Social DNA

Social DNA

ADAMs goal is to create a communication infrastructure that enables group agency without imposing a bias on how a group manages itself.

In a social group this could mean many different things like:

  • Rating systems: Deciding how a Post or a Comment's rating is calculated.
  • Access rights: Allowing certain Agents to do certain actions when certain conditions in the graph are met.
  • Shared definitions: Defining data structures in the graph that have a certain meaning.

Social DNA enable Neigbourhoods to add shared computer programs that run on the Agents device, based on the structure of the graph data. Using Prolog (opens in a new tab), a turing complete logical programming language, any kind of alorithm can be run based on the state of the graph.


Don't worry if you are unfamiliar with Prolog (opens in a new tab). You will normally either use AI to write it for you, or you will use some of our abstraction layers to avoid writing Prolog code yourself. (Think of ORMs vs raw SQL, Ad4m comes with its own implementation Subject Classes ).

Quick Intro to Prolog

Prolog is a declarative programming language commonly used for logic programming. It revolves around defining logical facts and rules to solve problems through logical inference. Here's a super quick introduction to the basics of Prolog with code examples:

Facts: Facts are statements that define relationships or properties.

likes(john, pizza).

Rules: Rules define logical implications or conditions using predicates and logical operators.

likes(john, X) :- pizza(X).

Variables are denoted with uppercase letters or by starting with an underscore (_).

Queries: You can ask Prolog queries to find solutions that satisfy a specific condition or to retrieve information.

?- likes(john, pizza).

Will return true

Creating Social DNA

Let's say we want to define top_posts as any post in the Perspective with more than 5 likes.

First let's create an Perspective, and add a post:

const myPosts = await ad4m.perspective.add("My Posts");
const expressionURL = await ad4m.expression.create("My first Post", "literal");
  source: "ad4m://self",
  predicate: "sioc://has_post",
  target: expressionURL,

Then let's use our Agent's DID to say that we like our own post 10 times:

const { did } = await;
for (let i = 0; i < 10; i++) {
    source: did,
    predicate: "sioc://likes",
    target: expressionURL,

Great! Now we have one post with 10 likes in our Perspective. Now let's define what top_comments mean to use in this Perspective:
top_posts(Post) :-
    triple(Source, "sioc://has_post", Post),
    link(Source, "sioc://likes", Post, Author, Timestamp),
    count_likes(Post, Count),
    Count > 5.
count_likes(Post, Count) :-
    findall(_, link(_, "sioc://likes", Post, _, _), Likes),
    length(Likes, Count).

After writing the Prolog definition, you can add the definition as Social DNA to any Perspective:


Querying the Social DNA

The Social DNA is now added to the Perspective, and we can easily ask the Perspective to give us the top comments:

const res = await perspective.infer(
  `top_comments(Source, Target, Author, Timestamp)`

What is interesting here is that the top_comments definition could have a different algorithm in a different Perspective, but the developer wanting to display these comments, would not need to worry about the actually reasoning behind what top_comments actually mean for this specific Perspective.

This gives us complete group agency!