Ethereum is Centralized - Twitter Live Debate


#1

Hello TruStorians!

To continue to experiment with the live Twitter debate format, we’ll be revisiting the “Ethereum is Centralized” claim. Preethi and I would love to get your input and involvement as we want this debate to be as community driven as possible. We’ll cover familiar ground from the first round, and would very much love to address any new points that we might have overlooked.

Below is a summary in bullet form from last time. Please do share any new angles/perspectives you have.

Main Claim: Ethereum is Centralized.

Sub-claim: Ethereum is reliant on a handful of private companies to survive

Back:

  • Assume you’re talking about funding of Ethereum infrastructure development (Eth 2.0, 1.x roadmap)
  • If you make this argument you’re focusing on: Ethereum Foundation, Consensys,

Challenge:

  • Ethereum ecosystem has out grown handful of private companies
  • Ethereum Community Fund – nonprofit initiative to fund critical infrastructure, developer tooling (8 founding members);
  • Enterprise Ethereum Alliance – member-driven standards organization – businesses and leaders collaborate to create decentralized web to benefit everyone
  • Community Initiatives :
    • Ethhub.io – completely independent, open sourced;
    • Ethereum Community Conferences
    • Community Ethereum Development Conference

Sub-claim: Block reward cuts can be agreed seemingly without objection

Back:

Challenge:

  • Rough consensus takes time
  • There are multiple layers before decisions get made.
  • EIP process is fairly arduous (too arduous?)
  • Block reward cut for Constantinople: EIP 858, EIP 1234, EIP 1295
  • Block reward cuts always always in the roadmap (e.g., mining difficulty adjustment since Metropolis 2017)

Sub-claim: Tokens likely remain in a few hands

Back:

  • Ethereum’s pre-mine results in the concentration of Ether in a few hands -> influence over supply / ability to crash currency (still unsure of this argument)

Challenge:

  • Ethereum Foundation Grants – Waves 1 – 5, roughly 14 million USD dispersed, and ongoing.

Sub-claim: Influence seems to be concentrated in few hands

Back:

  • Assumption: influence concentrated with ‘tokens held in few hands’

Challenge:

  • Again EIP: open, transparent, difficult to manipulate
  • Core Dev calls, recorded and shared
  • POS (CBC) cartel-resistant, discourage centralization

Sub-claim: Infura is an indication that Ethereum is centralized

Back:

  • Well acknowledged in the community. This is a concern.

Challenge :

Sub-claim: People can’t run their own full node because Ethereum’s blockchain is bloated

Back:

  • Generally true

Challenge:

Questions from the Community:

  1. How active is the community when it comes to making these decisions?
  • There is an active Core Developer and Developer community
  • Not much from a wider non-technical audience
  1. How much tokens would a group of high stake holders need to have to have a big influence?
  • Great question! Not sure having a high number of tokens would translate to influence whereas if Ethereum were to do on-chain coinvoting, that would be a different story, but I don’t think Ethereum is going in that direction.
  1. How thorough is the EIP process?
    Here’s a graphic (source: Dan Finlay: https://twitter.com/danfinlay/status/991521043939504129)

  1. Is there anything being done to remove/solve the barriers to making it easier to use a full node for an average user?

Yes, here are some initiatives:

  1. Since you guys dissected this topic, what essential elements do you think need to be looked after if Ethereum wants to stay decentralized as it was intended to be?

Some Ideas:

  • Progress on Ethereum Roadmap (i.e., Sharding, POS Casper, Scalability)
  • Adoption
  • Education
  • Governance that seeks to prevent concentration of power
  • Price (? Can we reason that higher price -> greater economic incentives for people to be validators -> more decentralized & higher security?)
  • Maybe everyone might not be familiar with how Ethereum started and what it was meant for initially, can we get a brief background?
  • Built on the weakness of Bitcoin’s limited scripting language
  • Wanted to be more than just currency
  • Ambition to be general purpose platform in which other things – a stable currency (Dai) could be built on top
  • Here’s a video from Devcon2 (2016) where Vitalik explains Ethereum in 25 minutes – provides a nice overview of what it was meant for initially https://www.youtube.com/watch?v=66SaEDzlmP4&feature=youtu.be|

#2
  • Decentralization vs. Centralization is on a spectrum
  • There is no silver bullet answer to determining whether something is centralized or decentralized
  • Instead, we can look at various aspects that make something decentralized and measure decentralize component by component

#3

Balaji S. Srinivasan and Leland Lee has published an amazing article trying to quantify decentralization for BTC and ETH. Their methodology is interesting:

They first use Gini Coefficient (Lorenz Curve) to measure coefficients across 6 sub-systems they defined;

*from article

And then they look for ways to evaluate the coefficients, for which they propose Minimum Nakamoto Coefficient – the number of entities we need to compromise in order to compromise the system as a whole.

BTC:

ETH:

Comparison:

Regarding the usage of Minimum Nakamoto Coefficient, Balaji gave some examples:

[For example, if one considers “founder and spokesman” an essential subsystem, then the minimum Nakamoto coefficient for Ethereum would trivially be 1, as the compromise of Vitalik Buterin would compromise Ethereum.]

[Conversely, if one considers “number of distinct countries with substantial mining capacity” an essential subsystem, then the minimum Nakamoto coefficient for Bitcoin would again be 1, as the compromise of China (in the sense of a Chinese government crackdown on mining) would result in >51% of mining being compromised.]

It’s good to quantify decentralization as Balaji himself mentioned about 3 benefits: Measurement and tracking, Improvement identification of sub-systems, Optimization using objective functions

I think a constructive approach to build on this forward would be:

  1. Define the goals of decentralization when used as a means
  2. Ponder if there’re additional sub-systems other than the 6.
  3. Understand which sub-systems are more critical that could cause the entire system compromising, or severely harms decentralization. Different tokens may also have different interpretation on the coefficients. For example, Bitcoin may value “Mining subsystem” more because mining power consolidation poses 51% attack threat. Ethereum, as a world decentralized computer, may value “Client” and “Developer” subsystems more.

#4

Some amazing insights in here which articulate the differences in decentralization between Bitcoin and Ethereum:


#5

agree wholeheartedly.


#6

Thank you for the resources @eddyso ! More stuff for @preethi and I to discuss :slight_smile: