Vitalik Buterin: Ethereum, Decentralization and Blockchain

Vitalik Buterin's IQ is an astounding 257. This was tested, this was verified. he was born in Russia in 1994 and his favorite toy as a child was Excel. Vitalik was able to speak fluent Mandarin Chinese within 3 months without even trying. While studying programming in Canada he came across Bitcoin.

He saw the potential before anybody else did and it paid off, he made $1,257,294 off of his Bitcoin investment, now he travels the world driving his car collection and does anything he wants. If that wasn't enough at 19 he created a new highly ambitious technology called Ethereum. Ethereum can only be described as the next step in the evolution of society and today we have a unique opportunity to make hundreds of thousands or even millions of dollars.

Imagine a future where much of the economy is controlled by decentralized organizations - you step into your self-driving car and you pay $20 per ride. Just from your car communicating with the other computers on the road. Ethereum is already being used by Goldman Sachs, AirBnb, Visa, Amazon and Tesla. We are still in the ages of infancy and because of the rapid uncontrollable growth for Ethereum and investors have been jumping in.

Chapter 1: Who is Vitalik Buterin?

Vitalik Buterin is the star child, he's a boy genius that learned Chinese in a few months at 6 or 7 years old, so it's a fascinating story. We have a hero already because Satoshi Nakamoto is this myth, he disappeared into the ether and no one knows who created Bitcoin, so it's impossible to know who's really behind this, but Ethereum has this very different appeal, a skinny, Russian kid named Vitalik Buterin who lives in the United States now and is the visionary only now in his early 20's.

One of the most buccaneering and fastest growing cryptocurrency, the number two in market capitalization is Ethereum and the interesting position it's taking rather than being another coin is the attempt to create a platform for other coins to launch off them. This is a cunning move, similar to Microsoft, instead of being the computer or application, they were the operating system for which all of the applications were run off.

That is why Ethereum has a lot of support and climbed very strongly quickly. Although Bitcoin paved the way. It had to deal with all of the first arising issues and skepticism, now that cryptocurrencies are being more widely accepted, it's allowed for really quick movement in coins that have a real interesting resonance.

Chapter 2: What is Ethereum?

Thanks to the power of modern communication we have the ability to create technologies that are decentralized, removing middle-men and allowing users to interact with each other directly through a global network.

Decentralized applications have been becoming more and more important in the past 10 years and have the benefits of massively reducing costs and barriers of entry, removing single points of failure, preventing censorship and ensuring transparency and trust between all of the parties in an interaction. BitTorrent, a file-sharing network developed in the early 2000s is arguably the first decentralized application to have been created.

BitTorrent allows anyone to share any kind of file to anyone else in the world, allowing people to distribute content quickly and easily even if they do not have the resources to pay for their own website or server. 5 years later, Satoshi Nakamoto came up with a blockchain, a sort of distributed database and used it to create Bitcoin, the worlds first decentralized currency. Decentralized currencies like Bitcoin allow people to send money instantly, anywhere around the world with no regard for national borders with negligible fees. Bitcoin is increasingly being used therefor for international remittances, micropayments and commerce online.

Decentralized applications for finance, cloud computing, messaging is distributed governance soon to come. Ethereum is a platform that is specifically designed for people to build these kinds of decentralized applications or DAPPS for short. The Ethereum client which we are calling the ETHERBROWSER will include the ability and peer-to-peer network for sending messages and a generalized blockchain with a build-in programming language. Allowing people to use the blockchain for any kind of decentralized application that they want to create.

Ethereum can be used to build financial applications that are fully trustworthy and transparent because they run on the blockchain. Online cryptographically secure systems for managing your property and contracts, social networking and messaging systems that allow users to maintain control of their own data, systems for trading underutilized computational resources like CPU time and hard-drive space and eventually tools for online voting and distributed governance. The most exciting applications for Ethereum are probably the ones that we have not even though of. As with all new platforms for innovation, like the protocols that underlie the Internet itself, it is not always easy to predict what they are going to be used for. Gmail, Facebook, Twitter, Instagram and the modern internet as a whole are all a result of early developments in the world-wide web and JavaScript, the programming language of the world-wide web from the 1990's.

Similarly, by providing a universal programmable blockchain and packaging it up into a client that anyone can use, the Ethereum project hopes to do the same for finance, peer-to-peer commerce, distributed governance and human collaboration as a whole. The question is, what will you build on top of Ethereum?

Chapter 3: The Ethereum Blockchain Ethereum was built to be a sort of new and improved version of Bitcoin and to do things that Bitcoin couldn't do. The original aim of Ethereum was to create not just a decentralized currency, but a decentralized global computer that could do computational problems. Ethereum has its own currency, Ether, the Ether though are only there to serve as only gas for the computations that are going on this network of computers that are running Ethereum.

  • July 30, 2015: Release date of the first live version of Ethereum.
  • $160 million: Peak value of Ether raised by the Decentralized Autonomous Organization (DAO).
  • $50 million: Value of Ether stolen in a back on the DAO.

"The original reason why I liked the name is Ether is there was sort of thing that scientist were hypothesizing about in the 19th century where people were thinking, sound waves travel through air, water waves travel through water wells, what do light waves travel through? They hypothesized this existence of a medium that kind of permeates the entire universe that in that of itself is kind of invisible but it provides this global reference frame and it is in some sense a sort of base layer that the universe runs on. I actually felt that was kind of a fitting analogy for what I wanted to do." – Vitalik Buterin "Bitcoin has the wrong approach. I would say Ethereum boasts features and opportunities to do things Bitcoin doesn't. It's like saying that a telephone can beat an orange." – Vitalik Buterin Ethereum is a decentralized application platform. Its apparent Bitcoin is a more of a platform for transmission and storage of value, so Bitcoin was essentially an experience in monetary theory gone wildly successful. You can think of Ethereum as a decentralized world-wide web. It will eventually be able to run all of the different kinds of applications that we see on the internet.

At the end of the day, it's a sort of fabric that what you could almost describe as being a new internet infrastructure. Some would argue you could run a Facebook-like application, the difference that it would be peer-to-peer, decentralized rather than a client server type architecture.

Ownership of data, ownership of communications or informational emissions will rest with the users of the system, rather than the owners of the company like Facebook. "I think Vitalik knows that he's building a giant, globally distributed, a reprogrammable supercomputer on a scale that mankind has never seen before and he's doing it through a mechanism which initially supports this concept of smart-contracts and issues a token just like Bitcoin so there's value being created by mining. The biggest difference is that Ethereum produces mines that are reprogrammable to tasks that we don't necessarily do yet." – Sam Garlinghouse, Billionaire Investor at

One of the highest ambitions of Ethereum from the beginning was that you would actually be able to program the functioning of entire corporations into this global computer and you would have not just a decentralized currency, but a decentralized company that would run according to the rules that you programmed into this network.

There was always this idea that Ethereum might allow what people call, "Decentralized Autonomous Organizations, or DAO for short. Decentralized Autonomous Organizations is this idea that you could have these kinds of anonymous entities that are written in code that are set on a blockchain that are governed through, "smart-contracts," it could be some voting mechanism, it could even be some kind of voting by prediction markets or some other new thing. It would be a sort of company that would be run, not according to human decision making or human roles, but according to rules that you programmed into the Ethereum network.

You could actually leave the sort of decision making up to an artificial intelligence. So if you had a board, you could say we can't come to a consensus, let's just refer this to an artificial intelligence and everyone just delegates their vote to the robot and the robot just makes their decision for them. Eventually, you can imagine that many boards are just run by AI and occasionally intervened by humans. DAO attracted somewhere around $150 million dollars in the course of a month, basically, on the first days it was operational, you had this attacker who came in and takes advantage of the contracts in a way that basically siphoned half of the money into an account under the control of the attacker.

The one saving grace in this was that the money that was siphoned off, because of the way the DAO was written, couldn't be used for 30 days and so began this countdown to try to figure out how to solve the attacker from getting the control to the $60 million dollars.

There was a time where we had to figure out if we're going to do something about it or just let it go. It set off this pretty really philosophical debate about how to deal with a problem like this in the realm of smart-contracts and virtual contracts. One of the great attractions of a blockchain technology, in Bitcoin, in Ethereum was that this record of transactions going back in time could never be changed.

That's one of the reasons why this was so valuable, you could look back and you know that a transaction happened and know that no one could tamper with that, no one could change it. Once you had the DAO, people wanted to go back and change these contracts so this person that seized all of this money, that those transactions wouldn't be valid. That struck at the very core of what Ethereum used to be. Chapter 4: Decentralization

  • The Advantages of Decentralization

I want to put some of these ideas into context and talk about different kinds of decentralization that we have and the ideas why decentralization is a good idea, why different kinds of decentralization make sense in some areas and why some different kinds of decentralization don't make sense in other areas. Then sort of see how these ideas and the benefits that you can get from these ideas make sense, particularly in the energy industry.

Decentralized in general is not a new idea. We've had human civilization for over 10,000 years, and over the last 10,000 years we've had lots of instances of centralized systems, centralized companies, centralized governments, centralized cities and we've had lots of instances of decentralized things. These are just some instances of decentralized applications in the software industry in the last 15 years:

  • BitTorrent
  • Blizzard Entertainment
  • FireChat
  • IPFS

It's important to point out that even before we had computers, even before we had IT, we had various instances of decentralized economies and decentralized ways of interacting. If you take just look at one particular example, just look at the insurance industry. Right now, we have large insurance companies and if you have insurance you generally have insurance either because it's some national government program or because you're buying a service from some insurance company that sells insurance as a product to very large numbers of people, but the way that the insurance industry started was actually because groups of people coming together and saying, "okay, we each have a ship and if any one of our ships sink or gets captured by a pirate, then we all agree to cover each other's costs." So decentralized models are something that's existed for quite a while.

Now if we look at the software industry, in particular, we've had BitTorrent, we've had decentralized networks for sending files from one computer to another. If you look at email, in some ways it's a semi-decentralized protocol, you can access email through GMAIL or through HOTMAIL, or some other email service. Theoretically, if you wanted too, you could set up your own server.

If you want to download some file, then BitTorrent is one great way to do that and it's, in fact, something that a lot of software developers use as a primary way of distributing data that they produce to their users. If you use open-source software packages generally the developers don't have the resources to actually set up and maintain the infrastructure to centrally upload and send their software to tens of millions of users and a lot of the time they even use decentralized networks as a way of solving the problem. A lot of these software companies actually encourage you toward using a decentralized network to download their applications, instead of downloading it directly from them.

That way users become part of your network. At the same time as they download their software, they help pass the software on to their friends. That's the model a lot of major gaming providers use to distribute patches to their users. It's a paradigm that's being increasingly used for messaging and for storing files. It's something that has been used for a while. Now in 2009, Satoshi Nakamoto came up with the idea behind Bitcoin. A peer-to-peer electronic cash system.

Computer scientists, cipher punks, lots of people have been interested in these ideas of decentralizing things for quite a while, but for the one major application that they really had a hard time with was actually money. They figured out how to do decentralized messaging, they figured out how to send files, but currency is this one application that for reasons that we're going to get into end up taking them quite a long time. Back in the 1980s and 1990s, there have been various forms of cryptographic e-money, but coming up with a version that doesn't rely on one central party to manage the entire system proved to be quite challenging. With Ethereum, the goal is to take the idea behind Bitcoin, the technology, the blockchain and try to make it more generalized. Attempt to come up with an architecture that you could use to build decentralized applications in a very general sense. It's an area that's fairly rapidly moving forward, but it's important to ask the question:

So what is decentralization, anyway? What are the benefits and what are we trying to get out of this? There have been lots of people trying to define the word, there have been lots of people trying to define distributed, define decentralized and figure out what the difference between the two are. Most people that are trying to do this tend to not come up with particularly good classifications and probably more people think that decentralized and distributed or synonyms. We're going to break down decentralization into 3 parts:

  • Architectural decentralization - is it one physical system or many physical systems? Does everything depend on one particular computer, one particular root server, one particular central architecture for managing things or is it spread out between many computers or many systems where even if a small portion of any of them go offline or get hacked, the rest of the systems keep working.
  • Political decentralization – Is it controlled by one person or entity or many? The idea here basically is to not just build the physical architecture or the physical distribution, it's about control. Is the system your protocol, is it controlled by a single person or one entity, or is it controlled by everyone together.
  • Logical decentralization – is the interface or data structure a singleton, or an amorphous swarm? The third one is hard to understand but I'll give a few examples and hopefully, it will become clearer.

You can come up with various combinations of this. Some of the combinations are harder to comprehend than other combinations, but as one quick example, if you look at a direct democracy, then it's architecturally centralized because you have one parliament where you have a lot of government buildings where decisions are being made.

You have police, you have the military, but at the same time, it's more politically decentralized because even though there is kind of a few chambers. The actual decision-making power is still widely spread among large groups of people. It's also logically centralized because there is one government, it does create one set of laws. Now if you take a look at the other extreme, you look at something like the English language.

It's politically decentralized, no one controls it, it's architecturally decentralized, there is no single place where it's taught or where it gets created and it's also logically decentralized. What that means is that the English language, spoken by one people on one side of the world and the English language spoken in other parts of the world may well actually be different.

You could easily have different English speakers disagree on a few words and their meanings, but even still it roughly works. Meanwhile in other cases, with the blockchain, you have something which is architecturally decentralized, politically decentralized but it is logically centralized and I would argue that actually has benefits. Quite a lot of the time when people talk about blockchains, they actually talk about the benefits of having been in a single database instead of having every single banker, every single institution has their own database and have some very complex and very inefficient process for reconciling between them.

The argument being logical centralization in many cases actually does have benefits. These kinds of decentralized have benefits in various kinds of situations, sometimes you might want logical decentralization, sometimes you might want logical centralization. Sometimes a blockchain isn't necessarily the right tool for you. In some cases, that it is. We can work together and figure out what is the best way and place to use them.

Chapter 5: Why Decentralize?

Here we have another categorization and it's split up into different reasons:

  • Efficiency – decentralization may be more efficient due to diseconomies of scale or spare capacity considerations.
  • Fault tolerance – use redundancy to reduce the risk of a system failing accidentally.
  • Attack resistance – the cost of attack sublinear, so splitting a system into components makes attacks more costly.
  • Collusion resistance – making it more difficult for a sub-group of a community to act for its own benefit at everyone else's expense.

In some cases, making a system decentralized actually is just the most efficient way to build it. In some cases, you want to have decentralized production because your customers are decentralized and you want to have one producer close to every set of customers. The physical distribution of something let's say restaurants is really obvious and should be decentralized because if all of the restaurants were in one city, then everyone would have to drive 500km every time they wanted to go to a restaurant. So that's just one trivial example where decentralization makes things more efficient. Now there are also other reasons.

For example, spare capacity considerations.

People talk about the efficiencies you can get with Uber and AirBnb, what are they talking about? There are people who already have their own cars, there are people that already have their own houses and they have resources that they're only using some of the time. They have the unused spare capacity.

The idea is you can get extra efficiency because instead of building new infrastructure specifically for renting it out for a day or giving people taxi rides, you can use spare capacity in infrastructure that already exists. So this is one set of advantages. Another set of advantages is fault tolerance. So decentralized systems, any system has some risk of breaking, but decentralized systems are less likely to break. There isn't one single thing where if the one single thing breaks then the entire internet goes down.

The third argument is attack resistance.

It's basically a matter of thinking how hard is it to bring the system down. How hard is it to cause the system to misbehave. Basically, the idea is if you try to build systems that have very expensive sensors, then even if you spend 100 times more resources making your sensor, it's not going to take 100 times more resources to attack. Basically the more you combined or bring components together into one highly centralized thing, in fact, the easier it is to cause the system to break. The easier it is to disrupt it. If you want to build something that's stable, then you want to try to distribute it to as many components as possible. Another argument is collision resistance.

This is one where political decentralization really starts coming to the forefront. Basically, the way to think about it is a way of making it more difficult for one particular subgroup of a community or a market, or an ecosystem to come together and act in its own benefit at everyone else's expense.

The simplest example is, you really want to avoid having monopolies. If you have a monopoly then everyone who owns the monopoly will be able to raise prices, get higher profits, but from the point of the ecosystem, from the point of view of the community in general, prices go up and there are lots of activities that really should be possible but stop being possible because prices go higher and it's generally bad for the economy. Now monopolies can do lots of things other than raising prices, and some of them in the long term may even be dangerous. There are some of the reasons why decentralized systems make sense.

A lot of these are reasons that people reading this should relate to. From the point of view of efficiency, this idea that forms of energy production like solar and wind are more naturally and become more efficient if you decentralized them. If you create this ecosystem where instead of relying on a few large firms provide all of the electricity, you'd basically let people put solar panels on their rooftops, put windmills on their rooftops and create this economy where not only do you have millions of consumers, but you also have millions of producers.

There's a lot of blockchain based decentralized storage or decentralized computation startups and they're basically trying to leverage spare capacity in the same way as well. In other natural use cases is the content delivery network. If you as a user try to access a website, where is the computer serving the network to you?

The closer you can find a computer or server to yourself, the faster you're going to get the site. That's one area where decentralization makes sense as well. So that's one example. From the point of fault tolerance, here's an interesting article from a few years ago.

Basically what this is saying is if you look at the decisions made by the United States supreme court and if you look at the citations that they make, for every time they have a link and you try to follow the link, 49% of the time, that link is broken. Now, why is the link broken? Basically, because the way linking to data on the internet works is it's a fundamentally centralized paradigm. It's a paradigm where the first part of the link says what website you want to go, what server you talk too, and the second part of the link says what do you ask them? The problem is what if the original server goes down?

Chances are if you right about something, 5 years later you might forget about it or you might put it up on some blogging platform and 5 years later the blogging platform shuts down. What do you do then? IPFS is a company trying to make a more decentralized standard for this where instead of linking to a piece of data by which particular server is supposed to host it, you would link to a piece of data by making a hash, making a cryptographic pointer to what that specific piece of data is. As long as there's one computer on the internet that hosts it, as long as there's one computer on the internet that has the data and is willing to send it to you, you would be able to get the data back.

This is one practical example of how decentralization and specifically how political decentralization can get you fault tolerance. Power outages is one example of how centralization can lead to fragility as well. From attack resistance, another example of one of the major dangers of centralization, the following is something that happened a few years ago and this about another piece of our internet infrastructure that ended up being quite centralized. It's certificate authorities.

It's the companies that if you access some major website whether it's Google or some major bank if you access them through https:// you get this green lock on your browser, this is actually that website, it's not some hacker that managed to man in the middle your connection. Well, the companies that actually provide the certificates, the public keys that make that system work, one of those companies got hacked a few years ago and basically a huge number of websites suddenly because vulnerable.

This is another thing where there's quite a lot of people that are looking to see if they could decentralize. Here's another instance of centralization being dangerous:

This is one high-profile example, but there have been 3-4 similar examples that happened just in the past few years. In general, when you have points of centralization, points of centralization become points of centralized data collection and a point of centralized data collection is something that can get attacked and could lead to very large data leaks very easily. We have data leaks almost everywhere, basically, if you use some online service, you can expect with high probability that anything you do within the service will at some point years from now, get leaked and the world will have access to it.

Ashley Madison was a particularly high-profile example because this website was basically a dating site for people to have affairs and you could expect if people had affairs you would expect people would want to have their affairs kept private.

The site got hacked and a few million people had very nasty surprises. It's also important to keep in mind there are two kinds of costs:

  • The seen – losses due to existing over-centralization. Security or faults happening.
  • The unseen – unrealized efficiency gains that do not happen because people are afraid of upgrading to network-enabled technologies due to their present centralization.

One very simple example would be online voting. Practically speaking that we still use pen and paper to vote in some cases is chaotic and an argument could be made that it's inefficient. But to a 21st century person it's pathetic. On the other hand, you see a lot of people are afraid to switch to a more electronic version because they're afraid of a huge number of very real security risks. That's just one example we can call an unseen cost of bad cyber-security. Collusion resistance and monopolies.

The ways monopolies are bad to tend to be very subtle as well, you have of the main issues you have is sort of "bait and switch" where people, entrepreneurs, businesses end up making themselves dependent on one particular centralized provider and the way that one centralized provider works suddenly ends up switching on you.

People make startups based on Twitter APIs and then Twitter changes their policies and your business model is gone overnight. One solution to all of these problems is federation. Where you can build protocols where instead of relying on one centralized provider, you would have this protocol where you could rely on a set of centralized providers.

The idea being anyone could set up one of these servers and anyone could use which server they could personally want and people that use different servers would be able to communicate to each other. One of the problems with this is a kind of market equilibrium that's proved to be very hard to sustain because as soon as there is any dominant provider, their going to have the incentive to embrace, extend and extinguish. They're going to have the incentive of making proprietary extensions to make themselves incompatible with everyone else and basically try to turn the entire thing back to a centralized system again.

Another problem is, in general, in federalized protocols, the diaspora is an example. What ended up happening is the largest server that people set up, ended up getting a 50% market share. So federation is something that does have a lot of challenges. Fully decentralized open protocols are one of those solutions that might make more sense and blockchains are one example. Blockchains don't rely on one people choosing one particular server, blockchains are protocols that work without any one particular centralized intermediary being necessary. How do these ideas fit into the energy industry? You have to think about what it is you're trying to decentralize.

  • Production of energy
  • Distribution (ie:// making and maintaining wires, energy storage, infrastructure)
  • Source authentication (ie:// is the power green? Is it financing negligent profit seekers)
  • Market mechanisms (ie:// allocation and incentivization)

Blockchains seem most well-suited for markets, though we can also think about source authentication.

We can also think about source authentication, supply tracking, supply verification and those kinds of use cases as well. Markets are interesting because in some way markets even as they are already an incentivized distributed computation algorithm. They're a storage of areas where a decentralized computation is an implicit sense is happening now. But the question is can we actually build on this and build decentralized market infrastructures that are even better. So markets, in general, are stately, in a sense, they are fairly complex mechanisms that often do have a notion of an internal state where they have to remember some notion of internal information.

So if decentralization makes sense then blockchains might as well be part of the solution. How could decentralization help? In one area it might help is fairness. In market designs, proving that these market designs are fair, that 60 years of math, game theory, all sorts of fairly complicated research, but what blockchains can do is prove that the algorithms that people think are being used are actually the algorithm that is being executed.

Also privacy preservation. In general, points of centralization become points of data collection and in general, blockchains don't provide privacy by themselves, but with strong cryptography, with decentralization as a holistic principle, then you can create architectures that do a much better job of preserving privacy.

Why preserve privacy?

It's important to understand the points that privacy isn't just about big brother paranoia, privacy is also a critical requirement for incentive compatibility of many classes of mechanisms. In many cases, it's a key part of making markets work. 

author default image - Managing Institutional Investments In Digital Currency

Buy, sell and invest in digital currency instantly with zero fees. A managed cryptocurrency portfolio with full custody solutions. Click here to create an account.