Skip to content
Exploring how cryptoeconomic mechanisms could help solve coordination problems in social settings.
Python Rust
Branch: master
Clone or download
Pull request Compare This branch is 29 commits ahead of vbuterin:master.
Fetching latest commit…
Cannot retrieve the latest commit at this time.
Permalink
Type Name Latest commit message Commit time
Failed to load latest commit information.
CTMPD.py
QVACG.py
README.md
charity_sim.py
charity_sim_v2.py
pos_hashcash_signalling_basic.rs
pos_hashcash_signalling_keynesian.rs
pos_hashcash_signalling_population_models.rs

README.md

QVACG: quadratic voting anti centralization gadget

  • https://ethresear.ch/t/quadratic-costs-and-collective-payouts-as-anti-centralization-gadget/2429
  • imposing quadratic costs may put pressure on the players of a game to move towards a maximally effecient equilibrium
  • the rules that impose that quadratic cost define the eqiulibrium points players move to, and thus the degree of centralization/decentralization
  • quadratic costs/benefits can be maximally effective in iterative games that are rolled out indefinitely (incentivizing participants to move to the equilibrium immediately as there no benefit in resisting)

CTMPD: charity through marginal price discrimination

  • original idea: https://vitalik.ca/jekyll/general/2017/03/11/a_note_on_charity.html
  • it seems like in a setting like Charity Through Marginal Price Discrimination the system works in isolation, but would be maximally effective in a setting with multiple participants all putting pressure on each other to opt into a shared mechanism.
  • moving towards a mechanism that can test and compare arbitrary populations
  • and then evolve populations to over iterated games (simulating people finding value in a mechanism and adopting it) Also considering how this could be implimented in the real world, and if the same mechanism might inspire cartell/monopoly type behavior if used for marketing/promotion/upsell mechanisms rather than community building. Attacks and defenses TBD.

PoS HashCash Signalling


Future Work

To really roll these out, we need to model both honest and malicious populations to determine up to what threshold attacks will fail or become profitable. This requires:

  • functions that can generate arbitrary populations with various metrics/parameters
  • functions that can iterate over those populations and test for thresholds of success/failure
  • functions that can iterate over populations over the course of repeat interactions over time and test for thresholds of success/failure

You can’t perform that action at this time.