. Since January 3rd 2009, the network has only been down for 14 hours — thus having an uptime of 99.98%. And, since the same launch, the basic rules about scarcity and lowering currency emission rate have remained the same.
In comparison to Bitcoin, all other projects have their tradeoffs — sometimes they offer interesting features but they come at the cost of lower security and resilience to political intervention. Nonetheless, it’s still important to dive into details and spread information about the main criteria to assess the quality of decentralization.
Why Does Decentralization Matter?
Why does anyone bother to create decentralized networks when we already have companies that deliver perfectly functional and user-friendly centralized products? What’s the point of waiting 10 minutes until the next Bitcoin block to get a confirmation when you can simply pay via PayPal and have instant global settlements?
Well, decentralization is mostly about censorship resistance and individual autonomy. While PayPal and banks can get political and give in to pressure from governments and NGOs, a network like Bitcoin’s is completely neutral and inclusive. As long as participants respect the consensus rules (run a non-contentious software), there is nobody who can block or censor them regardless of their reasons.
The argument for decentralization also extends towards the governments themselves. While we regard more or less fairly-elected officials as legitimate representatives of their peoples’ will, it isn’t always so. When corruption happens to rule over the land and all the financial instruments are under the control of the corrupt elites, breaking the law to preserve individual human rights becomes a moral duty.
In George Orwell’s “Nineteen Eighty-Four”, Winston Smith and his lover Julia need to trust Mr. Charrington with their money and O’Brien with their private data. The fact that they get dehumanized and tortured under a totalitarian regime does not justify or legitimize their treatment. On the contrary, their condition could only be improved by truly decentralized networks with nodes outside London and the entire Oceania.
The Case of Bitcoin — The King of Decentralization
If you want to understand what a decentralized network implies, look no further than the example of Bitcoin:
- anyone can become a node in the network without requiring a third party’s permission;
- nodes are able to individually verify the entire history of transactions and also participate in the process of securing the network;
- nodes are geographically and politically distributed in a way that prevents single points of technical failure and political censorship;
- all the software running on the network is open source, transparent, and easily verifiable by anybody;
- the distribution of monetary assets doesn’t generate a lot of inequality and market manipulation;
- launching an attack is costly and requires coordination from many nodes;
- there is no leader or centralizing figure to dictate development or protocol changes, it’s up to network participants to make decisions for themselves according to their own incentives.
While some metrics can be improved (for example, it would currently take 4 mining pools to control more than 50% of the mining power and potentially launch reorg attack on the latest state of the blockchain), Bitcoin is still an excellent example of decentralization. At any one time, you can find approximately 10,000 nodes that verify and validate each ongoing transaction.
Thanks to the preserved small block size, it is still possible to become a node on the network using old or underpowered computers — for about $100 you can get a Raspberry Pi and an affordable 1TB hard drive. This means that a great percentage of global citizens can afford to contribute to the decentralization of Bitcoin, while also enjoying the privacy and sovereignty benefits of full nodes.
Bitcoin is also king in terms of monetary decentralization, but could still see improvements in the coming years as adoption grows: as of June 2020, 14.46% of the supply is held by the top 100 addresses (but these addresses also include exchanges that hold user funds).
Ethereum — Not Decentralized Enough
Now let’s look at Ethereum, the world’s second-biggest blockchain network. At a glance, one can observe that it borrows some characteristics from Bitcoin: it’s permissionless, verifiable, censorship-resistant, open-source, and somewhat fairly-distributed.
It more or less ticks those boxes, but fails in a very important department: the known funders are central points of failure, and a conference call between them could be enough to stir development in a certain direction and make radical changes to the plans. While some fans will find virtue in benevolent dictatorship, it’s easy to imagine how governments can pressure the founders to change the protocol and even roll back the blockchain. Ethereum set an important early precedent for breaking immutability in 2016, and the fact that such an intervention can be done again is a bad sign.
Furthermore, running a full archival node on the Ethereum blockchain is resource-intensive and can’t even be done on a high-end laptop from 2019. According to BlockCypher data, not even the Ethereum Foundation keeps a full archival copy of the activity that took place on the Ethereum blockchain since launch. This can be dangerous because smart contracts and applications can simply disappear if they’re not supported by a node. If running this node requires industrial-level hardware, the decentralization premise is eroded.
For a blockchain whose ambitions extend to “DeFi” (decentralized finance) and dApps (decentralized applications), Ethereum is just not decentralized enough. And if you take into consideration the 72 million ETH premine (the developers held 12 million, while 60 million coins were sold in the ICO) and the fact that 34% of the supply is held in top 100 wallets, it’s easy to understand why the project receives so much criticism.
As a network, Ethereum is definitely not decentralized enough and it’s hard to imagine how some issues will get fixed without compromising the history and status-quo.
Other Altcoins — Are They Decentralized?
So if the second biggest network in the space has big issues with its claims to decentralization, then what about the projects that emerge and claim that they’re better than Bitcoin? Or what about the ones that aim to provide privacy or smart contracts without claiming that they also build sound money? Well, they may be excellent at one particular task but fall short in most other aspects.
On one hand, we have Bitcoin clones such as Litecoin, Dogecoin, Zcash, Dash, and Bitcoin Cash. From a computer science perspective, they are carbon copies of Satoshi Nakamoto’s original design and only feature a few notable changes. Unlike Bitcoin, they all have known active leaders and sometimes foundations that exert power and influence over development. Furthermore, they don’t have the same network effect as Bitcoin, so they benefit from the participation of fewer nodes. This also means that the underlying networks are less secure and easier for malevolent actors to attack .
Some of these altcoins will claim that they are faster than Bitcoin, but they don’t provide the same amount of security (which ensures transaction finality). On the Bitcoin network, it’s recommended to wait for 6 confirmations before declaring that a transaction is irreversible. To achieve the same kind of security on Ethereum, it takes 653 confirmations (2 hours, 25 minutes). Whenever you find yourself in doubt, check out howmanyconfs.com and do your own math for transaction finality.
On the other hand, there are the Ethereum clones: Tron, EOS, Tezos, NEO, and Cardano. While they all seek to increase scalability and speed, they fall flat in terms of decentralization. Comparatively, Ethereum still has more nodes and a greater network effect which guarantees better censorship resistance. But, in the aftermath of the DAO hack which created a precedent for breaking immutability, it’s unlikely that any Ethereum transaction remains final. The clones shouldn’t be considered more resilient either, as they are led by known leaders who act as potential central points of failure.
In between, projects such as Monero and Grin offer an interesting balance between decentralization and privacy. They are permissionless global networks which create interesting ways to preserve the anonymity of online financial transactions, but they still have fewer users and nodes than Bitcoin and Ethereum. Nonetheless, they are excellent and original pieces of open-source software that may be used in the future to boost Bitcoin’s privacy via Lightning Network atomic swaps or sidechains.
Decentralization, Sovereignty, and Personal Responsibility
Decentralized networks are some of the most significant discoveries of our time, as they act as protection shields for individuality. Their two greatest merits include holding centralized services accountable and competitive, and limiting the power of governments.
Yet users of decentralized networks also renounce certain conveniences. Sovereignty, defined as absolute independence from systems that find themselves under political or corporate control, entails caution and personal responsibility. There is no third party to protect you from your own mistakes when you’re using Bitcoin and if your funds get stolen, you can’t call a support desk to ask for a chargeback.
Under the “not your keys, not your coins” spirit of decentralization, you expose yourself to risks that don’t exist in the world of centralized services. Unfortunately, this part of sovereignty is intimidating to some people and keeps them bound to trusted third parties. Yet personal freedom is a worthy cause and we should learn to be less reliant on custodians.
With Trezor hardware wallets, you can be your own bank and carry your funds in your pocket with both security and plausible deniability. The PIN number, passphrase, recovery seed, multisig setup and Shamir backup protect your coins from malevolent actors. Conversely, the intuitive user interface, great compatibility with computers, and wallet integrations provide flexibility and true autonomy. Native support for the most significant coins guarantees that you can send and receive payments in any currency that helps you bypass Big Brother.
Trezor devices are designed to be unintimidating facilitators of sovereignty. With them, every Winston and Julia around the world can enjoy the benefits of decentralization without the need to take special classes in cryptography, computer science, and economics. You just need to know how to use a computer and take a good look behind you before typing in passwords or PINs. Also, thanks to the 100% transparent design, security researchers can contribute to the hardening of Trezors by disclosing vulnerabilities and suggesting fixes. Every day, hackers run out of more attack vectors, as the hardware wallets become more robust.
Decentralized networks aren’t easy to analyse and understand but at least their underlying assets can be stored in a provably secure way. When in doubt, always remember to compare the project with Bitcoin and remember that organically growing the number of users and nodes takes time. And, whenever you’re in doubt, check statistical resources such as arewedecentralizedyet.com and bitinfocharts.com.
