Skip to content
Switch branches/tags


The Truebit implementation of the Interactive Coin Offering Protocol by Teutsch et al.. This protocol attempts to solve two key problems in the current method for token salse, an initial coin offering.

  1. When the token sale is capped (i.e. a fixed number of tokens are sold), users that want to participate may not be able to if they don't attempt to buy in immediately. This prevents users from evaluating the sale before deciding to participate, and users sometimes have to take a leap-of-faith due to the fear of missing out.
  2. In an uncapped sale, it is unclear to any participant what percentage of the entire token supply they will actually hold when the sale is over.

Proposition. No token crowdsale satisfies that both:

  1. a fixed amount of currency buys at least a fixed fraction of the total tokens, and
  2. everyone can participate.

The proof-of-concept implementation from this work can be found here.

Directory Structure

This repository build on top of previous attempts to implement the IICO protoccol by Kleros and Modular. The two respective project are included as directories in this project as well (for ease of access).

Truebit Implementation

The Truebit implementation is located in ./truebit-implementation. The basic differences between the Truebit implementation and the Kleros impementation is the use of buckets instead of a linked list structure.

Briefly, the Kleros implementation consists of a linked list of all bids made. Each list nodes is a bid that specifies a maximum cap, the maximum valuation of the sale that this bid is willing to accept. If the the sale valuation exceeds this vaue, all bids whose max. cap is below it are considered inactive. At the end of the sale a cutoff bid is identified. The cutoff bid is the first bid found whose maximum cap falls below the valuation of the sale. The bid is pruned and all bids that lie ahead of the cutoff bid in the list is accepted (and redeemable) and all behind it are no longer valid.


A crowdsale mechanism optimizing price discovery information for buyers



No releases published


No packages published