Prioritising Societal Change

Starting from the Startup template

Motivation

Joshua Green describes Utilitarianism in Moral Tribes as a Pricing mechanism for different human values. We use monetary prices as a proxy to the value of different objects in society, based on Supply and Demand. We can use Supply and Demand to price Societal Change.

App flow

Goal

We want to compare the value add in different Universes:

• A world with an UBI > 100$ per month vs a world with no UBI
• A pro-choice world vs A pro-life world
• An open borders world ( i.e. no passports ) vs a world with borders ( as we have it now )

To do this we will have users answer questions about the various changes so that we can build estimates of the utility $C_{utility}$ and likelihood $C_{likelihood}$ of any change $C$ .

User Flow

Core Flow

Users will compare the value add of these changes and their likelihood. They will do it by getting a table such as:

Universe $\alpha$	Change	Universe $\beta$
UBI > 100$	1	UBI = 0$
Pro-choice world	2	Pro-life world
Closed borders	3	Open borders

Having seen this table, they are asked 2 questions:

1. Which Universe would you rather live in, $\alpha$ or $\beta$
2. Which of these changes is most likely to change over the next decade: 1, 2 or 3?

Advanced Flows

Users should be able to

1. Add Changes to an Ignore list, so that they are not asked about that change again
2. Add Changes to a preferred list, so that they see it compared with others more often
3. Propose New Changes
4. See the current best changes, in a Top Changes table. This will be based on ranking the changes by the $C_{utility}\times C_{likelihood}$ properties of a change $C$ , described below.

Business Logic

Estimating $C_{likelihood}$

$C_{likelihood}\sim \beta \left(C_p,C_n\right)$ will be described with a Beta distribution , requiring 2 parameters, $p$ and $n$.

Initialisation

$\alpha =\beta =1$

Update

Whenever a user submits a vote, they provide the information that $C_{likelihood}>A_{likelihood}\wedge C_{likelihood}>B_{likelihood}$ for some changes $A$, $B$ and $C$ . The result of this depends on details to be completed on the Beta Distribution note. Until that get’s finished, we will update the values of $A$, $B$, $C$ to $A^{\prime }$, $B^{\prime }$, $C^{\prime }$ via:

$X$	Old $X_p$	Old $X_n$	New $X_p$	Old $X_n$
$A$	$A_p$	$A_n$	$A_p$	$A_n+\frac{C_n}{C_p+C_n}$
$B$	$B_p$	$B_n$	$B_p$	$B_n+\frac{C_n}{C_p+C_n}$
$C$	$C_p$	$C_n$	$C_p+\frac{A_p}{A_p+A_n}$	$C_n$

Estimating $C_{utility}$

$C_{utility}\sim \mathcal{N}\left(C_{\mu },C_{\sigma }\right)$ will be modelled with a Normal distribution, requiring 2 parameters, $\mu$ and $\sigma$.

Initialisation

$\mu =1000,\sigma =100$

Updates

Whenever a user submits a vote, they provide the information that $C_{utility}>B_{utility}+A_{utility}$ or $C_{utility}<B_{utility}+A_{utility}$ .

Database

Tables needed:

1. User
   1. username: str
   2. email: str
2. Change
   1. submitted_by: FK to User
   2. option_a: str
   3. option_b: str
   4. details: str
   5. likelihood_alpha: float
   6. likelihood_beta: float
   7. utility_mean: float
   8. utility_sigma: float
3. Vote
   1. change_a: FK to change
   2. change_b: FK to change
   3. change_c: FK to change
   4. universe_chosen: Option ( $\alpha$ or $\beta$ )
   5. change_chosen: Option ( 1, 2 or 3 )
4. User_Change
   1. user: FK to user
   2. change: FK to Change
   3. ignored: boolean
   4. favorite: boolean