In the beginning was Bitcoin, the only blockchain we thought we’d need. Then Ethereum came along and launched us into a multi-blockchain world. Since then we’ve seen many new blockchains emerge with their various different protocols, financed by their various different tokens.
One of the the key doctrines of this era was Joel Monegro’s seminal “fat protocols” post, in which he explained that value creation in a protocol world happens in a very different way than in the Internet world. Specifically, in the Internet world, the protocols (think TCP/IP and HTTP) were conduits and not holders of value; it was the applications built on top of them that held the value. In the blockchain world, he argued, it’s the protocols that hold the value.
His post, in some ways, helped provide the fuel and justification for the protocol/ICO excitement we have witnessed over the past 6-8 months.
But the explosion of protocol ICOs (as well as ICOs that claim to be protocols) and the “fat protocols doctrine” have accelerated the timetable for a new layer of innovation for the decentralized economy. This new layer has been on the minds of many in the industry for quite some time, but on the proverbial backburner in terms of broader awareness. That’s now beginning to change.
The layer? Blockchain interoperability.
Fat protocols, thin protocols, and interoperability
Monegro’s core thesis is now coming under scrutiny. Teemu Paivinen argues in his recently posted essay on “Thin Protocols” that “[while] protocols in aggregate will continue to capture most of the value, individual protocols will in fact be quite thin and tend towards capturing minimal value, due to the combined effects of forking and competitive market forces.”
Meanwhile, blockchain blogger Evan van Ness wrote that “there’s no such thing as ‘fat protocols’” — for a few reasons. First off, he writes, Monegro doesn’t clarify what is an app and what is a protocol. Second, he points to examples blockchain platforms like Gnosis and Steem and reasonably asks, are they apps? Protocols? Or both?
Van Ness then refers us to a post by Stephan Tual, a leader within the Ethereum community who now is at Slock.it (and who has as more first-hand experience with decentralized innovation than most, since he was at ground zero of “DAO-gate”). Tual challenges conventional wisdom further by sharing his Web 3.0 Abstracted Stack.
You do not have to be a seasoned blockchain architect to see that these are not, in fact, fat protocols. Rather, they are a series of thin component protocols that work in harmony to deliver decentralized applications.
Paivinen, Tual, and van Ness are far more technical and knowledgeable than I am, but the basic gist as I read it is as follows:
The combination of the speed of innovation within the decentralized world, the ability to fork open source projects at will, and the economic incentive to serve more niches as the overall market opportunity grows is going to lead to an explosion of blockchains.
We are already seeing this occur. Given the strengths of each team and their vision, each blockchain-based protocol is going to do something really, really well, but no one blockchain can do EVERYTHING really well.
Bitcoin has a strong trust layer. Storj, Sia, and FileCoin have created strong storage layers. Replicating those strengths would take a lot of time and become redundant, and there is the well known adage “don’t roll your own crypto,” when it comes to security.
As a blockchain developer, you want to be able to take the best of everything and put it together so you can build the best application possible. That’s why interoperability is critical and why the market for it is starting to heat up.
The multi-blockchain world has created a lot of value and excitement, but moving between them can be challenging. If you own Numeraire currency, for example, and you want to get Bitcoin for it, you have to use a tool like ShapeShift or sell it on an exchange. You can’t just send Numeraire to your Bitcoin address without, at the moment, going through someone else. (Well, you can, sort of through something called cross-chain atomic swaps, but it’s still immature. Still, Decred and Litecoin pulled it off recently).
Another example is identity. Say you want to log into trading platform EtherDelta to trade your tokens. You’ll need a MetaMask extension to verify your identity. But what if you want to use the Civic verifier instead? You can’t. On the other hand, if you want to log into Prism, you need Civic and can’t use MetaMask. You need two “identity wallets.” Frustrating.
As an end user, you can use whatever identity provider you want, and the decentralized app doesn’t have to code for 10 different log-in possibilities. That’s value.
The cross-chain protocols and their approach
We’re seeing the first generation of interoperability players emerge to take on the task of intermediating between chains. Make no mistake about it, we are really early in this phase of development.
While it will take time for this segment to see maturity, what should get you excited is the fact that there are numerous projects out there aggressively making their case for becoming an interoperability standard. All of the innovation happening now may be a bit murky, but the arrival of hardened technology is going to enable public blockchains to work with each other and for private-public blockchain bridges to work better, while affording each organization the flexibility to choose the set of blockchain protocols best suited to its needs.
One of the key players in this cross-chain space, Polkadot, has a roadmap that puts its own genesis block in 2019. In the meantime, it has already raised $140 million to make it possible, so unless the team is out ordering caviar and Dom Perignon every day, they should have enough cash to deliver.
ICON is another big name in this field. The Korea-based group closed a $42 million token sale in just six hours in September.
Other players include Cosmos (which calls itself the “Internet of Blockchains”) and new entrants like Aion, Lamden, Metronome (focused on cross-blockchain currency), and a totally open protocol put forth originally by Ripple called the InterLedger Protocol.
If you want to get really deep on the tech side of cross-chain protocols, check out Jackson Palmer’s great explanatory video. He does a really nice job. Using Polkadot as an example, here’s roughly how they work:
Polkadot has three main components.
- A relay chain to coordinate consensus and transactions between separate chains
- Parachains — these are the constituents (let’s call them the chains that have “bought into” the interoperability solution offered)
- Bridges, which are links to blockchains, like Ethereum, that have their own consensus mechanism
Such a system requires the same type of game theory and mechanism design that a regular blockchain would and needs to account for ways to:
- Cryptographically secure a new block to the chain
- Ensure members of the network are honest
- Uncover and remove dishonest actors
It is not too difficult to foresee an academic or professional discipline that studies “blockchain interoperability protocol design” and certifies that such protocols function as they should. We are definitely going to need that.
When will we see the first cross-chain protocols?
My crystal ball is as unclear as anyone else’s in this industry, but the one thing that consistently surprises those of us in the middle of it all is the pace of innovation.
If the Polkadot roadmap is any indication, we are about 18 months out from the first deployment. That gives all of us plenty of time to think critically about what the components are of a solid interoperability standard and which one is most likely to emerge victorious.
We also have time to begin developing plans for even more robust, dynamic, and powerful decentralized applications that leverage the “best of the best.”
The lag time until we see that first deployment also provides adequate warning to the establishment that decentralized technologies are maturing and, once bolstered by interoperability, are going to get closer and closer to mainstream use cases and adoption.