[Pre-proposal Discussion and Draft] Ethereum Scalability Layers Index (L2 Index)

1

UPDATE : Draft for a “L2 Index” proposal :

CONTEXT

In the light of the CryptoKitties mania back in 2017, and the more recent DeFi and yield farming craze during Summer 2020, it becomes obvious that Ethereum’s 15 transactions/second limit cannot keep up with its popularity.
From a few cents per transactions, high gas prices resulted in transactions prices peaking up to 15 USD equivalent for a single token transfer and much more for complex operations.
Moreover, if current average gas prices allow basic transactions to be mined for one or two dollars, it can be useful to remember that in a lot of developping countries the monthly income is below 500 USD equivalent.
This situation makes Ethereum only accessible to the wealthier people, and only accessible to the “whales” in times of congestion.

Although the ETH2 Phase 0 release is imminent, it comes with some misconceptions :

  • Timing : ETH2 - the one everybody is waiting for, the one that brings scaling - is for late 2021 or 2022. ETH2 phase 0 is “only” a beacon chain, a parallel chain able to “watch” the ETH1 chain, but not able to interact with it.
  • Nature : Although the switch from PoW to PoS might itself increase the transactions per second, in ETH2 most of the scaling is envisioned horizontally, through the concept of shards orchestrated by the Beacon Chain. Put simply, each shard will be it’s own chain so each validator can focus on a specific shard or set of shards. Cross-shard composability (Compound on one shard, Uniswap on another shard) is yet another challenge that will add some complexity.

Ethereum Scalability Layers - commonly referred as Layer 2 or sometimes Layer 1.5 - allow vertical scaling on Ethereum. Although they can feature an additional consensus mechanism, they all share the following properties :

  • They allow more transactions per second.
  • Funds are still secured by Ethereum’s L1 chain.

Why Ethereum Scalability Layers :

  • They are here already : Some are already developped and used, others are in active development.
  • They are here to stay : As they don’t replace Ethereum’s L1 but live ontop of it, they’ll have the ability to migrate to live ontop of ETH2 in the future.
    A bonus point is that some of those scalability layers can also act as a layer of abstraction, requiring to rethink dApp logic and code today, but facilitating migration to non-EVM L1s in the future (ETH2 eWASM or any so-called ETH-killer).

PROPOSAL

It is proposed the creation of an Ethereum Scalability Layers Index. The goal is to have an asset that will see its price appreciating as different Ethereum Scalability Layers are being developed and adopted.
To fullfill that purpose, it is proposed that the index will include tokens that are directly linked to projects whose main goals is to develop Scalability Layers on Ethereum.
Different approaches, specifications and implementations exist. Some of the most famous are State Channels, Plasma, Zk-Rollups and Optimistic Rollups. However projects might implement their own version or a combination of those, as long as finality is achieved on Ethereum mainnet.

SELECTION CRITERIA

It is proposed the following criteria for token selection :

  • The project must be developing a Scalability Layer that inherits from the Ethereum mainchain security guarantees against loss of funds while allowing higher transaction throughput.
  • The project must be in active development.
  • The project must feature a ERC20 token with both enough market cap, liquidity and distribution, ensuring a price coherence accross platforms. Those concepts being subjective, numbers can be adjusted throught formal or informal governance.

SELECTED PROJECTS

OMG Network (OMG) :

  • Website : https://omg.network/
  • Intro : OMG Network uses More Viable Plasma to enable fast and cheap ETH and ERC20 transfers. The vision is to capitalize on this transfer layer to enable cheap payments in Asia and the rest of world. OMG is currently used by some exchanges as a cheaper alternative to withdraw ERC20 tokens.
  • L2 type : More Viable Plasma
  • Coingecko : https://www.coingecko.com/en/coins/omg-network
  • Token ERC20 address : 0xd26114cd6ee289accf82350c8d8487fedb8a0c07

Loopring (LRC) :

  • Website : https://loopring.org/
  • Intro : Loopring uses zkRollup to provide Loopring Exchange (orderbook-based ETH and ERC20 exchange) and Loopring Pay (instant ETH and ERC20 transfers) on https://loopring.io/. An AMM exchange is also being developed. The vision is to improve performance of financial operations on Ethereum with almost no compromise on security, data availabiliy or fund accessibility.
  • L2 type : Zk-Rollups
  • Coingecko : https://www.coingecko.com/en/coins/loopring
  • Token ERC20 address : 0xbbbbca6a901c926f240b89eacb641d8aec7aeafd

Matic Network (MATIC) :

  • Website : https://matic.network/
  • Intro : Matic provides an EVM-compatible sidechain that capitalizes on the Ethereum mainchain by publishing regular checkpoints via Plasma. The vision is to provide a scalability layer with very little friction for both developers and users : dApps that work on mainnet can be moved/copied on Matic, and using those dApps is as simple as adding a Custom Network on your Metamask.
  • L2 type : EVM Sidechain + More Viable Plasma
  • Coingecko : https://www.coingecko.com/en/coins/matic-network
  • Token ERC20 address : 0x7d1afa7b718fb893db30a3abc0cfc608aacfebb0

Celer Network (CELR) :

  • Website : https://www.celer.network/
  • Intro : Celer Network created State Guardian Network (SGN), a multi-purpose “watchtower” network that is used as a security and availability layer for their State Channel network for transfers/payments, but also Celer Rollup (Optimistic Rollup) for dApps. The vision is to build a complete set of interoperable L2 solutions that use the SGN, each of those solutions having different compromises and being optimal for specific use case needs.
  • L2 types : State Channels, Rollups
  • Coingecko : https://www.coingecko.com/en/coins/celer-network
  • Token ERC20 address : 0x4f9254c83eb525f9fcf346490bbb3ed28a81c667

Raiden Network (RDN) :

  • Website : https://raiden.network/
  • Intro : Raiden Network uses State Channels to enable cheap and near-instant payments. The vision is to be the “Lightning Network” of Ethereum, enabling micropayments at a scale. Raiden’s routing system allows to operate off-chain whenever there are intermediate channels (even non-direct) between the sender and the receiver.
  • L2 type : State Channels
  • Coingecko : https://www.coingecko.com/en/coins/raiden-network
  • Token ERC20 address : 0x255aa6df07540cb5d3d297f0d0d4d84cb52bc8e6

WEIGHTING

A market cap + volume dominance weight analysis would result in the following :

image

It is proposed an initial manual weighting that rebalances the asset by rounding down the big caps to give more weight to the small caps :
- 50 % OMG
- 30 % LRC
- 12 % MATIC
- 5 % CELR
- 3 % RDN

image

LIMITATIONS of a “Ethereum Scalability Layers Index” (ESI) :

  • Price of those tokens might not reflect the performance of a given project. It can be impacted by tokenomics (token distribution, utility of the token), and future decisions of the team protocol, token utility, money printing/burning policy. In any case should they be considered as money or currency, nor equity. Although this is true for most ERC20 tokens today, it is important to keep in mind that very few L2 projects actually need a token to function properly (as opposed to L1 projects). In fact, a lot of intersting L2 solutions are being built without the introduction of a token and are following a different business model (StarkEx, Aztec, Optimism, ZkSync, …). Unfortunately, that also means that a “L2 Index” cannot capitalize on the success of those projects.
  • L2 development is evolving quickly, with different technical solutions being tested, and sometimes adopted. Moreover, the topic is often technical, thus maintaining an “Ethereun Scalability Index” might be challenging.
  • On the practical level, it might be more difficult to mint new $ESI because of the low liquidity of those tokens on Uniswap (except LRC). The tokens will likely needed to be minted “manually”, involving Centralized Exchanges, or involve slippage.
  • Also because there aren’t that many projects working on L2’s and featuring a token, 80% of the weight of the index corresponds to only two projects, which reduces diversification.

PROJECTS NOT SELECTED (along with reasons) :

  • 0x (ZRX) : Orderbooks are shared off-chain (via Relayers or the more recent 0x Mesh), but trades are executed directly on-chain. Hence, there is technically no increase in transactions per second.
  • DeversiFi (NEC) : Using StarkEx L2 implementation but not involved* in its development, very low number of holders. * StarkEx is also used in other projects such as dYdX.
  • xDai (STAKE) : EVM-compatible sidechain. Horizontal scaling but doesn’t benefit from any Ethereum security mechanism (no Plasma implementation).
  • SKALE (SKL) : EVM-compatible sidechains network. Horizontal scaling but doesn’t benefit from any Ethereum security mechanism (no Plasma implementation). No circulating token yet.
  • Loomx (LOOM) : Recently pivoted to Enterprise Data. Up-to-date information harder to acquire.
  • Liquidity Network (LQD) : Project allowing to set up centralized and non-custodial hubs inside which zero-fees payments can be made. Market cap (0.5M) and volume are too low.

SOURCES

NOTE : This proposal is a draft. Feel free to propose changes or discuss about the points you are not in line with.

ORIGINAL POST

This goal of this discussion is to determine which projects should be included in a potential future Ethereum Scalability Layers Index (aka L2 Index).
Ethereum Scalability Layers are being created to help solve the congestion and gas fees problem on Ethereum.

  • Each of these solutions either solve an application specific problem like transfers/payment/exchange to make this specific usage more efficient (ex: OMG, Loopring, Deverisifi).
  • BUT some solutions are targeted to be multi-application and interoperable (ex: Optimism with their OVM).

Those technologies are very promising as they represent the real short-term scalability solution for Ethereum. It means that there will be adoption of some of those. It means that short-term might become medium-term (network effect) and eventually long term. In short, Scalability Layers are very promising.

However, for building an index, several group of questions must be answered :

  • Which projects must be considered as Scalability Layers Projects ? This is very technical and few actually understand.
    Examples : Loopring uses zk-rollups, which means it runs directly on the Ethereum main chain. Optimism’s OVM also runs directly on the Ethereum main chain (but with less interaction with it ?). OMG uses Plasma, which uses side-chains and uses the main chain as an arbitration layer. Matic uses a different version of Plasma along with Tendermint (?). xDAI seems to be mostly a side-chain and I cannot find information about it using the main chain for security. Does that mean it should not be considered as a L2 ?

  • How to be exposed to the technology ? zk-rollups and optimistic-rollups are two examples of promising technologies that have an adoption potential on Ethereum. However those technologies rely of the Ethereum main chain and they don’t need specific tokens. Two months ago one could have said that investing in Loopring was a good solution to be exposed to zk-rollups. However, (1) Loopring token (LRC) is specific to the Loopring Exchange platform and Loopring Pay application, and (2) more and more projects are planning to adopt zk-rollups (ex recently Curve.fi). Deversifi is also a DEX using zk-rollups through StarkWare, but their token NEC is not as hyped as LRC. Same issues goe for Optimistic Rollups. I personally invested in SNX because I found that they were experimenting with Optimistic Rollups. Today however, Uniswap is also working on a version running on the Optimistic VM. ChainLink as well might be working on something. Those are their own project (for which the token is more or less exposed to the project itself) on the way to adopt L2’s, but they are not L2’s by themselves. Does that mean that they should be put out of an L2 Index, or at the contrary an L2 index should represent the projects that are adopting L2’s ?

I will paste multiple polls below to try to get your opinion. You can tick multiple answers for each one.
But none of this is truly binary so I’d like to get a discussion started. Don’t hesitate to comment if I’ve been inaccurate or to add more options to the polls

What can be qualified as a Scalability Layer on Ethereum
    • Side-chains that use the main chain for security (ex: plasma-like).
    • Side-chains that use only their own consensus and are not directly connected to Ethereum (ex: xDAI).
    • Rollup technologies (ex: zk-rollups, optimistic rollups)

0 voters

How to get exposure to Scalability Layers on Ethereum
  • L2’s base tokens (ex: OMG, Matic, STAKE)
  • Tokens of applications those main value is adopting L2’s (LRC, NEC)
  • Tokens of existing applications planning to adopt L2’s soon (UNI, SNX, CRV, and probably a lot of DeFi tokens)

0 voters

2 Likes
2

Interesting one this, but I’ll explain why I voted for everything.

I try to think about it in terms of the idea that the index is expressing. In this case it’s that scaling is beneficial to the Ethereum ecosystem and the benefits will be reflected in the price of the tokens that enable it.

To select one solution over another, or one type of token over another feels like trying to pick winners, which in my view is the opposite of what holding an index aims to achieve. I think if you set a broad criteria based on some future outcome, then a rising tide will lift all boats.

Beyond that there may be some technical reasons to include/exclude some projects. It is also interesting to consider at what point you include a project that announces adopting a layer 2, but hasn’t yet done so. The index methodologist would need a clear set of criteria for what constitutes a token addition.

4 Likes
3

@neozaru Nice idea.

Strong theme and focused on an area that I think people would like to increase their exposure.

I would not consider users of scale up technologies to be suitable members of the index. I fully expect UNI, SNX, and many others to adopt L2 solutions with different partners and they would drown out the smaller cap focused projects.

A stray thought wonders if some of the ETH protocols would add anything to the set. Rocketpool is the obvious 1, but I suspect there are others. Obviously not L2 scaling layers, but ETH2 will give more TPS.

4

After digging for quite a while, I found that a set of promising L2 projects associated with a token were OMG, Loopring, MATIC, CELR and Raiden.
They are especially interesting in the sense that they all implement different L2 approaches.
I updated the original post to include a draft of a proposal.

This is still a work in progress, and I am still debating the weight methodology, as I feel like bigger caps would have a disproportionate weight. Good thing ? Bad thing ? Something I am missing ? Don’t hesitate to comment below.

1 Like
5

Hey! PMing you about this.

1 Like
6

I am all for sector indices but I think the “limitations” section is the most important here. Granted, my understanding of these tokens is limited so everything I say should be taken with a grain of salt.

Challenge 1: is this an idea that has a limited window of opportunity? Roll-ups + sharding on ETH 2.0 should give us all the scalability we need. In the meantime, projects will just use Optimism and ZK rollups. What’s the potential of this index beyond say 2022?
Challenge 2: In line with the above, given rollups, how big is the market opportunity here?
Challenge 3: Assuming ETH 2.0 runs into issues and gets delayed, is this index going to benefit? Or will projects migrate to Polkadot, Kusama, Near, Cosmos, etc? None of those are ERC-20 so that’s another problem.
Challenge 4: Liquidity. We have run into this issue working on the Metaverse index. The only token with sufficient liquidity here is LRC. Assuming we need to allocate $5 million of capital into the tokens, how do we do it on-chain and how much does it move the market?
Challenge 5: Limited number of tokens in the index. If LRC and OMG are close to 90% of the index, I can just go and buy those two. Realistically, is this a space that will see more projects in the future given rollups and ETH 2.0?
Challenge 6: I’m not an expert on rollups but Vitalik’s recent blog seems to suggest that state channels have certain drawbacks that could make them a narrow solution.

Channels cannot be used to send funds off-chain to people who are not yet participants. Channels cannot be used to represent objects that do not have a clear logical owner (eg. Uniswap). And channels, especially if used to do things more complex than simple recurring payments, require a large amount of capital to be locked up.

Would love to be convinced that most of these challenges can be addressed and would not limit the potential of the proposed index.

7

I am personally less and less convinced of the relevance of such an index, mostly because of the limitations mentioned in the original post :

  • In what do we invest by saying “investing in L2 solutions” ?
    a) Investing in the projects maintaining those solutions : That sounds impossible, since most of those projects don’t have or need a token.
    b) Investing in the projects that are using a L2 or migrating to a L2 : That sounds too broad as - and this is my thesis - most of the projects will end up migrating to an L2 (or multiple ones). I think a significant portion of the projects in the $DPI (DeFi Pulse Index) will be living (or capitalize) on an L2 by the end of 2021.

Of course there are exceptions : MATIC and OMG feature a token. Loopring develops both it’s L2 and the application layer (exchange, AMMs, …).

But too me at this stage :
(a) fails to be exposed to the value created by L2 development : If 90% of the ecosystem ends up running on Starkware, ZkSync and Optimism - and if there is no token or investment channel for those - an $ESI won’t capture 90% of the value.
(b) is too broad and might result in focusing on a specific sector, which other indexes might do very well (and in this specific case, $DPI would be already very close to that imho).

The other limitations mentioned in the original post still applies.

8

Just for clarity, most of these solutions will likely end up launching a token. zkSync is for sure, they have already said that publicly.

Starkware and Optimism? Not as sure about those.

1 Like
9

Hi, I would like to do a follow-up on this topic, it seems to have been kind of dormant. I am convinced that a scalability/layer 2 index for Ethereum would make a great product. Here are a few thoughts, trying to add value to what was previously said here:

technical complexity

  • For the members of the community who are not very comfortable with the layer 2 technologies, I would highly recommend to read a recent article from the Ethereum Foundation https://ethereum.org/en/developers/docs/layer-2-scaling/
  • In the proposal, the technical complexity was presented as a challenge for the index maintenance. I think this technical complexity and lack of understanding of the technologies by lambda investors is actually a very good argument for the existence of such Index. Investors need this index because they are not able to understand and select projects by themselves.

timing

  • The crazy rise of Matic/Polygon got the attention of a lot of investors
  • We can expect that several projects will launch a token this year (zkSync, maybe Optimism after receiving a16s investment)

scope / methodology for token addition

  • One of the main open question presented in the proposal and debated in the comments is regarding the right scope for this index. My opinion is that we should adopt the most simple approach: add only the “L2’s base tokens”, the projects whose main purpose is developing L2 solutions (frameworks, platforms or tools to use/implement the layer 2 solutions), and not the projects using Layer 2 for their product. This is the best way to focus on the core concept and not overlap with DPI. With this approach, Loopring would not be added to the index, despite actively developing a L2 solution, because its core product is an exchange and not a Layer 2 solution. None of the other defi projects would be added, despite their work to implement a layer 2 solution for their product (Uniswap, dydx, …).
  • ETH2.0 staking solutions: as @overanalyser suggested, we could also add staking solutions, like Rocketpool or Lido, but I think these projects in the long term will be more like financial intermediation products than actual scaling solutions, so I think they should not be included
  • Lack of Liquidity on Uniswap: this is a real issue, Raiden for example could not reasonably be included in an index. What I propose here is
    (i) to define a minimum liquidity threshold for adding the token. This threshold can be increased with the increase of the AUM of the index
    (ii) use a uniswap liquidity weighted factor ( maybe more relevant than total volume)

@neozaru @ansteadm @DarkForestCapital @verto0912 @overanalyser would you like to come back discussing on this subject? or was it kind of discarded? if some of you are actually working on it, I would be glad to join. I quit my job in the banking industry 2 weeks ago and Indexcoop is one of the project I want to focus on

2 Likes
10

I think that this type of product will be ideal to have around July or August when there are more L2 assets to support and move forward with.

Right now there’s just a handful that have tokens, and not all of them are actually doing anything live.

We just need to time this appropriately and then we will own this market.

4 Likes
11

Would there be any benefit to broadening this just a bit to include both Ethereum privacy and scaling? Doing so might allow you to add Aztec/zk.money and Tornado.cash (Torn). The common theme is that these and most of the other scaling solutions being discussed rely in some way or another on zkSnarks, which addresses both scalability and privacy. (I’m no expert; just repeating what I’ve heard.)

Personally, I agree that it might be better to wait a bit while still keeping a close eye on this space. Both zkSync and Aztec have all but confirmed they will have tokens, but not when. zkSync’s logically might drop in August, just ahead of zkSync 2.0. And, the Tornado.cash team has a token but have some big things in the works, including a wallet, a v.3 and DeFi integrations. I also agree that it would be better to exclude things built on top of scaling/privacy protocols, and just include the protocols themselves.

I assume some/all of you are Zero Knowledge podcast listeners? There likely will be additional candidates discussed there.

12

Yup, Aztec would be one of the assets to consider. zkSync will be launching a token as well. Tornado cash would not be included in this as they aren’t the infrastructure for L2. Aztec would be, but not Tornado Cash. Tornado Cash would be a product that would most likely launch on zkSync tbh.

I’m pretty active in the zk/L2 space, so I’m constantly keeping an eye out on what’s going on there :wink:

1 Like
13

Re Torn: That makes sense. Thanks for the reply.

1 Like
14

just noticed this in the BanklessDAO: Scaling Eth tokenset - General - Bankless DAO

3 Likes
15

Glad to see that this is being proposed. I suspect Matic’s dominance is pretty large right now if it was purely volume weighted.

Maybe for the zkrollups that don’t have a token themselves then eth could be their proxy as that is what they will presumably pay gas fees in?

16

Most zkRollup implementations will be natively gasless, and rollup operators will swap those assets back into ETH to pay for storing data onchain. Many of them (optimistic rollups and zkRollups) will launch their own
governance tokens.

17

Where does this proposal stand? Is anyone working on this? The market is ripe for a simplified layer 2 index, this product should have a dedicated team and focus.

With a 3+month product pipeline and many priorities, would assume this wouldn’t be off the ground until Q1’22.

Happy to take it forward if open.