OmiseGo & Plasma

Notes and reflections on the SF Ethereum meetup on OmiseGo & Plasma

OmiseGo:

Intriguing presentation by Jun Hasegawa (OmiseGO), but they didn’t want to reveal much. What I would have wanted to know is: how are they going to do decentralized crypto-to-fiat? The bar here will be much higher than just decentralized crypto-to-crypto (solved by 0x, Kyber network, etc). For decentralized crypto-to-fiat, they would need to obtain KYC information on their users as well as obtain Money Transmission Licenses (MTLs), at least in the US. I’d have loved to know how is OmiseGo launching this in S.E. Asia (Thailand and Singapore) without KYC and MTL equivalents.

Plasma:

Joseph Poon, founder of Lightning Network and Plasma talked about scaling transaction rates on Ethereum blockchain by several orders of magnitude: O(millions/sec) compared to the current O(10/sec). They will achieve this by building multi-level deep side chains where most transactions are settled. The hash of a transaction from the side chain immediately above is settled onto the main chain. This now opens up the blockchain to fraud attacks. A fraudulent miner can lie about a transaction by publishing the hash of the transaction but not the witness (i.e. the transaction) itself. To detect this fraudulent transaction, the onus now falls on a new group of third parties who submit “fraud proofs” to indicate that a particular transaction hash is fraudulent (is missing some info about a transaction that rolls up to its hash).

Third party fraud-proof originators

What Joey Poon and Vitalik didn’t describe were details of this ecosystem. Who are these third party fraud proof originators? Are these fraud proof originators also responsible for providing some form of insurance (a la’ FDIC) to Ethereum owners?

Plasma brings a reversal and dispute mechanism to Ethereum

In a lot of ways, Plasma is bringing well known reversal and dispute mechanisms from the fiat world to the crypto world. A consumer making a payment using a credit card can reverse it by calling up the card company and getting their funds back. Similarly, if I use Ethereum to make a payment, I can now reverse it by submitting a fraud proof. In the fiat world, the recipient (merchant) bears the risk of paying back the transaction. In Ethereum now, we will have to build a similar reversal mechanism. However, it is not clear how Plasma is going to support the reversal mechanism. My guess is, it will be via some form of a smart contract.

What about confirmation times in Plasma?

only a small number of extra confirmations (to be precise, around two to five) on the faster chain is required to bridge the gap; hence, the 17-second blockchain will likely require ten confirmations (~three minutes) to achieve a similar degree of security under this probabilistic model to six confirmations (~one hour) on the ten-minute blockchain.

Excerpt from: https://blog.ethereum.org/2015/09/14/on-slow-and-fast-block-times/

Another important point that was not brought up is: transaction confirmation period. In Bitcoin (or Ethereum), you wait for 6 (or 10) confirmations to be sure that the transaction is confirmed. This implies a wait time of 1 hr or 3 minutes for Bitcoin and Ethereum respectively. What this confirmation period establishes is that there can be only one blockchain i.e. single source of truth. With Plasma, this deterministic confirmation period is now probabilistic i.e., I can no longer be sure that my transaction went through up until the dispute period. Depending on how the dispute period is defined and how long it is, this can be problematic.