What Decentralization Is, What It's Not, and Why It Matters

September 28, 2021

It’s George again, with the latest on everything you need to know about decentralization - and, more importantly, the many imposters that often go by its name.

Right now, there are many heated debates on decentralization, and I’d like to take a moment to provide some clarity to all of the people following Offshift and my previous blog posts. All of you are deserving (and capable) of possessing the necessary knowledge to understand who is giving you the facts and who is trying to fool you for the purposes of furthering their own personal agendas.

As always, let’s break this down to the basics, and proceed point-by-point:

So, what is the key difference between centralization and decentralization? The answer lies in the meaning of trust.

In a centralized system – for example your bank network – there is a trusted authority acting as a validator that determines whether your transaction is valid. These authorities withdraw money from a sender’s account and deposit it to that of a receiver. Governments trust them (lol) and you do as well. They are in a higher position of power relative to you: they can freeze your money or lock you out of the primary system of exchange, and it’s on you to show proof that they are wrong. As should be painfully clear, they are the ones in power - not you.

They also have duties (well, on paper anyway), and are responsible for issuing a refund if something goes wrong, like if one of your cards gets stolen. In theory, trusted authorities are supposed to be held accountable for their mistakes. Obviously, there are a lot of arguments and angles to be addressed, but the defining element of a centralized system is that it requires you to trust a single entity with a great deal of your livelihood.

In this instance, I use the bank network as an example because it’s very familiar for most people, and it is easily comparable with the blockchain-based systems we are building in the crypto space. But there are countless other centralized ecosystems we use and rely on in our everyday lives.

Let’s get to decentralized systems: here, the key piece (trust) is taken from the opposite angle, that is, the goal is to achieve a trustless ecosystem.

But what’s the meaning of trustless, anyway? Trustless means that the system works without the need to trust any individual or entity.

So how do you achieve consensus amongst participants in such a system? How does everyone come to agreement on what has to be done, and which rules to follow?

This is where blockchains come into play. To join a blockchain-based ecosystem, a predetermined set of rules exists that you must agree to in order to join the network with your node, be it a validator or just a relay.

At this point, you might be less than pleased, seeing as you have to agree to a predetermined set of rules when interacting with banks as well, and at least with banks you can avoid ludicrous gas fees and even get a refund for mistakes and fraudulent charges.

Unfortunately, in a lot of decentralized systems users often have to purchase miscellaneous hardware and develop an intimate understanding of the technology; whereas in conventional, centralized systems, the user experience is seamless and rarely labor-intensive.

A major challenge for those of us that are building and pioneering decentralized systems is convincing users to migrate and adopt our platforms.

What makes blockchains special is that they provide mathematical proof that stored data cannot be tampered with. There are no signatures from the executive board, politicians, or CEOs; it’s just plain mathematics.

Most of you, regardless of your field of expertise, will recognize the importance of being certain that the data sets you read and submit are tamper-proof (this topic deserves a dedicated article in and of itself).

In a blockchain-based system, you agree to the established set of rules directly. You choose to support or reject hard forks with your relays or validators, and you can contribute to code or submit improvement proposals. And the proposals, by the way, are evaluated by peers - not a centralized committee or consortium.

In centralized systems, your agreement to the set of rules that determine your state’s laws is indirect. Once every X number of years, you cast your vote for the politician you believe fits your beliefs the most, and then you trust him or her to make decisions according to your purportedly shared beliefs. But you can’t just become a politician and let everyone in the country know about your beliefs; whereas in crypto, you can dive into the most controversial issues with pull requests, improvement proposals, and other mechanisms of meritocratic participation right away.

But wait, it’s not all black and white - decentralization has shades too! In fact, although you might not realize it, our society already employs some degree of decentralized services.

Let’s take the government for example: you have a legislative branch in Congress, an executive branch in the president and his cohorts, and a judicial branch in the Supreme Court. Which one rules over the others? Just as in rock, paper, scissors, there is a series of checks and balances. These checks and balances help to ensure that no individual branch can accumulate enough power to rule over the people in tyrannical fashion - essentially the IRL version of preventing a 51% attack.

We have looked deeply into the foundation of systems founded on centralization, decentralization, and some form of a hybrid (a decentralized system with centralized actors). Now it is time for an experiment:

Let’s try to stress-test each system by performing attacks in a theoretical scenario. Imagine that a group of attackers intend to take a system to its knees.

First off, let’s set the scope of their attack: they want to tamper with data. If they can compromise the integrity of otherwise trustworthy data that is used to decide important matters, the attackers will be able to decimate any organization which relies on that data for decision making.

In a centralized system, trusted parties act as validators by giving their word that data is to be trusted, providing a pre-established list of rules for participants to follow, and laying out punishments in the event the rules are broken - only so as to ensure the safe operation of the system and to protect its participants, of course (lol).

Anyway, the aforementioned attack would surely be crafted around the weaknesses of one or a few trusted validators. For example, the attackers could bribe or blackmail them, which has happened an awful lot in recent geopolitical history. Once attackers get a foothold in a centralized system, it becomes very easy to tamper with data and rig the system to their advantage.

So as you can see, a centralized system is only as resilient as the weakest member of its trusted party. To me, this isn’t good enough to sleep tight.

So what about soft decentralization then? It’s far more resilient, but still has the same issue: if attackers understand the particular system of checks and balances, they can compromise trusted parties and convince them to collude. In the case of the United States government, 2 out of 3 is enough to do the job.

Here’s the bottom line: these entities, balanced as they may be, are still composed of and ruled by humans, and therefore remain vulnerable to human weaknesses.

Now, let’s examine a fully decentralized system. As you might have guessed by the arguments presented above, decentralization fixes what is an otherwise unfixable problem. Bribery and blackmail are not effective, and government bans only strengthen a decentralized system by conferring attention to its competitive advantages. Bitcoin miners, for example, always have the monetary incentive to behave as good actors, and if attackers were to spend enough money to compromise the majority of nodes and tamper with data, the entire system would become worthless and useless.

In a decentralized system, you don’t have to trust anyone, and you can participate in the network freely; but there’s one major advantage that does not receive the attention it deserves: censorship-resistance.

This is not to say that you can make derogatory remarks about China’s president without consequences; it just means that there’s no way (again, it’s based on math, not a promise from a third party) to selectively deny access to the network because there is no single person or entity that can be bribed or blackmailed. For every miner willing to accept a bribe, there will be several unwilling miners.

BUT, BUT, BUT - we must continue to stick to the facts. And on that note, where we now stand regarding cryptography is concerning, to say the least. We are actually regressing and reinstating problems that created the need for blockchain architecture in the first place.

Bitcoin’s Lightning Network is a perfect example. Sure, people want BTC, and they want to make transactions quickly and cheaply. But here’s the thing: expensive and inefficient mining processes are exactly what makes the network secure in the first place. If you aren’t going to leverage the system’s advantage of decentralization, you might as well revert back to servers with simple nodes - they’re even cheaper and more efficient than Lightning.

What I’m saying is, the benefits of decentralization are effectively removed by off-chain scaling solutions such as Lighting Network, which centralize control over user data and create single points of failure that are vulnerable to pressure from regulators and other government agencies.

And by the way, Ethereum is no different. When I see Layer 2s arguing with each other about who is the most decentralized, I don’t know whether I should laugh or cry. Carrying out anything of major importance, especially main-chain transactions, off-chain is not what blockchain was about, and certainly not what it should look like today.

But wait - shades of decentralization have shades too… it’s a decentralinception!

A remarkable mention on this matter goes to Matter Labs, the organization that maintains zkSync. Their values are very similar to those we hold close at Offshift. Is zkSync an L2? Yes, but not the kind you are used to.

zkSync is committed to user privacy and adoption; they propose an L2 where validators have to post proofs on-chain and have no room to act maliciously. They just execute calculations.

Is it as decentralized as Ethereum? No. Is it set up to scale in the direction of Ethereum? YES, and that’s what matters: all of their code is open source and free to use - no hidden contracts that require purchasing a license, no endorsements. And they have tutorials and guides everywhere because they want users and developers who utilize their tech to remain private. They don’t ask for money, and most importantly, they haven’t set up their environment in a way where it will be dependent on their team in the future.

Now, their audience is different from our own at Offshift, and seeing as our platform will live on L1, we won’t be able to use zkSync. My words here are nothing but an honest compliment towards a team whose values, conduct, and actions I respect.

I want to quote this bit from zkSync’s ”Mission, Values, and Culture” page :

There are an infinite amount of ideas that are worth fighting for. And there are infinite ways people can use blockchains to create systems that uphold and defend such ideas. At Matter Labs, we have chosen to center our mission and work around freedom.

In a world of fiat moneys, adversaries can exert powers destructive to freedom in more subtle and unrestrained ways than ever before in the history of other mediums of exchange. To preserve and extend human freedom and prosperity, Matter Labs is accelerating the mass adoption of crypto to give everyone in the world the choice of financial sovereignty.

zkSync belongs to you. Utilize it, govern it, own it.

As team members of cryptocurrency projects that stand against tyranny and centralized control, we have chosen what is worth fighting for. Sure, it’s about the money, but it’s not just about the money.

As the user, you have a crucial role - one that is far more important than you might realize. You are the only one who can push project teams toward a fair(er) world supported by decentralized systems.

Speak up and use the platforms and cryptocurrencies you believe in - the ones that are aligned with your principles.

What’s landed us in this dilemma of centralized blockchain architecture has indeed been the uninformed user’s complaint:

It’s too hard, it’s not cheap enough, it’s not instant, and it’s unbearable to be held accountable for misplacing my private keys and losing my coins.

On the surface, all of the above complaints appear to be fair game. But think about it: you already have all the comforts of a convenient user experience in conventional systems, but it’s still not enough. Considering deeply what you have read in this article, would you be investing your time and energy in crypto if the conventional systems you use and participate in were serving you properly?

If you want to contribute to a world that is truly fair, open, and free, help us (crypto) make it: do your part, stand for what matters.

About Offshift

Offshift is leading private decentralized finance (PriFi) with the world’s first Private Derivatives Platform. It leverages zero-knowledge (zk) proofs and sources reliable, real-time price feeds from Chainlink’s decentralized oracle network to enable users to mint zkAssets, an unprecedented line of fully private synthetics. Offshift’s mostly anonymous team has developed a trusted reputation for their thorough privacy research, development and execution.

To learn more and get involved, visit the links below: