Web3 is just expensive P2P

Web3 claims to be the only way to save us from commercial entities defining what we can see and what not. Yet, it does exactly this: It results in commercialized entities defining what we can see and what not.

More on the blockchain ecosystem in this overview and especially in the simple, humorous, yet thorough article «Hitchhiker’s Guide to the Blockchain».

What is Web3?

According to its proponents, Web3 is going to „increased data security, scalability, and privacy for users and combat the influence of large technology companies„. How it does this is not well defined. Cryptocurrency and blockchain supporters envision these goals to be achieved by basing Web3 on cryptocurrencies and the blockchain, as a Decentralized Autonomous Organization.

Moxie Marlinspike’s attempts at creating Web3 distributed apps were sobering:

  • It was not decentralized, it required you to go through one of a few gateways to make the app practical.
  • These gateways exert at least as much control over what you can do and cannot do than traditional web hosting sites.
  • Doing the operations on an actual blockchain is very expensive.
  • Most users actually don’t care.

Distributed systems

Between centralized to decentralized (or distributed) systems, there are federated systems, sharing properties of both.

Networked systems, like a lot of the Internet and its applications, typically come in one of three forms:

  • Centralized applications, which run one one system (or at least appear to). E.g. your bank’s e-banking or most commercial Internet services like Zoom or Slack.
  • Federated systems, where a lot of people/organizations run a lot of centralized systems; however, these centralized systems can talk to each other. The most well-known example is email: Anyone can operate a mail server, yet you can send mail both locally (on the same server) or to any other (remote) server. Each of these servers is responsible for a (list of) domains and everyone can find out which server to contact for a particular destination domain.
  • Decentralized or distributed applications, where responsibility is much more fluid. E.g., the early peer-to-peer (P2P) systems for file sharing, where all participants were connected to each other, directly or indirectly. Searching for a particular file would result in that query being sent to many (or even all) of the systems; anyone with answers would reply.

The blockchain is also a distributed system. In this particular case, all the blockchain data is copied (replicated) among the nodes, i.e., everyone (conceptually) has a copy of all the data.

Web3 vs. P2P distributed apps

During the heyday of Peer-to-Peer systems around two decades ago, people contributed spare resources (computing, storage, network bandwidth) willingly to what they believed to be worthy causes. Today, a handful of Watts can go a long way with tiny, efficient, and inexpensive computers such as the Raspberry Pi, making a personal server extremely affordable.

So, essentially, the P2P community established something like a tiny, lightweight communist bubble, to the benefits of everyone. It wasn’t perfect, but it worked, if someone could install the software and keep it running.

The Web3 community approaches this from the opposite side: Instead of a lightweight communistic take, they follow a heavyweight libertarian path: Everything should be monetized.

For this, they built a complicated, expensive underpinning, and added more layers on top of this, each more complicated, failure-prone, and expensive. Everyone ends up spending more money and other resources. However, the server administration is now done by someone who gets paid for this, but it is still up to you to get the actual Web3 software running (and writing a correct smart contract is among the most ambitious software projects, complicated and error prone).

Which is better?

P2P never made it to the mainstream; instead, centralized systems still (or, again? even more?) dominate the landscape. So, because it failed, should we try Web3? I doubt it, as it is more likely to fail, due to higher complexity and more dependencies.

So, unless the setup and maintenance trouble of the initial setup can be reduced (and there are community efforts out there), the best way to go is the centralized model: You pay someone specific to do this. Maybe set up a not-for-profit association to provide the necessary money.

Neither in the centralized, P2P, nor Web3 world, things can be left unmaintained. Anything else is an illusion.

What else?

In the 15 months of its life, Molly White’s Web3IsGoingGreat reported on burnt/lost/stolen/… assets valued ~10 billion USD.


This article was inspired by Cory Doctorow in two ways:

  1. The topic was inspired by his article «A useful, critical taxonomy of decentralization, beyond blockchains».
  2. His Memex Method explains how to use a blog post as a bookmark which you can actually reuse. I’m trying this here.

Further reading

Schreibe einen Kommentar

Diese Website verwendet Akismet, um Spam zu reduzieren. Erfahre mehr darüber, wie deine Kommentardaten verarbeitet werden.

Web apps