Blockchain Puzzle Pieces—Understanding Application Stacks and How Value is Created
This is a write-up of my presentation from the Chicago Blockchain Meetup Event on 9/28. To find out more about future events, check out the group here.
When I first started reading about blockchain, one of the most confusing aspects to me was the set of interactions amongst different components of the “blockchain application stack” — blockchains, protocols, tokens, and applications.
Various books, guides, and Medium posts did an excellent job explaining these components individually, but how the components worked together was a mystery that I’ve tried to spend some time unraveling over the past few weeks.
By reading various whitepapers and attempting to deconstruct projects in the space, I put together a basic framework for understanding (see below) — any feedback or comments are greatly appreciated!
The Various Components of the “Stack”
The following diagram, taken from Preethi Kasireddy’s post on Hackernoon “Bitcoin, Ethereum, Blockchain, Tokens, ICOs: Why should anyone care?” was helpful in providing a basic framework:
On the surface, the above diagram seems really simple, but when I began exploring different companies in the space, I found it difficult to easily apply this universal framework.
Another important concept to grasp is that the decentralized web will almost definitely ascribe value differently than the web today. One core example, outlined by Joel Monegro (formerly of Union Square Ventures; now at Placeholder) was explained in the post “Fat Protocols” from August 2016.
At least today, nearly all of the value in the blockchain application stack is captured at the protocol layer (Bitcoin, Ethereum, Litecoin). Since this post was published, the global market cap of the protocol layer has skyrocketed, going from $11B at the time of the post (8/16) to more than $150B today.
The value of the application layer has also greatly increased — with Coinbase, as a blockchain application, worth $1.6B as of its latest funding round. Many other application ICOs have also garnered significant value, but, regardless, the value gap between the two sides continues to be huge.
This is very different than the web today — where the protocol layer (TCP/IP, SMTP, HTTPS, etc.) holds no direct value, and the combined value of applications built on this layer (Google, Facebook, Salesforce, etc.) are measured in trillions of dollars.
Here’s a brief description for each of the “components’ of the blockchain application stack:
(1) What is a Blockchain?
From Will Little’s excellent post on Hackernoon,”A Primer on Blockchains, Protocols, and Token Sales” — A blockchain is a very long list of transactions stored on a network of computers (i.e. a distributed ledger).
(2) What is a Protocol?
From “Bitcoin, Ethereum, Blockchain, Tokens, ICOs: Why should anyone care?”:
Protocols are the special sets of rules that nodes in a network use when they transmit information. For blockchain applications, protocols are “cryptoeconomic rules” that are enforced by a blockchain in order to maintain distributed consensus across the blockchain’s peer-to-peer network.
(3) What is a Token?
Tokens are more complicated, and I’ll plan to write a completely separate post to explain this concept in its entirety. The varied nature of tokens adds a lot of complexity to understanding the technology stack.
It’s important to understand, though, that there are different kinds of tokens. This post from Nick Tomaino breaks tokens into 4 types, which is a useful shorthand for starting to think about the different use-cases:
- Traditional Asset Tokens
- Usage Tokens
- Work Tokens
- Hybrid (usage + work tokens)
(4) What is an Application / Decentralized Application / dApp?
dApps are applications built on top of the blockchain. These applications have backend code running on a decentralized peer-to-peer network (like Ethereum) and can have frontend code and user interfaces built in any number of different languages. The backend running on a decentralized network is the key differentiator.
How Do These Components Work Together?
The following low-fidelity diagram is my attempt to show how a few different blockchain application stacks are constructed, by looking at the pieces that make up the different components explained above.
One of the important takeaways is that applications can look similar on the surface, but have very different infrastructures constructed from a mix of shared and custom components.
In the first example above, Ethlance.com is an early, full-functional dApp that is built directly on the Ethereum blockchain + protocol, and uses ETH (Ethereum protocol token) rather than having any token of its own. This is a great example of a very simple application that lives ONLY at the application layer, relying on Ethereum for the rest of its stack.
Steemit is a popular, and valuable, dApp often referred to as “Reddit for Blockchain”. Like Ethlance, Steem.it appears to end-users as a standard web app, but belies a much different infrastructure than Ethlance. Steem.it is built on its own blockchain, with its own protocol, its own token, and it’s own dApp. Building a complete infrastructure, versus just the application layer, undoubtedly makes Steem a more difficult platform to create than Ethlance, although there are definite advantages to owning the value throughout the entire stack.
The next use-case, Augur, is a highly-publicized “Prediction Market on Blockchain”, worth ~$278M at the time of writing. Augur features a much more complicated blockchain stack than either of the two previous use-cases. In my oversimplified example, Augur leverages Bitcoin’s blockchain and protocol layers, as well as BTC as a token. Augur, though, also involves two additional tokens, REP and internal exchange tokens.
REP is the token that Augur ICO’ed and is most visible to investors, but the use-case for all 3 tokens is outlined in Augur’s whitepaper. Complex blockchain platforms like Augur are much more likely to feature complex infrastructures, which are often needed to build the right incentive, capital, and procedural structures for the platform to function effectively.
The final example, AdEx, is a stack I’ve seen in a number of applications, with a shared public blockchain (like Ethereum or NEO) providing the blockchain and protocol layers, and a dApp with its own token built on top. Sometimes, in these cases, both tokens are needed for the platform to function, whereas in other cases the unique token is used simply for capitalization.
In AdEx’s case, its token (ADX) is used to buy and sell advertising space, making it a functional token rather than one just used for capitalization. ADX was ICO’ed, though, and the AdEx platform is notably the first dApp built on NEO’s blockchain.
So What Does All this Mean?
Blockchain application stacks, at least in their current state of maturity, feature a tremendous amount of variance that might not be apparent to an end-user, casual investor, or blockchain beginner.
The development market for dApps is still at a very nascent stage, and it remains unclear how much value will be captured long-term at the application layer versus the protocol layer.
The whole discussion makes me think of a cover of The Economist from earlier this year:
As long as data is held at the shared protocol layer (Bitcoin, Ethereum, NEO, etc.), it’s safe to assume that protocols will continue to hold significantly more value than their value-less predecessors (TCP/IP, SMTP, HTTPS, etc.).
Decentralized applications, though, are certainly not without value. While dApps remain at a more nascent stage than blockchain protocols, Coinbase is worth $1.6B, Augur is worth $278M, and AdEx is worth $64M.
Historically the value split was essentially 100:0 (applications vs protocols). In the early days of the decentralized web, though, that ratio has reversed and stands at something like 20:80 (this is a guess — I don’t know the cumulative value of all applications).
The big questions I see are:
- Will applications find a way to build value to compete with protocols on the decentralized web?
- If the value stays in protocols, will more apps opt for the Steem.it approach, owning both the protocol and application layers?
- Can apps like Ethlance, built on Ethereum without a token of its own, appreciate over time, or will creating a token be a requisite for creating value?
- Will standard infrastructure stacks develop over time? Or will this variance become the standard for blockchain application stacks?
- Can data-rich companies like Google, Facebook, Amazon, etc. compete with their existing models? Or will they be challenged by decentralize competitors?
I’d love to hear any thoughts on the post above! Reach out to at Daniel Mason with questions / comments. Thanks for reaching and, if you live in Chicago, I’d love to see you at my next Chicago Blockchain Entrepreneurship Meetup.
