Ethereum: Re-decentralizing the Web

Everyone who enters this space generally starts with Bitcoin, Nakamoto’s white paper provides a nice foundation. After Bitcoin, people notice Ethereum because it’s got the second largest market cap (and supposedly gaining steam). While understanding Bitcoin certainly does help the transition, Ethereum is much wider in scope— and certainly goes beyond the financial use case.

At a high-level, Ethereum is best understood as part of a movement to re-decentralize the Web. To understand why, we have to revisit the early-90’s when the Web was commercialized.

Web 1.0

The Internet comprises packets of information sent through loosely connected networks of computers. The Web, built on top, was conceived as a way to organize information (read the original 1989 proposal here).

Initially, the Web only allowed for information to be consumed (read-only). And yet, this allowed anyone to have a presence.

At the time, this was revolutionary — anyone could have their own piece of cyberspace (remember that term?). Back then, one had to be a columnist for a magazine or newspaper to have an outlet. So, when this ‘thing’ comes along and allowed anyone to project their idea for the world to read — it was revolutionary.

Web 2.0

Fast forward to after the first dot com crash (~ 2005), the Web went through a shift. It was when Tim O’Reilly and associates popularized the term ‘Web 2.0’ that the idea of Web platform business models began to take shape, seeding the digital giants we’re all familiar with today (FANG: Facebook, Amazon, Netflix, Google); (GAFA: Google, Apple, Facebook, Amazon); (BAT: Baidu, Alibaba, Tencent) and (FAAMG).

Many of these ideas, outlined in O’Reilly’s “What is Web 2.0?” essay, defined the era and shaped the competitive landscape. And it was the successful application of these ideas that propelled centralization.

The first principle was Web-as-platform. The main idea here is that ‘a service automatically gets better the more people use it’ (O’Reilly, 2005). And the more people use it, the more important database management becomes. For Web platforms, software is only as useful as the ‘scale and dynamism of the data it helps to manage’ (O’Reilly, 2005).

Another core idea was that users added value, so it was important to capture (and control) user data. This would give businesses a competitive advantage. Although many of the ideas behind Web 2.0 advocated for an open Web (i.e., using peer-to-peer protocols, open source) as companies sought to cultivate network effects, the move to tighten control around user data heighted.

Increasing Centralization

As companies realize the value in data, they developed sophisticated ways to lock user in. Google’s platform emphasizes familiarity. The cottage industry around search engine optimization (SEO) and google analytics bred familiarization further prevented advertisers / users from switching.

With Facebook, not only are all your friends there, the platform becomes your online identity, allowing you to sign-in to other applications, which perpetuates lock-in. Even if you wanted to move your contact list, status updates and content (text, pictures, videos) to another platform, you wouldn’t be able to. The terms of service meant signing away control of your content.

Once locked in, it becomes about keeping users on the platform. Over the last six months, Facebook has developed a suite of products — from messenger games, to Instagram stories, to ordering food in-app — so that users never have to leave. Leaving will be more difficult over time as Facebook figures out ways to extract more data.

Thus, vendor lock-in, coupling of user data with applications and centralization go hand-in-hand.

To further secure their advantage, companies began excluding external networks from participating, preventing users from linking to other platforms. Thus, the walled gardens became increasingly siloed.

Why is this bad?

Facebook offers an invaluable service. No other platform allows me to keep in touch with friends who have long since moved across geographic borders. It’s not an accident these companies are worth billions.

However, there is a trade-off.

As we sign the terms of agreement, we give up our data, privacy and security to these firms. As more of our data is stored with a few major players, society become more vulnerable. These digital giants, like any ‘trusted’ third party are security holes. The Snowden leaks have revealed to the world, not just US citizens, that large government agencies like the NSA have access to our data and are spying on everyone.

In 2015, Apple was ordered to build a back door to extract user data — contacts, photos and calls history — from locked iPhones (see FBI-Apple Encryption dispute). To their credit, they refused. But the fact that they had to make that choice should give all Apple users pause. Opening a back door for law enforcement also opens it up to nefarious actors.

Unfortunately, it’s not just large corporations attempting to centralize (control) the Web; it’s government, China and North Korea, being the most obvious examples. The Thai government tried to pull a similar stunt recently (fortunately that’s been scrapped).

Another side effect of Web centralization is lost net neutrality; there is increasing evidence of government interference in democratic processes and widespread misinformation.

Finally, this has stunted innovation and progress. In some ways, these companies are doing what they’re supposed to do in a capitalistic endeavor. They’re building sustainable competitive, highly defensible moats around their business and are exceedingly effective at it. But over time, this results in less choice for users.

Re-Decentralization

As a result, there’s been a movement back toward decentralization. Tim Berners-Lee has been critical of the Web’s current state. In 2016, a summit was held to get the Web back to its founding values.

Several things need to happen.

First, there needs to be a separation between applications and (user) data. Users should be able to choose where to store their data (if, for whatever reason, they want to take their data to a different application, they can). Thus, people gain control over their data, while still accessing needed services. Berners-Lee refers to this as exposing users to the Web of Data (in addition to the Web of Documents). Simply put, users become data owners.

Ruben Verbough describes a decentralized version of Facebook where different parts of the interaction could be stored in different data servers. For illustration, this means the data for a profile picture is stored in one location; status updates might be stored in another location; ‘liking’ and commenting might be stored, on yet, another location — a newsfeed can be divvied into separate data storages, all chosen by the user. While this sounds complicated, software can streamline this operation so that this happens behind the scenes, after the user has made an informed decision on how to manage their data.

When users are in control, applications have to ask for permission to access user data. Users can choose which application they want to display their data as well as where to keep their data. This forces service providers on both ends to step up their game.

Solid, a project led by Berners-Lee, provides tools to create social applications giving users true ownership of their data. Because of its modular design, users can switch apps and storage providers without losing any data along the way. In turn, developers can reuse data that already exists to create new applications.

In a decentralized environment, users require authentication to allow (or deny) access to their data. Solid uses the WebID 1.0 (Web Identity and Discovery) standard to provide users with universal usernames/IDs. WebID uses public-key cryptography as the basis of identity management.

Finally, standards must be written to ensure that data on the open Web can be linked, similarly to what hypertext transfer protocol (HTTP) did for documents (see Linked Data platform). This will prevent platforms from excluding access, curbing the tendency towards data silos. Moreover, it allows unexpected, yet surprising discoveries to happen when data is linked in meaningful ways.

Nevertheless, before a decentralized version of Facebook (or any of the dominant social networks) can be created, people have to adopt the Solid protocol first. The question is:

How do we get people to adopt this protocol?

For all the technical features of the Solid project, this is a question that has received a less than satisfactory answer. Tim Berners-Lee, knows better than most, how difficult it is to get a protocol to take off. As important as Solid is to Web re-decentralization, one thing is missing: economic incentives.

Ethereum: Protocol Adoption

The biggest weakness of the early Web was a lack of payment system. Thus, protocols like TCP/IP produced immense value without being able to capture it. All the value went to applications built on top of these protocols.

This is reversed for distributed ledger technologies.

Here, the protocol layers capture most of the value. In a decentralized web, user data, separated from the apps, is distributed across applications which makes the space more competitive; this levels the playing field for new entrants.

But decentralization is not enough for adoption.

A native token can be used to incentivize early protocol adoption by developers, entrepreneurs and speculators. Because they’re invested, they will build applications on the protocol to positively affect token value (see Joel Monegro’s highly cited post for details).

This economic incentive is what is missing from earlier protocols like TCP/IP and even Solid. If we are to have any hopes for decentralized social networks to compete with the incumbents, there needs to be a way to incentivize early adopters, entrepreneurs and developers to create a vibrant ecosystem.

The Ethereum Project, which has its own blockchain, is poised to be a potent force towards a decentralized internet. Ethereum is designed to be a general purpose, agnostic platform that allows developers to build decentralized applications (dapps) on top of the blockchain. For authentication, Ethereum developers are building an identity management system (much like Solid’s WebID) into the platform (uport self-sovereign identity system is another option, also built on Ethereum). Ethereum has a native token system (Ether) that can be used to incentivize participation and protocol adoption (Ether is needed by developers looking to build apps on the Ethereum blockchain and interact with smart contracts).

Furthermore, Ethereum is one third of an ensemble (along with Swarm and Whisper) designed to support in a decentralized Web. While Ethereum facilitates applications, Swarm, a peer-to-peer data storage and distribution network, allow users to host and fetch their data without a centralized server. Whisper allow for private, secure messaging between nodes.

There is a level of synergy that comes with using the Ethereum platform. Developers can create apps. Users can store their data on Swarm (for alternatives: IPFS, Storj, or maidsafe) and communicate with other nodes through Whisper. Finally, there are browsers built on Ethereum (MIST, Toshi).

This is not to say that projects like Solid and Ethereum could not collaborate. There could be room for Ethereum to adopt Solid’s Linked Data principles, while Solid could benefit from Ethereum’s flourishing ecosystem.

It’s too early to tell which protocol will provide the underlying infrastructure for a decentralized Web; while protocols like Solid look promising, the economic mechanism inherent in Ethereum will take it further.