Eth2.0 Implementers Call 10 Agenda

[djrtwo]

Eth2.0 Implementers Call 10 Agenda

Meeting Date/Time: Thursday 2018/1/17 at 14:00 GMT

Meeting Duration 1.5 hours

YouTube Live Stream Link

Agenda

1. Client Updates
2. Research Updates
3. PoW -> PoS ether transfers
4. Test formats discussion
   • ssz
   • tree hashing
5. Spec discussion
6. Open Discussion/Closing Remarks

[AlexeyAkhunov]

I would like to discuss the details of moving Ether from PoW chain to PoS chain, because it affects impact analysis for for the change of PoW. There are 3 main things:

1. PoS finality gadget on PoW chain, which is based on the beacon chain
2. Tapering off PoW rewards gradually once finality gadget is in place
3. Risk of beacon chain validators stopping other validators to emerge (by not coming to a consensus on accepting further deposits from PoW chain)
4. Censorship resistance of deposits (the idea described below)
5. Reorg attack resistance of deposits (also the idea described below)

an idea on the simply way to make censoring of deposits much harder by PoW miners. The beacon chain needs needs to allow counterfactual deposits. It means that I have an address A with some ETH on it on PoW chain. I choose a salt, and generate a CREATE2 contract address based on that salt, some fixed factory address, and some standard init_code, which creates a contract that can only self-destruct and burn (properly burn, by self-destructing to itself) ETH. I send 32 ETH to that address, that to everybody who does not know my salt looks just like any other address. Once the acceptance threshold is reached (i waited for 1024 or 2048 blocks), I reveal my salt to the beacon chain. It simply checks that CREATE2 address == sha3(0xff | factory_address | salt | sha3(init_code)) and accepts my deposit, without requiring me to actually instantiate the contract and burn my ETH - because nothing else can be done with it anyway
so my deposit transaction on PoW will be indistinguishable from just sending ETH between two non-contract addresses. And miner will have to censor all such things (which means pretty much halting the chain) to stop those
it also allows me to wait the time of MY choosing before I reveal my deposit to the beacon chain. If I want to be sure that my deposit will not be reverted by a deep reorg, I will wait sufficiently long
which makes deep reorgs attacks much less of a threat to PoS stake depositing

[5chdn]

I would like to discuss the details of moving Ether from PoW chain to PoS chain, because it affects impact analysis for for the change of PoW. There are 3 main things:

1. PoS finality gadget on PoW chain, which is based on the beacon chain
2. Tapering off PoW rewards gradually once finality gadget is in place

Thanks for raising this. If we assume a Serenity mainnet is ready for launch in (early?) 2020, we have to discuss transition details on both sides. I would also add this to the Eth 1.x agenda once we are done with Constantinople upgrade and the ProgPoW debate.

[jannikluhn]

Could we briefly discuss the test file formats for SSZ and tree hashing defined here and here in the next call?