Another good space today! 👍
Thanks Alex
@BeamCTO and Max
@maxnflaxl.
These are my notes about what was said!
Alex started with an announcement:
📢 Beam will be present to the
#CryptoXR exhibition and conference (
cryptoxr.fr), on Jan 31st and Feb 1st, in Auxerre, France. We didn't get a speaker slot, but we will have a booth, and DB
@dbadol2000 will be there to welcome anyone interested in Beam!
Also, Alex reminded that if anybody has questions or ideas for the next Space, please do tell him. It can be any subject about Beam, about crypto or about privacy in general.
There were two subjects to discuss today: First some comments on privacy tech. And then some of the dilemmas and decisions we have about the upcoming Beam L2.
1⃣ About Privacy tech
- The question of the day is: What is an "Anonymity Set"?
- The best anonymity happens when we are in the middle of a large group of people behaving the same way as us, and doing the same kind of things. The larger the group, the strongest the anonymity.
- That's what a mixer does: it tries to confuse the outside viewers about who is doing what.
- We can imagine many people in a room, some want to send money, others want to receive money. If there are only 2, 3 or 4 people in the room, it will be easy to guess who is doing what. But with more people, it becomes more difficult.
- But that's not all. Even if there are many people, if you are the only one who is always there, while the other people change constantly, then you will finish by standing out.
- This "side effect" idea is very important in crypto, security and privacy.
- That's why we need not only a large anonymity set, but a large "effective" anonymity set.
- Different coins are using different methods to increase the anonymity set, and hence, ensure the users' privacy.
- And to make it simple, we can consider two types of privacy: The privacy of values, and the privacy of identity.
- We then see the following, by instance:
1) Beam transactions = privacy of values AND privacy of identity.
2) Beam smart contracts = visible values BUT privacy of identity.
3) Secret Network (TEE-concealed smart contracts) = privacy of values BUT visible identity.
- ZCash uses a type of Zero Knowledge Proof, and is probably the best theoretical privacy tech.
- Indeed, the theoretical anonymity set of ZCash is infinite. Because when you use a UTXO you prove it's one of ANY of ALL the existing UTXOs!
- However, it's not that straightforward. Because not all UTXOs are spent in the same fashion...
- By instance, some user might usually spend their UTXOs in one or two days, while another user will keep them untouched for years. The "effective" anonymity set of the former user is much lower because to monitor them, we only need to monitor the UTXOs that "move fast".
- Lelantus works in a similar fashion, with the UTXOs in a "Shielded Pool". The difference is that in Lelantus, the size of the pool is not infinite but limited to 64k.
- To limit the "side effect" aspect in Lelantus, Beam Wallet tries to automatically use the oldest shielded UTXOs rather than the newest ones, in order to increase the "effective" anonymity set of the user.
- Monero, on the other hand, uses a specific and constant number of decoys in each transactions (16 today).
- It's fixed but it's enough and it works very well. All the more so that Monero is the most used privacy coin today (which is good as the particular behavior of the user could still appear through the decoys if the overall usage of the coin was low).
- In any case, whatever the tech, if you leak info about your transactions (by instance by switching from transparent to private transactions, like in ZCash or LTC MWEB) you will reduce your effective anonymity set.
- That's why Beam has privacy transactions ONLY.
Now, to the second topic of the day:
2⃣ Decisions about our future L2
- The objective of the Beam L2 is to have a faster Beam alternative (e.g. 1 second block instead of 1 minute).
- To allow that, we are going to change the consensus mechanism of the L2 to Proof of Stake (PoS).
- Many people don't like PoS because, contrary to PoW, you only need money to make... more money. You don't need to spend any other resource.
- And indeed, PoS usually creates a situation where the founders hold a very large share of the coins and of the power.
- There is also a problem of "value". Because, in a standalone PoW, the value of the coin is linked to the cost of the resources spent to mine the coins (e.g. the cost of electricity). In PoS, this basis does not exists.
- An example of a weakness: Imagine we allow bridging NFTs from ETH into Beam. We can then imagine a very expensive NFT coming into Beam, and an attack on Beam blockchain not being expensive enough to protect that NFT.
- The same situation can happen between our L1 and L2, since we want to allow easy exchanges between the two! So we need to be sure that the L2 is not too weak compared to the L1...
- For this Beam L2 PoS, we have 3 options: Not use any coin, use BeamX or use a brand new coin.
1) Without any new coin, fees will be the only reward to the validators. If the fees are enough, and the cost of being a validator is low, it can work.
- In a few months we will have a proof of concept L2 to play with and we will be able to see how much it's going to cost to be a validator.
2) With a new coin, which cannot be Beam (because its emission schedule on the L1 is already set in stone), we could decide to use BeamX.
- Indeed, we have quite a lot of BeamX available. When created, there was a share planned for investors that was not used, and we also have some planned reserve for liquidity and ecosystem incentives.
- So the DAO could decide to take some of these coins to reward the L2 validators. And it would make sense because the L2 is a DAO product, and all the applications in L1 will be quickly ported into the L2 and will profit from it.
3) The last option is to create a new coin (codename "BeamW") that would be minted on the L2. Such a scheme would make much sense if it allows bringing investors and external funding, but we are not going to do that. So we will need another mechanism to define the price of this new coin (which will then also allow justifying being a validator).
Some last interesting questions that have been raised:
a) Who else is building an L2 for their own L1? Do we know any other example?
b) If we have a fast L2, why would we use the L1? We need to think about that...
c) Can we keep the L1 relevant without artificially crippling the L2 (by instance if we were to only allow creation of tokens on the L1)?
Many questions, much dilemas... but a very interesting subject and very exciting times! 🤩