Scaling the OG blockchains Ethereum & Bitcoin

BbyA...nMtW
22 May 2023
144

As crypto adoption as continues to grow the diversity of users has grown along with it. Once purely the domain of Layer 1 blockchain token trading the crypto ecosystem has expanded into many different use cases. So much so a new term emerged to better describe this broader church it had become: web3.

With this expansion has come significantly increased demands on the blockchain technologies that underpin the decentralisation and permission-less features web3 embodies.
For Ethereum and Bitcoin this has been a huge problem. Neither chain was originally designed for the transaction volume or cost needed to compete in a crypto market that wasn’t a niche industry.

The eventual solution is of course to evolve the core technologies but with both sets of dev teams stuck working at a pace (because of legacy + decentralisation) far below that of the growing number of new high volume low cost Layer 1 chains they have
turned to separate but connected blockchains as a solution to remain competitive and solve for their scaling issues.




Scaling Ethereum

As the historical epicentre for most of the innovation in crypto (and all of it in the early days) Ethereum has struggled to scale to meet demand. This has allowed a number of competing Layer 1 chains to emerge who are built to serve much larger transaction volume at a much lower cost. Many of which are EVM compatible so are able directly access the value on Ethereum.

Ethereum’s response has been to split its ecosystem into separate layers to concentrate on delivering the volume and cost metrics in order to stay competitive.
The scaling project on Ethereum has been focused on the Layer 2 ecosystem which incorporates a growing number of technologies and designs. Broadly there are two categories: Rollups and Sidechains.



Rollups

Whilst rollups can exist anywhere, they are most prevalent on Ethereum. Rollups are blockchains that use Ethereum validator set for ultimate consensus but achieve volume and cost improvements by sequencing transactions (rolling them up) first before submitting that finalised transaction bundle to Ethereum to validate.
You can’t be a validator on a rollup chain. That’s why you can’t stake token like OP or ARB on their respective Layer 2 chains.
There are two categories of rollups: Zero Knowledge and Optimistic.


Optimistic rollups

Optimistic rollups take a optimistic approach to transaction discrepancies. They have a 7 day wait until they submit a transaction bundle to Ethereum to be finalised. As a user all you see is your transaction confirmed almost instantly. Where you’ll better see the underlying mechanism at play is bridging assets from Ethereum to an optimistic rollup chain. This process can take up to 7 days (but often can be sooner).



Current the Layer 2 rollup ecosystem is pretty much Optimism and Arbitrum in terms of users and protocols. Each are similar but approach the optimistic design in their own ways. Important to note is a key security aspect for optimistic rollups is fraud proofs. These are transaction checks that are done regularly at the Layer 2 level in order to catch fraudulent activity. Neither Arbitrum nor Optimism, despite being in production, have these live on their blockchains yet (though both have them in their roadmap to deploy very soon).



Zero Knowledge rollups

Zero Knowledge rollups do what optimistic rollups do only much faster. They use what are called zero knowledge proofs to confirm and condense transactions to then submit to Ethereum.


Recently there’s been a lot of excitement around zero knowledge rollups. Polygon (zkEVM) and Matter Labs (zkSync) have both released their zero knowledge rollups into testnets very recently. You’d be forgiven for thinking this was the first time we’re seeing this technology deployed on top of the Ethereum blockchain. But it’s not.

Loopring is a zero knowledge rollup that’s been live since Dec 2017. It’s never really seen much traction, despite using zero knowledge tech.



Immutable X is another zero knowledge rollup that’s been live since April 2021. Its focus has been entirely on web3 gaming via providing high performance NFT marketplaces. It has acted as both infrastructure, incubator and investor for its stable of web3 games with some games being developed in-house by the Immutable teams.
It’s also not really seen that large a take up from users, but in this case that can be partly explained by how long the development cycles have proved to be for web3 AAA gaming.



Recently Immutable announced they would be partnering with Polygon to deploy non NFT marketplace gaming use cases onto their new zkEVM blockchain.

It’s a bit of a headscratcher as to why so many people are so excited about the most recent zero knowledge rollup chains that have launched given, to date, the performance of this technology has been very lacklustre when compared to optimistic rollups (Arbitrum now regularly processes more daily transactions that Ethereum).

We’re yet to see if these new zero knowledge chains actually will bring a different value proposition to the space and avoid the failure of the existing incumbents to capture market share.




Sidechains

Sidechains are scaling Layer 2 solutions that borrow parts of Ethereum security but do not rely on it for consensus. They instead have their own validator set. This has led to higher throughput and lower transaction costs than rollups.



Polygon POS is most well known and most successful sidechain on Ethereum today. Polygon POS uses Ethereum as a data availability layer and to host staking assets of its validators. Sidechains have their own consensus security i.e., they don’t rely on Ethereum consensus. That’s why you can stake MATIC as its a part of consensus mechanism on Polygon and also why you require it (and not ETH) to pay for transaction fees.



Sidechains do not inherit the full Ethereum security (as rollups do) so they can be less secure if they have a smaller validator set than Ethereum.
Ronin is another popular sidechain of Ethereum created by the company Sky Mavis to facilitate transactions for their popular game, Axie Infinity. Ronin also has its own native token, RON, that's used to pay for transaction fees and participate in governance on the network. The Ronin chain was hacked for $622 million dollars highlighting the dangers of the security compromises sidechains make for speed and cost improvements.





(reluctantly) Scaling Bitcoin

The OG blockchain, Bitcoin, has seen it’s own scaling solutions start to emerge in recent times. As more people come into crypto there is a growing desire to be able to use BTC in more complex financial transactions and tools.

The issue is Bitcoin blockchain technology has traditionally struggled to accommodate added functionality (“struggled” being an understatement. Bitcoin has largely, deliberately, ignored smart contracts so far).

Up until very recently, and still mostly to this day, the best way to access greater functionality was to wrap Bitcoin in an EVM compatible contract and essentially remove it (and its' transaction fees) from the Bitcoin network.
Not a great solution for the longevity of the Bitcoin blockchain as miners see no return for providing the security that made wrapping that Bitcoin so attractive in the first place.
Thus scaling solutions have emerged as a solution to this problem.
‌



‌The original Bitcoin scaling solution

The Lightning Network runs on top of the Bitcoin network, which means that it relies on the security of the underlying Bitcoin blockchain to keep it safe. When someone wants to start using the Lightning Network, they need to lock up some Bitcoin in a payment channel. This Bitcoin acts as collateral and helps ensure that both parties in the payment channel will act honestly. Furthermore, each person has their own password controlling their funds, known as a private key, which is necessary to keep safe. With the Lightning Network, users can make super-fast payments while still benefiting from the security of the Bitcoin network.

The Lightning Network has been around for ages, much longer than Ethereum Layer 2s, so why hasn’t it taken off in terms of defi and NFTs like Ethereum’s Layer 2s have?

The reason is it’s not really an expansion of the underlying Bitcoin technology, it just makes it faster and cheaper to use your BTC.

Given the significant issues smart contract Bitcoin Layer 2 chain Stacks (see below) continues to have with speed we may well start to see future scaling solutions built on top of the Lightning Network in order to achieve cost and speed metrics that would make them competitive with other smart contract enabled chains.


‌


Smart Contracts guaranteed by Bitcoin security

Stacks protocol uses Bitcoin for its security by rolling up its transactions and anchoring them to the Bitcoin network.

Stacks is a smart contract blockchain that allows developers to build decentralized applications on top of the Bitcoin network.

To achieve this, Stacks uses a "Proof of Transfer" (PoX) consensus mechanism that requires miners to send Bitcoin to the Stacks blockchain to mine transactions.
In return, the miners receive STX tokens, which are used for governance and utility purposes within the Stacks network. By committing Bitcoin to verify transactions, participating miners contribute to the security of the network while remaining incentivized to maintain its integrity.



Stacks continues to struggle with providing the speed to properly enable the kinds of applications the vast majority of existing crypto users. It has seen a influx of users in recent times driven by both the renewed interest in Bitcoin brought about by new use cases like Ordinals and by the work the Stacks ecosystem has done in developing bridging technology and a truly cross-chain permission-less wrapped Bitcoin: sBTC (not to be confused with Synthetix’s sBTC, a synthetic Bitcoin price pegged token).



Sovereign Rollups: Bitcoin’s “sidechain” solution

A Sovereign Rollup is a scaling solution for Layer-1 blockchains that allows batching of transactions that can be "rolled up" off-chain into a single on-chain transaction. This significantly reduces the cost and increases transaction speeds on the base chain. A sovereign rollup does not require a a settlement layer for validation (like Arbitrum needs Ethereum to achieve finality), and thus offers "sovereignty" like a layer-1 blockchain. This is achieved by verifying small chunks of data on the base chain to ensure data availability. In many ways this is the balance of security vs cost & speed we see from sidechains on Ethereum.

Sovereign rollups on Bitcoin make sense on Bitcoin, in a way they don’t as much on Ethereum, as it’s incredibly hard to fully integrate with the Bitcoin blockchain given the state of it’s technology and the pace at which it has been able to modernise.




The most “side” of sidechains: Bitcoin timestamps

‌Babylon has an interesting take on leveraging Bitcoin security, without relying on Bitcoin for consensus. This, as we saw with sovereign rollups, like sidechains in Ethereum as it’s not using the consensus mechanism. In this case it’s a very little integration as its just using the Bitcoin timestamp.

The advantage here is the flexibility that gives any chain to integrate with this aspect of Bitcoin security.

There are significant downsides to this as a scaling solution though.

  • There's little upside for the Bitcoin miners (who secure the chain)
  • It doesn't actually scale BTC perse as it’s not contributing much in the way of activity on the Bitcoin network (unlike a chain that is leveraging a lot more of Bitcoin security and thus the block space).
  • See Primary Usecase: Securing PoS Chains | Babylon Blockchain for more information




Sidenote: Scaling the validators

Whilst not an OG Layer 1, the Cosmos ecosystem is unique competitor in this space as it’s taken the modular approach that scaling solutions represent at their core but applied that to an entire ecosystem of independent Layer 1 chains that share fundamental interoperability standards.

The reason I mention this here is this has run into issues scaling, not from a transaction volume or cost stand point (new Cosmos chain Sei may actually be the fastest, cheapest blockchain to date), but from a security scaling stand point.
How do you get enough quality validators to secure a new Layer 1? It’s a hard problem to solve.

Cosmos has taken an interesting approach to this particular scaling problem with their interchain security (also known as replicated security).

How it works is Cosmos Hub, the chain secured by ATOM stakers and the largest chain in the Cosmos ecosystem, shares their validator network with another Cosmos chain. Cosmos chains have always been able to easily move tokens between chains due to their shared technology standards but that process was only as secure as the weakest partner (chain) in a transaction.

Added to this every new Cosmos needed to spin up it's own validator set (as all Cosmos chains are Layer 1s). This made it hard for smaller, community driven, projects to get up and running. By leveraging the large fleet of Cosmos Hub validators it makes it much easier to launch and scale a Cosmos chain.

The caveat is you'll need to integrate with the Cosmos Hub validators so removing some flexibility in your blockchain design (a flexibility that is the main reason people build in the Cosmos ecosystem).





Why even bother scaling “old” technology?

With all the ecosystem excitement around the various scaling solutions for Ethereum and Bitcoin, the question has to be asked: Why?

  • Why do we need all this complexity to provide scaling?
  • Why can’t the Layer 1 chains, Ethereum and Bitcoin just scale?


Solana, Fantom, Aptos etc.. these are all Layer 1s that are scaling their chains now to the speeds and costs that the bulk of new users will need if they’re to be onboarded to crypto.

So why are we focusing so much time and energy on building fixes (Layer 2s, and in some cases Layer 3s) for core issues on the Layer 1s?
I don’t know the answer, but I have two broad theories why.



Security

There is a widely held assumption that Bitcoin and Ethereum offer the best security out of all blockchains. Ethereum offers the best POS security and Bitcoin the best POW.
I don’t have enough technical knowledge to explain each of these arguments in detail, but having immersed myself in both sets of arguments I think that this is true if you want it to be true.

There is are perfectly sound argument to be made, for certain high value transactions, in small enough numbers, for either Layer 1 being the best option security wise.

However just as credit card security isn’t “the best” but it’s been good enough to enable trillions of dollars of transactions and underpin a large part of the global economy I think there are a lot of “good enough” crypto use cases where either Bitcoin or Ethereum security is overkill. And over-priced.


Money

As cynical as this sounds I think money is the main reason why we’ve seen so much energy devoted to scaling the original Layer 1s, over just building and expanding new Layer 1s.

It is becoming fun to watch experts say in one sentence that we are only at the very early stages of crypto adoption and we’re yet to see the 99.9% of users arrive and then in the next sentence say that the only place to build is either Bitcoin or Ethereum as that’s where all the users and money are. Even very knowledgeable people can find it hard to see past the nose on their face.

Bringing non-crypto users into crypto is hard. Crypto has been dominated by technical people for the longest time. There is a dearth of the marketing and user experience skills in crypto needed to onboard users to these very different ecosystems and product than they’re used to.

It has been, traditionally, much easier for technical people (which crypto has mostly been) to stay in their comfort zone and focus on building tech for existing users they don’t need to explain stuff to. In that sort of climate it makes logical sense to just go whether ever the current user base is and build there.

It’s why we still see so many NFT projects and protocols on Ethereum despite it being a horrible user experience (sky high fees) for the vast majority of NFT users.





A UX that isn’t sustainable long term

At the heart of both Ethereum and Bitcoin lies a user experience which is (currently) pretty bad.


Ethereum costs way too much

$10-150 per transaction on the Ethereum mainnet is not going to fly with the general public.
$0.1-0.9 per transaction on Layer 2s won’t either if you want to serve any use case other than what credit card payment providers do now (and even then that might be too high).
It definitely won’t serve the web2 transaction use cases we see done at scale for tiny amounts of money i.e., transactions in games or on social media platforms.

Maybe some of these work when we go to Layer 3, but now we’re at 3 layers. Going up and down those 3 ladders is complicated and will be expensive. It’s also likely going to be time consuming. What if users ever need to go back down through Ethereum? If you’re a kid with an $80 account on some gaming platform and you get it wiped out by gas fees because you weren’t paying attention that won’t be a good time. Have that happen a few thousand times and you’ll see users stay away in droves.


Bitcoin is way too slow

Wait times of over 30 mins (often hours) on Bitcoin will not fly either.

ALEX is currently the most popular DEX on Stacks (I have funds staked there) and it recently saw a large influx users chasing the airdrops that ALEX advertised. Those users were not ready for Stacks. I’ve been watching it unfold in the Discord.

Now things haven’t been helped by other issues (Stacks hard fork and popular wallet Xverse having some significant bugs) but mostly the transaction wait times are driving people crazy. ALEX tells you there are wait times of between 10-30 mins but that is nearly always incorrect. I’ve used Stacks for months and you’ll regularly see times go into hours. If Bitcoin gets congested it goes slow. Like glacial slow. I’ve occasionally had to wait overnight for a transactions to go through.

There are very narrow use cases (share trading or large asset ownership) which can be still be served even with long wait times, but those are already well served in the non blockchain world and those non blockchain providers offer a bunch of security guarantees that the vast majority of people transferring high value items value far greater than permission-less and decentralised technology.




Where the buck is going

There’s a saying plenty quote (often misattributed to Wayne Gretzky, but in fact it was his father, Walter): “Skate to where the puck is going, not where it has been.”

Easy to say. Very hard to do.
Every instinct will tell you: “Look! There’s the puck. Just go there.”

I liken where we are now in crypto to when Nokia ruled the early smartphone phone world. The prevailing wisdom then was you needed to build for Nokia as that’s where the users were, and it wasn’t viable to build for any other ecosystem unless you had niche product.
In hindsight the smart play was to realise the limitations of the Nokia tech (which were obvious even at the time) and instead spend your time building for things like Palm. Palm and similar phones held a tiny portion of the market but were clearly better tech and the smartphone market was small overall.

The tech that people like Palm pioneered morphed into the smartphone tech we know and love today and Nokia is now merely a brand asset that doesn’t represent any of the underlying tech it once dominated so completely.

That all said very few actually built for that Palm tech. It was the “smarter“ play to stick with the market incumbents, even though the overall market was many times smaller than the market it would eventually become. Even though many experts at correctly predicted the smartphone market would become much larger overall, the prevailing wisdom was: “Go to where the users are now.”
Sound familiar?

My opinion is that old blockchain tech, like Bitcoin and Ethereum, is the past. That it is dominant today is like saying Nokia dominated the pre iPhone smartphone market.
If you believe that crypto is going to be magnitudes larger in market size than it is today, then it’s not going there with the tech that came to define the market when it was only thousands active users. The tech that will dominate in the next phase needs to serve first millions and then billions of users. Probably 100’s of billions of bots too. Most of which will probably be legit use cases of AI technology.

It may not be Solana, Sui, Fantom or NEAR that solves for those use cases. But those types of chains are the design patterns that will win.
If you’re building and thinking for those type of apps then you’re moving to where the puck is headed. Not to where it is now.




Don’t throw the baby out with the bathwater

Just because I may think that Ethereum and Bitcoin aren’t the ultimate future of crypto doesn’t mean there isn’t a lot to learn from both in the present. Especially this is the case when it come to their respective scaling solutions.

Optimism and Arbitrum are both home to a lot of cutting edge defi innovation. Many of the ideas being refined on both these chains today will, in my opinion, be the leading designs and technologies in the near future in crypto. I’m invested directly in multiple projects (and part of the Discords) on both chains.

Similarly I think the problem Stacks is trying to solve is applicable to other areas in the future. Trying to securely scale a technologically rigid system of high value assets sounds suspiciously like the problem that many traditional financial assets will face when moving on chain. Transaction speed will not be an issue for plenty of these asset classes but security definitely will be. That’s the reason I’m invested in the ALEX project and are part of their Discord.

And maybe I’m wrong about Ethereum and Bitcoin. Maybe they can eventually solve the technical challenges holding them back. For that reason I hold exposure to both chains, especially Ethereum who birthed defi as we know it today and is still home to some of the most influential projects in crypto today.


You can have some cake

Cakes On Solana
Enjoy this blog? Subscribe to andrew-

14 Comments