2019-11-21 Eth2 implementers call

2019-11-21 Eth2 impl call.txt

Nimbus updates

core / Testnet:

- Cleaned up leftovers from 0.8.x spec: shard, crosslink, …

- We have our weekly testnet setup in place:
  rebooted on Tuesday morning and stable for a week.

- 0.9.1 compatriot, waiting for 0.9.2
- Compatibility with zcli 0.9.1 genesis next Tuesday

- Complex sync issue to debug
=> working on better logs as it stresses our setup

BLS:
- benchmarked on Raspberry Pi Herumi (without OpenSSL GMP):
  21 ms for pairing

TODO
- Fuzzing effort -> ncli

Networking - Libp2p
- Working on native Nimbus discv5

Eth1
- Finished Istanbul hard fork implementation
- Go wrapper for switching part of Status client to Nimbus instead of Geth

————

https://github.com/ethereum/eth2.0-pm/issues/101

———— 


Testing updates

Sigma Prime: grant for beacon-fuzz -> ramping up efforts.

Namespace clash between Zrnt and Prysm
Working with Lighthouse, Trinity, Zrnt at the moment

Next couple days: Nimbus and Prysm

Fork choice test were not prioritise so far
update soon

—————

Client updates

Prysm: Terence

testnet relaunched with v0.9
55 peers only a couple from Prysm

Same config as minimal config
Danny: update the shared repo so that it’s easier to test


Lighthouse: Paul
Validator on boarding
Hot-Cold database
Testnet ready but not publicised yet, pending db
large refactor of validator clients
—> want to beat Nim
—> BLS implementation with Herumi

Lodestar: Marin
- 3 BLS EIPs implemented in JS
- Discv5 huge progress
- 0.9 spec done

Felix: Q on Sybil resistance to the DHT
—> Adrian: limit IP addresses per subnet

Nimbus updates

Parity: Wei
—> revive the Substrate runtime
New name of the runtime “Place”

Trinity: Alex
Passing all the 0.9.1
Started on attestation aggregation
Perf improvement on PySSZ

Artemis: Joseph
Split in 2 teams:
Research: Harmony + part of Pegasys
Production Readiness: Artemis
—> working on sync
—> finalizing aggregation
—> 0.9.1 next
Looking into Docker images of other clients
—> Discv5

Harmony: Mikhail
Progress on GossipSub simulation
Lot of time spent on merging teams

———————
Research updates

Vitalik
-
phase 1:
lots of work by Proto on plugging it in the testing
Discussion on data availability: fraud proof or stark proof based?
Hash research for SSZ: binary vs prime field

Youtube chat Question on BLS standardisation

Carl
-
BLS standardisation presented at IETF meeting
No issue raised which is one of the precondition for it to be a blockchain standard
Last call: submitted to CFRG for crypto, ~6 months to go.
—> for blockchain standard we can start now since no reservations were raised.
Next internal blockchain meeting in 2 weeks time

Justin
-
EF is looking for someone to manage the validators
Need someone with devops and operational security skills to setup
Potentially multi-client, complementary with Coinbase.

Phase 1, Dan Boneh: reelection mechanism
Issue with a hash function for data availability, would take years to validate

Will
-
Scout to test multiple shards
Smart contracts EE
Q Mamy: original smart contract in Rust/Assemblyscript
—> can target it from Nim
https://medium.com/@william.j.villanueva/ethereum-2-0-phase-2-progress-7673b57eabff

Danny: new calls
Eth1x, organised on ethresear.ch forum
Research on Eth1, and transition to Eth2
Phase 2 community call: https://github.com/ethereum/eth2.0-pm/issues/103
on Tuesday 2019/12/03
Probably monthly networking call, would like a member of each team
Light client task force: https://github.com/ChainSafe/lodestar/issues/555
Wed dec 4, 3PM UTC

———————
Networking updates

Felix: no update, was on Geth

Harmony simulation: publication by the end of next week

——————————
signing_root
—> Proto proposed to remove it for full hashTreeRoot
Feedback wanted: https://github.com/ethereum/eth2.0-specs/issues/1487

—————
End Q:
weak subjectivity period