Vitalik Buterin:What is in Ethereum 2.0?


整理自 youtube 视频 Devcon4 Keynote - Vitalik Buterin

What is in Ethereum 2.0?

  • PoS (Casper)
  • scalability(sharding)
  • VM improvements(EWASM)
  • improvements to cross-contract logic(eg.abstraction)
  • improvements to protocol economics

2014:the long road begins

slasher:a punitive proof of stake algorithm
slasher is a useful construct to have in our war chest in case proof of stake mining becomes substantially more popular or a compelling reason is provided to switch.

abunch of aborted ideas…
“proof of proof of work” (2014 summer)
hub-and-spoke chains(2014 summer)
hypercube chains(2014 summer)

but some progress…
weak subjectivity
certainty that algorithms with much stronger properties than existing PoS are possible
an understanding that with PoS scalability can be done through random sampling...somehow

and a roadmap

ether sale-frontier-homestead-metropolis-serenity

and an increasingly resolute commitment to proof of stake

2015-16:winter
a series of abortive attempts at solving the core problems,while research on casper quietly begins…

consensus by bet

with long readmes…and PoCs

first attempts at “rent”

the scalability paper(notes on scalable blockchain protocols)

featuring...complicated escalation games

featuring...cops and robbers analogies

featuring...deep state reversions
the fundamental problem:fisherman’s dilemma

but even still,we were on our way…

meanwhile...EWASM quietly marches forward

the industrious age
minimal slashing conditions( feb 2017)
casper FFG(fall 2017)
casper CBC(2017-18)
Erasure coding data availability(initial idea spring 2017)

casper FFG paper

new sharding design

casper CBC paper

casper FFG PoC

new beacon chain spec

meanwhile…
LMD and IMD GHOST fork choice rules
VDF research
more ideas on abstraction
rent...now renamed to “storage maintenance fees” for political correctness reasons

and research
synchronous cross-shard transantions with consolidated concurrency control and consensus (or how i rediscovered chain fibers)
cross-shard contract yanking
DRAFT:position paper on resource pricing

and development

soon...serenity begins.

serenity is …

  • a realization of casper(pure proof of stake),sharding, EWASM and other ideas from protocol research between 2014-2018
  • a new blockchain,though with a link to the existing PoW chain
  • “the world computer”as it’s really meant to be
  • still decentralized-and more than today

expected phases

  • 0:beacon chain PoS
  • 1:shards as data chains
  • 2:enable state transitions(EWASM)
  • 3 and beyond:iterate,improve,add tech

expected features

  • pure PoS consensus
  • faster time to synchronous confirmation(8-16 seconds)
  • “economic finality”(10-20 minutes)
  • fast VM execution via EWASM
  • ~1000x higher scalability

Post-serenity innovation

  • layer 2 execution engines(privacy,cross-shard transactions,“semi-private chains”)
  • proof of stake improvements(casper CBC when ready?)
  • upgrade to STARKs(signature aggregation,erasure code correctness(for data availability checks),state execution correctness)
  • stronger base-layer cross-shard tracsantions?
  • even faster single-confirmations?

medium-term goals

  • eventually stabilize at least the functionality of layer 1
  • think about issuance,fees…
  • agree on the guarantees that people expect from the protocol
  • think about governance…

what is immediately next?

  • stabilize protocol spec
  • continue development and testing
  • cross-client testnets
  • security audits
  • launch