Skip to content
This repository has been archived by the owner on Jun 15, 2021. It is now read-only.

Commit

Permalink
Update research.md (#18)
Browse files Browse the repository at this point in the history
  • Loading branch information
orishim authored and tspoff committed Jun 21, 2019
1 parent 01ecf93 commit 97dee9c
Showing 1 changed file with 2 additions and 4 deletions.
6 changes: 2 additions & 4 deletions research.md
Expand Up @@ -17,11 +17,9 @@ An overview of the features one would conceivably want a bonding curve to suppor

### Core Curve Features

- *Curve Function*: Linear, Polynomial, Logarithmic, etc.
- (More complex math and computational costs demands a reason imo - I'd like to see some solid reasoning for one curve or another.) [This article discusses](https://medium.com/thoughtchains/on-single-bonding-curves-for-continuous-token-models-a167f5ffef89)
- Ideally you would be able to make your own equation of arbitrary complexity
- *Curve Function*: Linear, Polynomial, Logarithmic, etc. [See thorough discussion of trade-offs here](https://medium.com/thoughtchains/on-single-bonding-curves-for-continuous-token-models-a167f5ffef89)
- *ReserveRatio*: % relationship between buy and sell prices
- *Spread/Split* (separate buy and sell curves)**:** Perhaps someone would want more control here
- *Spread/Split* (separate buy and sell curves)**:** Difference between buy and sell curve or amount that goes to beneficiary

### Financial Features

Expand Down

0 comments on commit 97dee9c

Please sign in to comment.