In the web3/blockchain ecosystem, Joel Monegro’s post on fat protocols has been so influential that it has become standard to believe that value accrues at the protocol level.
I dissent: there’s no such thing as “fat protocols.”
It is the early days of the decentralized internet. So without a doubt much of the infrastructure needs to be built so that awesome consumer apps can be built on those foundations. But there’s no reason to think that value intrinsically accrues at the protocol level.
I modified Monegro’s graphics:
App or protocol? Heck if I know!
What is a protocol and what is an app? Monegro doesn’t define the terms; I can argue both sides to this question for nearly every major project.
- Are Gnosis and Augur protocols (because they want devs to build on their platform) or are they apps (because they’ll be consumer facing)?
- Is Steem a protocol (it’s a blockchain you can build UIs for) or an app (it may have the largest user base today)?
- Is Status an app (it’s literally a mobile app) or a protocol (it’s a mobile platform for Ethereum)?
What about Bitcoin? A few years ago, the maximalists talked about Bitcoin as a protocol that would enable everything to be built on Bitcoin. However, lately it seems to have settled on being a digital gold app rather than a protocol to build stuff on it.
For argument’s sake, I’ll discuss Ethereum from here on out, both because it is what I pay the most attention to, but also where most blockchain projects are choosing to build.
Below is an expansive view of the web3 stack from Stephan Tual, but note that there will be plenty of layers of things built in what Tual lumps together as “decentralized applications.”
Ethereum replaces the server, not TCP/IP
Monegro proffers that, “previous generation of shared protocols (TCP/IP, HTTP, SMTP, etc.) produced immeasurable amounts of value, but most of it got captured and re-aggregated on top at the applications layer.“ Maybe – I’m not convinced by this analogy because the reason that TCP/IP won over previous standards is that it was free.
Ethereum is replacing the need for a server, not TCP/IP. We don’t talk about this as much lately, but in the early days of Ethereum all the talk was about enabling “serverless” applications.
Think about what a blockchain is – it’s just a database! For example, Ethereum will let you replace the backend of your internet app with logic and value stored on the EVM, data in Swarm, and messages in Whisper.
Ethereum is just a platform for decentralized software to be built using a blockchain instead of a server.
Databases and servers get commoditized eventually
In the long-run, Ethereum (or whatever fully allows decentralized software to be built) will also become commoditized. That’s what happens in technology. It’s inevitable. Even if they do achieve world domination, blockchains won’t run the world forever.
In the medium-term, we’re still far away from blockchains becoming commodities. It’s a nascent technology that is still being built. Largely the pace of commoditization will depend on scalability and interoperability.
Even in the short-term, the market value of Ether is about $30 billion, and the market value of all the tokens is about $7 billion. Ethereum has been live for over 2 years, but none of the tokens are live. If tokens work, those token values will go way up. That seems to indicate that the current market isn’t convinced that value accrues to protocol.
Intrinsic value of Ether depends on scalability and interoperability
The inherent value of Ether is due to the fact that it buys computational power on the decentralized database. The price of Ether is thus determined by supply and demand for that computational power.
Demand is largely speculative at the moment, but long-term demand for Ether will depend on how many things get built on top of Ethereum.
Supply is determined by how scalable blockchain can become. 15 transactions per second isn’t good enough.
Eventually blockchains will solve scalability and you’ll be able to use any of them. Since it hasn’t happened yet, we don’t know when or how this will happen. If it does happen, much of the value will accrue to that scalable chain in the short-term. But eventually other chains will figure out better ways of doing it, there will be a copious supply, and it will become commoditized.
Or perhaps millions of chains will exist and scalability occurs through the interoperability of those chains. Either way, the supply of computational power will force prices down to a commodity level.
When these foundational elements begin to become a commodity, all that value will begin to accrue up the stack, just like it did in the Dot Com era and in the web2.0 era.
This is a good thing™, because it means we can’t even imagine yet the apps that will get built.
Thanks to Dunning_Krugerrands, Daniel Zakrisson, and Corey Petty for feedback and conversations around this post.