Editorial Policy — Safe Haven Universe

Mission

The Safe Haven Research Project documents game systems through repeatable experimentation, direct observation, and player-side testing.
We focus on mechanics, behaviors, psychological design patterns, and environmental logic—not speculation, rumor, or leaks.

1. Evidence Sources

We use three types of data:

A. Reproducible Gameplay

  • Behaviors that can be triggered consistently by a typical player.
  • AI patterns observable in controlled conditions.
  • Level interactions that respond predictably to player input.

B. Observational Field Notes

  • Individual playthrough logs.
  • Screenshots or video of test runs.
  • Experiment journals submitted by contributors.

C. Comparative Trials

  • Repeated attempts under varying player conditions.
  • Cross-platform validation (PC / console / mobile).
  • Patch-to-patch consistency checks.

If a finding cannot be reproduced by multiple testers, it is not published as fact.

2. No Insider or Proprietary Material

We do not accept:

  • Developer leaks
  • Private playtest content
  • Datamined assets
  • Decompiled files
  • Internal documents
  • Paid access leaks
  • Discord “insider” claims

Any submission containing proprietary material is rejected immediately.

3. Classification of Claims

To avoid misinformation, every entry falls into one of three labels:

Evidence

Confirmed through repeated trials.
Must be reproducible on demand by average players.

Inference

Logical conclusions derived from patterns and cause–effect testing.
Presented with supporting examples.

Hypothesis

Emerging or unstable interpretations under testing.
Explicitly marked as unconfirmed until verified.

We never present hypothesis as fact.

4. Player-First Design

We write for real humans, not robots or ad networks.

  • No AI auto-generated spam
  • No keyword farming
  • No filler content
  • No “top 10” trash lists
  • No page padding to manipulate SEO

Clarity > virality.
Accuracy > search volume.

5. Corrections Protocol

If a published claim is proven false:

  • The entry is flagged.
  • Revision notes are added.
  • Historical context is preserved.
  • Updated experiments replace outdated ones.

We do not quiet-edit and pretend nothing changed.

6. Contributor Integrity

We reject:

  • Speculation disguised as evidence
  • Sensationalized claims
  • Second-hand rumors
  • Content copied from third-party sites
  • Reddit or YouTube hearsay without verification

Contributors submitting falsified data will be permanently banned.

7. AI Usage Declaration

We do not publish AI-generated dumps or content-farm output.
AI may be used for:

  • editing
  • formatting
  • summarizing internal notes

But all mechanical findings originate from human gameplay.

8. Non-Affiliation Statement

This project is not affiliated with MOB Entertainment or any IP holder.
We do not present unofficial content as canon.
We do not speak for developers.

9. Audience Responsibility

Readers are expected to:

  • Evaluate claims critically
  • Compare results in their own sessions
  • Understand that behavior may vary by patch, platform, or hardware

If you want guaranteed outcomes or official doctrine,
you’re in the wrong place.