The Appchain thesis has grown into a counter-narrative to Ethereum’s multi-chain future. Especially with the development of Cosmos 2.0 and dYdX igniting the movement by moving from the Ethereum layer 2 solution Starknet to Cosmos. I’m going to write this article under the assumption that you are familiar with Cosmos and if that is not the case you can read about it here.
So what is the Appchain Thesis?
Appchain Thesis
The appchain thesis is that as every protocol built on a blockchain wants to scale, they will seek to deploy as sovereign chains that would give them full control of their protocol without relying on an inherent L1. Why would you want to do that? Imagine that you are a DeFi protocol on Solana that is trying to maintain your operations but the network constantly experiences downtimes hindering your business in the process.
Note: There was no way Solana wasn’t going to catch a stray bullet. Sorry, not sorry.
This would make you look for other options that provide more resilience. Protocols that opt for the sovereign route would be in full control of:
Maintaining their own consensus
Security (could be exchanged for Interchain Security aka ICC)
Validator set
Uptime
Instead of being dependent on the underlying L1 running smoothly, they would acquire full accountability for running their sovereign blockchain. This can both be a blessing and a curse so let’s dive into the pros and cons as I do my best to provide a fair case for both sides.
Arguments For The Appchain Thesis
Sovereignty is Strength
One of the drums that get banged against the appchain thesis is that there is a loss of composability. However, an appchain believer would argue that full composability creates a cascading effect that falls like a domino throughout the system, leading to more damage. The thesis is built on the insufficiency of composability as it will make sense for some DeFi protocols to become sovereign (and for others not so much). A DeFi protocol with a lack of demand and users would most likely take a different route after watching the Dotsama ecosystem. We’ve seen instances where the sovereignty has been working perfectly, such as the Terra debacle.
Benefits Interoperability
The Appchain thesis can be argued as misconstrued as it can be bullish for anything that uses a generalized messaging system as it facilitates interoperability. Essentially any relevant data can be sent from one chain to the next one. This would also imply that the facilitators of this infrastructure will be among the main parties benefitting from the Appchain thesis. Some of you will be able to read between the lines after that sentence and I will get more to that further down in the article.
More Room For Speculation
One of the reasons why L1s have amassed such astronomical value is that they have incredible ecosystems built on top of them, which also makes them hard to value from a financial standpoint. This leaves room for more speculation and potential growth. If we compare this to standard DeFi applications (which probably will port over like dYdX or build from scratch) they are much easier to give a rough estimate of what a fair valuation would be. From there more rational consumers would appear that wouldn’t buy the token at a premium (this might be utopian though because we know how irrational the market is).
Further Token Utility
The Appchain thesis also provides further utilities for tokens as an app’s native token now would be used for gas fees to pay validators that secure the network. This would create an additional sink which currently is not apparent for most protocols as they are experiencing an influx of tokens instead due to incentives leading to selling pressure.
MEV Resistance and No Rent
MEV (Miner Extractable Value) resistance is another core benefit of appchains as it makes it much harder to perform MEV when you have a customizable validator set where you can ensure forced behavior. We’ve already seen instances of this in the Cosmos Hub where transactions forcefully have to take place in batches limiting the ability to perform MEV (batching forces the miners to execute transactions irrelevant to the transaction order).
Applications pay for gas on a typical L1 (you can view it as rent) for fighting for the same block space as other applications. As you are already aware by now, in times of high demand this causes gas prices to spike and turns into a negative experience for the end user. This would not take place on an appchain as they are in control of their network.
Avoid Regulatory Hurdles
Lastly, one of the strongest arguments for appchains is that they can be perceived as more decentralized. The reason for this is that if you draw a comparison of protocols deployed on an L2 such as Arbitrum that have a centralized sequencer, that is a potential choking point that becomes a liability when Gary Gensler tries to protect you. An appchain with countless validators becomes more decentralized as it’s difficult to identify an actor to target which is why protocols that are afraid to be branded as securities might seek refuge in this manner.
These are the cases I currently have for appchains, let’s look at the counterargument.
Arguments Against The Appchain Thesis
Lack of Composability
The Appchain thesis is premeditated on the fact that an application doesn't need composability and other apps become superfluous as the market will gravitate towards the winners. You could view this differently. Composability has been one of the main drivers of DeFi’s growth and along with that you already have sidechains that build towards specific applications that don’t inherit Ethereum’s security. Moreover, protocols have successfully deployed on L2s that are more scalable while being secured by Ethereum as well.
Applications are isolating themselves by fragmenting liquidity and losing the composability that has been working for them so well. Liquidity fragmentation does not necessarily become a problem if the ecosystem works well but in reality, UX isn’t great and there are some security concerns to consider as well.
Security Issues
The Cosmos SDK can be considered clunky to build on in comparison to the EVM. Applications being responsible for their own security becomes a major risk. Sticking to teams with proven experience with the Cosmos SDK seems like a safer bet than risking punting money on inexperienced devs that will entail smart contract risk. Obviously, interchain security is a way to mitigate this so that’s one to look out for.
Also, there is no denying that bridges have been proven to be dangerous. While bridges have enabled consumers to use assets on different chains, it has also decreased security considering the ridiculous amount of hacks that have taken place in the industry. If we want the crypto industry to be taken seriously, the value lost in hacks needs to be decreased. The bridges are only as secure as the developers make them. This doesn’t mean that the IBC is deemed insecure. However, some heavy investigation took place after the BNB incident, and a vulnerability that would affect all IBC-enabled chains was found. Cudos to the Cosmos team for finding this before it did any more damage.
The Cosmos Go Module (used to implement IBC) was audited in January 2021 by Informal Systems and while I don’t doubt their capability as they are much smarter than me, it would be nice to see an independent party audit the module. Informal Systems didn’t build the module but they directly benefit from it as their business is built on top of it. This becomes a conflict of interest.
Cannibalization within the ecosystem
The cannibalization within the ecosystem is mainly a detriment to the ATOM token. Multiple protocols are trying to acquire the different forms of value accrual available in the ecosystem as the Cosmos Hub has been a slow mover in regard to their implementation. This is not necessarily bad as it can be deemed as making value available for anybody instead of being a gatekeeper that keeps it in-house. Each to their own. The main one to keep an eye on as a direct competitor to the new ATOM 2.0 roadmap is:
Celestia - Celestia is a modular blockchain that allows other chains to connect to their existing consensus instead of requiring them to create their own consensus mechanism. This provides shared security between these chains. The value proposition is that it will be easy to deploy on Celestia in comparison to using the IBC and ICC. While also being able to scale along with the number of users that join their network through their data availability sampling that solely downloads random samples of data from blocks instead of the whole block —> proves that the whole block is available and verifiable. Blocks are able to grow in size with the number of users (light nodes) as the base layer doesn’t need to download the whole block and thus does not require constantly increasing hardware requirements. If you want to learn more about Celestia, RainandCoffee has a great article about it here.
Complexity for developers
One of the main reasons why competitors such as Celestia have appeared is because of the clunkiness of the Cosmos SDK it becomes a larger task for developers to run and maintain their own chains on Cosmos in comparison to solely launching on Ethereum. Clunkiness and isolation probably might not be a good idea for most apps that need to benefit from the convenience of composability.
Also, the current Ethereum roadmap is already building towards what could be deemed appchains and roll-ups that prioritize this. IMX is an example of this and Arbitrum Nova is as well. Nonetheless, competition is healthy for the growth of the industry as end-users benefit from quality products.
So who benefits from this?
The obvious and boring answer here would be the ATOM token as they intend to launch interchain security that would allow a better value capture mechanism, but there is more to it. When I argued for the appchains I mentioned that it benefits interoperability and some of you probably have probably figured out that Synapse might be worth looking at. The team is probably licking their lips in case a cross-chain future solidifies itself as it is poised to capitalize better than any other bridging protocol.
Synapse already started building Synapse chain, which is preparing a cross-chain messaging system that securely sends any arbitrary data between chains. It will provide an interface that will facilitate the developer’s process of building cross-chain use cases. Sounds quite convenient, doesn’t it? Especially considering what the current landscape looks like.
Synapse also contains the widest bridging infrastructure outside of IBC (as IBC is within the Cosmos ecosystem) considering it enables you to bridge between 18 different L1s/L2s.
Key Takeaways
The appchain thesis is currently the main counter-thesis that makes sense from a blockchain standpoint. While you have other blockchains such as Solana and Avalanche, they are also proof-of-stake blockchains that don’t provide significant differentiation apart from speed. You could argue that they are optimizing for a problem they don’t even have regarding users.
The Cosmos ecosystem offers something different and still has a long way to go despite impressive developments that have taken place so far. However, there is no doubt that the infrastructure being built provides a net good for the space as it doesn’t really matter what chain applications are residing on from an end-user standpoint. People will use it as long as it’s easy to use and can attract demand. So be it on Cosmos or Ethereum.
However, I would have some precautions with protocols that don’t utilize interchain security when it gets released. Especially if they don’t have a lot of proven experience with the Cosmos SDK as it leaves room for vulnerability.
Lastly, this is the first iteration of cross-chain infrastructure and there is no doubt in my mind that it will become more seamless as time goes on.
What’s your view on a cross-chain future?
Well done if you managed to congest all that information, I hope you enjoyed the post. Don’t forget that you are more than welcome to leave feedback or drop any questions in the comment section.
Sharing this post helps it reach out to more readers which would be great. If you are interested in future posts from me, do not hesitate to subscribe. Thanks for taking the time!
Disclaimer: All Content on this site is information of a general nature and does not address the circumstances of any particular individual or entity. Nothing on the site constitutes professional and/or financial advice, nor does any information on the site constitute a comprehensive or complete statement of the matters discussed or the law relating thereto. I am just a random degenerate sensei sharing an opinion.
Nice article. Out of all the benefits, high performance like high throughput with low latency can be crucial for dApps in choosing appchain approach. In the meantime, app-specific rollups can be a better replacement of appchains, to provide better derived security and composability. You can check out StarkNet L3 thesis and Altlayer's flash layer for references.
nice article until it mentions Synapse