Analysis Featured

A Discussion About Internet 3.0 , Decentralized Everything

Internet 3.0
Fig. 1. A map of some of the decentralized space from here.

Internet 1.0 is HTML websites.
Internet 2.0 is a social network and user-created content.
How is Internet 3.0 coming along?

What is Internet 3.0?

Are you familiar with Napster, Kazaa, and BitTorrent? Today, Bittorent has met Bitcoin and given birth to the following startups, networks, or organizations:

  • Decentralized computing power. Golem, among others, is a peer-to-peer market for putting your computer’s excess CPU power to use for other people. It works because there is no easy way to pay anybody on the planet fractions of a dollar for having used their CPU for 1 minute. This is, however, possible via blockchain.
  • Decentralized exchanges. Ether Delta, among others, is a cryptocurrency exchange that operates in a decentralized way (i.e., without a central counterparty). Decentralized exchanges allow peer-to-peer trading, which means that when a trade is executed the items are exchanged directly between the traders without touching any third party, and without the traders being able to stop the exchange. This approach eliminates counter-party risk entirely. On the other side, it also allows people to trade completely anonymously.
  • Decentralized protocol approval. Tezos, among others, is an open-source platform for assets and applications and allows the participants to vote to change its rules and protocols. Participants can choose to change the fee structure, rules, the protocol APIs, nearly everything. This protocol change-mechanism is built within the network rules, and nobody has the right of veto or override. Imagine if eBay merchants could vote to reduce the eBay fees without the eBay management being able to stop it. Of course, this opens the doors to politics, and also to oligarchies as having more Tezos coins obviously gives you more power to influence the votes.

Other similar companies include but are not limited to:

  • Decentralized file storage (Filecoin)
  • Decentralized domain naming (Namecoin)
  • Decentralized cloud storage (Storj)
  • Decentralized databases (BigchainDB, IPFS)
  • Decentralized internet address allocation (JACS)
  • Decentralized Video Encoding and Streaming (Livepeer).
  • Decentralize financial services (Bitcoin, Litecoin, etc.) and more.


Internet 3.0

Fig. 1. A map of some of the decentralized space from here.

Business models

Centralized online marketplaces, like Amazon, Uber, E-Bay or Lending Club, typically earn roughly 10%-35% of the value exchanges through the platform.

Other online platforms like Facebook or Google don’t share any of the ad revenue earned from the personal-data exchanged through the platform. They keep 100%.

In addition, all centralized marketplaces and platforms exert full control over who can advertise, who and what can be sold, to whom, where, etc.

Their full control, when the company is young or fragile, is not being exercised much. They want to attract users and customers. However, as the company grows, and pressure from investors and the financial markets increases, the platform position of the de facto monopoly in their sector is usually leveraged to increase fees and to control who and what can be transacted on the platform. For example, Google has a history of banning certain ad categories on its platform. Most people agree that the bans, so far, have been legitimate and are targeting harmful or mostly fraudulent industries from selling their products and services. However, Google’s power of life-or-death over entire industries is troublesome.

In comparison, decentralized networks and organizations have so far mostly tried a few different business models.

Financing and crypto coins

Traditional , centralized, startups sell their equity to investors. Equity is scarce by definition, to 100%. And once sold, investors typically have a contractual right preventing startups from creating more shares and diluting them without their approval.

Equity is a problem in a decentralized project. Equity to what? What does equity holders control?

Most decentralized organizations mentioned above have created their own crypto coins in order to finance their creation. Their usual business model is to make the coin, artificially or legitimately, a required part of each transaction on their network. As the number of transactions grows and the coin inventory is limited, the coins become more valuable. And the network itself uses its own inventory of coins to finance its expenses. In addition, some decentralized networks also take a percentage of the value exchanged on their platforms.

However, the token approach has, so far, failed to work for most networks.

The most successful tokens today have thousands of active daily addresses.

active network addresses

Fig. 2 : Number of active addresses per network per day, log scale, for MakerDAO ( purple), Tezos (blue), Binance (orange).

This is not surprising. All these decentralized organizations are new startups. It takes time for startups to build traction. A handful of them will have millions of users after 3-5 years. Most startups may still be viable businesses even though they only have hundreds of daily active users, but their tokens will not have any real value due to over-inventory. Therefore, maybe relying on token activity and scarcity to finance all decentralized projects may not be a viable way to finance these projects.

I believe an alternative token model is needed for most of these projects. A model that will have significant returns to investors even if the network only achieves the modest success of 100s of transactions per day. However, this may require an increase in network fees.

The 4 Open questions of decentralized entities

As I think of decentralization, many questions are on my  mind:

  1. What are these entities? Are businesses, networks, organizations, protocols, or something else? The concept of Decentralized Autonomous Organization, or DAO, has been used in the past. But to my knowledge, no actively operating entity using a real DAO model is live and generating revenue today. All entities have executives, employees, bank accounts, offices, etc. Or is it? The Bitcoin network itself, with all the developers in various organizations who are trying to contribute to it, is fairly decentralized.
  2. Governance: Leaders in centralized entities are required. Often, leaders aren’t any good at taking decisions, but making some decisions is often better than not being able to make any decision. Many an organization has died because nothing at all was done. Are decentralized organizations able to make decisions fast and efficiently over 5 to 10 years while they grow?
  3. Are decentralized networks cheaper to run, and do they have a disruptor advantage over centralized networks? It is not clear. Lending Club, one of the first P2P lending startups, argued that their cost structure was cheaper than banks’. However, it turns out the cost of capital lending and cost of customer acquisition were under-estimated and banks have cheaper capital and cheaper customer acquisition. Lending Club’s profit margins are not impressive. Neither is Uber’s. Nor are Amazon’s. I believe there is no single answer to this question, but assuming that a decentralized entity is more cost-effective than a centralized entity is not obvious. In human history, disciplined centralized organizations (armies, empires, …) have clearly been more successful than federations, communes, etc.
  4. Is there value built, and where is it? The startup/VC model has worked since the Dot Com boom because it was a profitable model for everybody involved. VCs made money, and successful entrepreneurs attracted more smart wannabe entrepreneurs. It is very important to see the founders and investors in these decentralized organizations be successful or there will be no second-generation decentralized entities.


What is the innovation here?

I believe that an exchange that can work without counterparty risk is a real innovation.

I believe that a method to pay fractions of dollars efficiently to anybody on the planet is a real innovation.

I believe one day we will see the Netflix of Internet 3.0 bankrupt the Blockbuster of Internet 0, 1.0, or 2.0.

However, questions remain. Is decentralization in a business similar to communism in politics? Does this model really work? In 1990, in Moscow, everything was rationed, bread was extremely scarce. When a communist leader asked the London mayor who is in charge of the bread supply to London so they can learn their secrets, the mayor, confused, answered “Nobody!” Our modern food supply is a decentralized market, and fewer and fewer people are going hungry.


George Popescu
George Popescu


default image

"Your daily letter is great!" , Ram , Founder and CEO, PeerIQ

default image

"Hi George - just want to tell you that you are doing a great work with Lending Times;-) Brgds, Kasper" , Kasper, Partner and Co-founder at Dansk Faktura Børs A/S

default image

"I've been following your newsletter for some time now and have been very impressed with the content." Charlie,Co-Founder | Bolstr

default image

"Hey George, I must say I really enjoy your site. It has inspired me to do some changes at our platform and we are the biggest consumer lender in Sweden." , Ludwig, CEO @ Savelend Sweden AB

default image

"Your daily email is very useful. It gives quick update on what's going in the market. Thank you very much for all that info." Yann Murciano, Head of Base Metals Trading at Morgan Stanley

Our daily p2p news digest

Daily News Summary Digest Sent Daily To Your Inbox