From Goal List to Goal Graph#
Goals in the real world are rarely isolated.
A Goal may:
- Belong to a larger, longer-term Goal;
- Depend on other Goals;
- Conflict with another Goal;
- Be composed of multiple sub-Goals;
- Be changed by new evidence;
- Hold different priorities at different times.
A linear Goal List cannot carry any of the relationships above. It can only record "what needs to be done," not "what relationship these things have with each other." This is why we need to move from a Goal List to a Goal Graph.
What a Goal Graph Contains#
A Goal Graph may contain the following nodes and edges:
- Parent Goal;
- Child Goal;
- Dependency;
- Conflict;
- Shared Resource;
- Actor;
- Agent;
- Evidence;
- Risk;
- Decision;
- Outcome;
- Learning.
Parent Goal and Child Goal form the hierarchy; Dependency and Conflict form the lateral relationships between Goals; Shared Resource reveals where resource competition actually sits; Actor and Agent identify who is executing; Evidence and Outcome show whether reality is actually moving; Risk, Decision, and Learning record how this structure gets revised over time.
A Goal Graph is not a more complicated to-do list.
It is a computable expression of a person's or organization's "structure of future intent."
This distinction is substantive. A to-do list manages actions; a Goal Graph manages intent and its downstream structure of consequence. The former resets to zero once things are done; the latter produces Outcome and Learning once things are done, which in turn changes the graph itself.
Goal Object#
A possible Goal Object contains the following fields:
goal_id:
owner:
title:
intent:
desired_future_state:
why_it_matters:
priority:
time_horizon:
deadline:
status:
confidence:
dependencies:
conflicts:
constraints:
risks:
required_capabilities:
assigned_humans:
assigned_agents:
budget:
evidence_required:
current_evidence:
next_best_action:
outcome:
learning:
This is a research-stage schema draft, not a final product requirement.
It is worth noting that this schema pairs assigned_humans with assigned_agents, evidence_required with current_evidence, and outcome with learning. These paired fields show that a Goal Object is not just a to-do record β it simultaneously carries responsibility for division of labor, evidencing, and retrospection.
Where Goal Graph Sits in Goal OS#
Among the six conceptual layers of Goal OS, Goal Graph is the third:
Identity
β Memory
β Goal Graph
β Execution Engine
β Evidence
β Reflection
The question it answers is:
Which futures do I want to happen?
What dependencies, conflicts, and hierarchies exist between Goals?
Above it sit Identity and Memory β "who I am and what I value" and "what has happened, what decisions have been made"; below it sits the Execution Engine β "which people, Agents, tools, funds, and relationships need to be called on" β followed by Evidence and Reflection, which use evidence to judge whether reality is actually moving toward the Goal, and decide from there whether the Goal should continue, be revised, be paused, or be abandoned.
In other words, Goal Graph is the layer in the system where intent gets structured. Above it sit values and history; below it sit resources and execution. Without this layer, Identity cannot translate into action, and Execution loses the basis on which it is judged.