Some thoughts on Beam L2

Some thought on L2
These are just some ideas. Not claiming to be the final best option or whatever. Comments and alternative ideas are welcome.

TLDR

CAs: CAs should have the same ID on both L1 and L2, ensuring they exist on both chains. Preexisting CAs on L1 are assigned to L2 at genesis, with no initial emission until bridged. New assets will be registered on both chains with the same ID, and registration/unregistration will be synchronized.

Validators: Validators must run full nodes for both Beam and L2, and double as Bridge nodes. They stake Beam or BeamX and can receive BeamX incentives over XX+ years. Beam itself could serve as the gas token for L2, with fees distributed among validators. Validators could vote on TX fee levels, creating a market-driven balance between TX fees, demand, Beam price, and node costs.


Full Version

Assumptions:

  • L2 is an extention of the existing L1 system. Ideally, both form an integrated common ecosystem.
  • L1 POW has the higher security and better decentralization. It is considered the leading chain.
  • The Beam coin has to remain the centerpiece of the ecosystem. Any changes to the structure have to benefit first and foremost Beam itself.

Confidential Assets on L1 and L2:

The goal is to maintain a simple, orderly CA ecosystem across both L1 and L2. Each CA on Beam has a unique Asset ID. Adding L2 without special measures introduces the risk of diverging asset IDs between L1 and L2, which will cause confusion. Additionally, some tokens might exist only on one chain but not on the other. Later they might get bridged but with a different ID. Lots of undesirable side-effects which should be mitigated.
Therefore, asset IDs should be synchronized across both chains, ensuring that CAs have the same ID and exist on both chains simultaneously.

Preexisting CAs:

All preexisting CAs on L1 will be assigned to L2 during the genesis block. These CAs will exist on L2, but their initial emission will be zero. Emission on L2 will occur only after bridging from L1.

New CAs After L2 Genesis:

New assets will be created on both L1 and L2 simultaneously with the same Asset ID. This ensures a consistent ecosystem across both chains, with assets being registered (but not emitted) on both chains.

Registering / Unregistering:

To maintain the ability to unregister CAs, this action must be synchronized between L1 and L2. Alternatively, we could deactivate the unregister option entirely.

CA Emission:

  • Option 1: Emit assets only on L1 and bridge to L2. This is simpler and takes advantage of L1’s stronger security.
  • Option 2: Emit assets on both L1 and L2, but emission is restricted to one chain at a time for each asset. Assets will be registered on both chains, but emission can occur only on one. Some contracts on the L2 might require the ability to register and emit new assets, which would make this option more favorable.

Cost of Asset Creation:

The same Beam locking rule should apply on both L1 and L2 to create an asset. The fee is paid once, regardless of where the asset is emitted.

Bridging:

The asset registration process should automatically enable bridging between L1 and L2, with all assets in the Beam ecosystem being bridgeable from the start.

Final thoughts on CAs:

Synchronizing asset IDs across L1 and L2 provides a clear UX benefit and reinforces the idea of L1 and L2 as an interconnected ecosystem.

Validators:

Node Requirements:
Validators must run full nodes for both Beam L1 and L2 to enhance node availability. They may also validate the Bridge between both layers if needed.

Staking / Gas Token / Staking Rewards:

  1. Validators Stake Beam:
    Validators stake Beam, and transaction (TX) fees are paid in Beam. An additional incentive in BeamX may be distributed to validators over a long period of time e.g. 50+ years and with halvings.

  2. Validators Stake BeamX:
    Validators stake BeamX. Otherwise same as above.

Gas Token on and TX fees on L2:
Beam could perfectly serve as token for TX fees on L2.
Generally, one option could be to have validators have some power to influence the level of TX fees. Market forces might a balancing efect on Tx fees vs. Validator costs vs. usage demand.
Ideally, validators should earn enough from gas fees to avoid relying on large treasury emissions.

Important Notes:

  • The BeamX supply is limited, so it cannot be emitted indefinitely. If there is any distribution, it should follow a long-term emission schedule with frequent halvings to ensure sustainability.
  • Increased BeamX emissions could cause selling pressure on the token.
  • Delegated proof of stake could incentivise regular non-whale user to participate in staking and reduce overall selling pressure.
  • L1 / L2 Contract interoperability, also something to consider
  • What makes POS chains successful? The incentive to hold and fast and dirt cheap transactions. How can get those things while keeping Beam at the heart of it?
3 Likes

I feared that the Layer 2 would devalue beam since the L2 would be faster and give the same privacy benefits thus if it had a separate coin then the Beam on L1 would feel inferior. But if it’s the gas token for the L2 then it will increase its value which is good of course.

Staking Beam to receive Beamx would require BeamX to have a use. It’s a governance token that isn’t being used at all thus the incentive feels empty. I feel like people love proof of stake because they can make money by just holding a coin. But if the incentive is Beamx I’m not so sure how many people will care. The main purpose of BeamX would still be to sell it for profits rather than holding it for its purpose or because it might go up in value.

Staking BeamX for Beam would give people a reason to hold BeamX but the real issue is that BeamX needs to be used consistently for something worthwhile. Even Beam for Beam would help people to care but for BeamX incentives there just needs to be a good reason to hold it.

Off topic but I also wanted to say that we all want Beam to succeed but we shouldn’t fake enthusiasm for the sake of wanting it to succeed because this behavior makes the project look sketchy.

1 Like

I understated it makes sense from a logical math point of view but the problem with L2s is the sequential steps you are you imposing on users i.e. buy beam then bridge it back and fourth to L2. It’s just ugly and a cheat and exactly why I never use Lightning.

Whatever happened to the GhostDAG thing?

The “Beam Warp” L2 is the replacement for DAGKnight idea.
This way we achieve high TPS and fast finality without changing the underlying consensus and most importantly the emission schedule.

You have to buy BEAM to do anything on Beam anyways, I don’t see why this would be “ugly and a cheat”. Don’t worry, the UI/UX for the bridging will be very simple.

The L2 is not a Lightning equivalent, we have Laser Beam for that.

Genuinely curious, in which ways is it not equivalent to Lightning?

The insertion of a step, albeit one step of bridging, is the ugly part and unnecessarily if the base chain is fast and scalable.

My view is, L2s steal from the base chain. For Lightning it is stealing from miners. It is, in effect, the creation of a middleman. Any, just philosophical, just curious how it would be different.

Laser Beam only supports BEAM and simple transactions, you can’t do anything else.

Going for a L2 is the logical step after we saw that DAGKnight is impractical for the reasons above.
We deliver on the promise made to improve speed and throughput.


Yes, that is indeed something we have to figure out and we will discuss this in the Space tomorrow. There are a few ways to achieve this.
Specifically the problem of DEX liquidity, is something we’ll have to carefully look into.


That’s not exactly correct, you still have to pay a tx fee on channel close/open.
Laser Beam even less in this regard, since the channels are P2P unlike Bitcoin Lightning where you don’t need a channel with every recipient.

The use case of Laser Beam is different than that of the normal transactions, as the information about intermediate transactions is never broadcasted. The number of off-chain transactions, frequency and amounts are never revealed.

2 Likes

I see. I’d very much like to hear that Spaces but will likely miss it. Hope there is an archive elsewhere e.g. yt

You can listen to the Space on Twitter afterwards too, we haven’t uploaded the recordings to YouTube.

Adding some ideas and comments here:

I had a discussion about the Beam L2 project with a friend who is knowledgeable on the subject. Here are some notes about his remarks.

  1. He thinks that the idea of a non-cannibalistic L2 is very interesting and would be a nice feature of Beam if we manage to do it well.

  2. He believes that we should look into the Optimistic Roll-up approach, which has proven to be quite efficient.

  3. The only close example he thought of for a non-competing L2 was Metis (the L2 in which Vitalik’s mother is involved). It’s an Ethereum L2 rollup whose sequencer is distributed over the L1 validators. That way, not only does it manage to decentralized the sequencer (which is very centralized in all other L2s), but it also ensures that the L2 fees go back the the L1 validators (contrary to other L2 where the sequencer is money grab that does not contribute to the L1).

  4. Thus, he thinks that our L2 should definitely use BEAM as gas. The same Beam that is minted on the L1 (and that would thus be bridged into the L2).

  5. Second, he thinks that the validators of the L2 should be selected among the L1 miners. This would allow the fees from the L2 to reward the same nodes that secure the L1. Also, it would radically reduce the cost of the maintenance of the L2 validators (since the nodes would already be running anyways, and the surplus due to L2 validation would be minimal).

  6. Regarding PoS consensus system, he thinks that Tendermint is the best one (it’s provided by the open-source Cosmos-sdk project).

  7. Lastly, he advices to build the L2 in a way that, later on, we will be able to provide the tools needed for anyone to create their own L2 over Beam, using the same mechanisms. As proven by the Optimism Superchain Stack, creating new L2s is something people want to do (cf. Base, Lyra, Worldchain, Kraken’s Ink, etc.). And if we provide the tools to do it properly (and in a non-competing way to the L1), we might attract some projects that want to have their own fast privacy chain with smart contracts.

2 Likes