Skip to content

README.md

Latest commit

 

History

History
74 lines (55 loc) · 3.59 KB

README.md

File metadata and controls

74 lines (55 loc) · 3.59 KB

Example: Causal State Machine with Effect Ethos

This example demonstrates the integration of a Causal State Machine (CSM) with an Effect Ethos to add a layer of deontic (normative) reasoning to a reactive system.

It models a simple temperature monitoring system that triggers an alert when the temperature exceeds a threshold. However, the action of sending the alert is subject to approval by the EffectEthos, which evaluates it against a set of predefined norms.

How it Works

The example is structured to highlight the difference between a purely reactive system and one governed by deontic rules.

1. The Components

  • CausalState: A state that becomes active when a specific condition is met. Here, it checks if the temperature data value is greater than 50.0.
  • CausalAction: An action that is triggered when the corresponding state is active. In this case, it's a simple function that prints an alert to the console.
  • CausalStateMachine (CSM): Manages the link between states and actions. When a state evaluates to true, the CSM triggers its associated action.
  • EffectEthos: A deontic reasoning engine that contains a set of norms (Teloids). It evaluates a ProposedAction to determine if it is permissible, obligatory, or impermissible.
  • Teloid (Norm): A single rule within the ethos. In this example, we define a norm that makes the High temp alert Impermissible. This simulates a scenario where, for instance, alerts might be suppressed to prevent spamming or if a manual override is in effect.

2. The Scenario

The main function executes the same logic twice to provide a clear comparison:

A. Execution without EffectEthos

  1. The CSM is initialized with the high-temperature state and the alert action, but without the EffectEthos.
  2. The CSM is evaluated with a temperature of 60.0, which exceeds the threshold.
  3. The CausalState becomes active.
  4. The CSM immediately fires the CausalAction, and the alert "Alert! High temperature detected!" is printed.
  5. The result is Ok(()), as the action was executed without restriction.

B. Execution with EffectEthos

  1. The CSM is initialized with the state, the action, and the EffectEthos containing the impermissibility norm. A TeloidTag (High temp alert) is provided to link the evaluation to the correct norm.
  2. The CSM is evaluated again with a temperature of 60.0.
  3. The CausalState becomes active.
  4. Instead of firing the action directly, the CSM creates a ProposedAction and submits it to the EffectEthos for evaluation.
  5. The EffectEthos finds the active High temp alert norm, which renders the action Impermissible.
  6. Because the action is forbidden, the CSM does not fire the action. Instead, it returns a CsmError::Forbidden containing a detailed explanation of the verdict.

How to Run

From the root of the deep_causality project, run:

cargo run -p csm_examples --example csm_effect_ethos_example

Expected Output

--- Running CSM without EffectEthos ---
Alert! High temperature detected!
Result without ethos: Ok(())

--- Running CSM with EffectEthos ---
Result with ethos: Err(Forbidden("The final verdict is Impermissible....
The outcome is Impermissible because at least one impermissible norm was active and undefeated, which has the highest precedence."))

This output clearly illustrates the core concept: the EffectEthos successfully intercepted and blocked an action that would have otherwise been executed, providing a powerful mechanism for building safer and more robust systems.