Moderation and Enforcement Overview

[ajbott]

Infraction Reporting and Enforcement

Percentage for Infractions

• Process:
  • If a certain percentage of users report an item, it triggers removal.

High Level Reports

• Immediate Suspension:
  • Reports on serious offenses like NSFW, child pornography, etc., lead to instant suspension pending investigation.

Reporting System

• Reportable Items:

  • Messages, photos, video calls, audio calls.
  • Specific categories: photoshopped, filters, memes/cartoons, pets, group photos, community guidelines.

• Data Retention Reminder:

  • Users are informed they have X days to report to maintain data retention.

• Local Moderators:

  • Designate local moderators for specific geographic regions, similar to Reddit.

Punishments

• Bans:

  • Permanent Ban: Account is permanently locked out.
  • Temporary Ban: Account locked for X days.
  • Suspensions: Account locked for a longer period than a temporary ban.

• Profile Tags:

  • Used to tag profiles for specific behaviors.

Ghosting Punishment Example

• Report 1:
  • Guilty, not intentionally: Issue a warning.
  • Guilty, intentionally: Issue final warning or account suspension.
  • Not guilty, not recorded (insufficient evidence).
  • Not guilty, evidence of false report: Issue final warning or ask them to be careful.

Dating Related Crimes - SAFETY/ENFORCEMENT

• Community Notification:
  • System notifies the community of recent dating crime events in the area (opt in/out).

Threat Interaction Notification - SAFETY

• Community Alert:
  • Notify the community of threats they may have interacted with.

Dynamic Community Reporting

• Community Reporting:

  • Users can report a wide range of bad behaviors.

• Sub-Features:

  • View most reported items.
  • Community votes on disciplinary actions after X votes.

Community Rules Page

• Rules Display:

  • Displays rules, reportable items, and disciplinary actions.

• Sub-Features:

  • Dynamically updates with rule changes.
  • Lists high-level offenses.

Automated Enforcement System

• Phishing Detection:

  • Learns chat history patterns to detect phishing attempts.

• Other Detection Systems:

  • Fake account detection, duplicate account detection, not cell phone number detection.

Automated User Moderation Triggering

• Relationship Abuse Prevention:
  • Checks if a user enters into too many relationships back-to-back to prevent abuse.

Moderation

• Two Levels of Moderation:

  • Volunteer Moderators can assist if there are too many reports.

• During Investigation:

  • Investigators have access to report details, chat logs, and dispute details.

Enforcement

• Offense Levels:

  • High - Bannable Offenses
  • Medium - Suspension/Warning
  • Low - Warning

• Offense Types:

  • Ghosting, over-reporting, bad behavior, player mentality, false reporting, duplicate accounts, advertising/selling, filter reports, wrong filter settings.

• Investigation Tags:

  • Applied during investigations for transparency.

Banning

• Account Bans:
  • Banned accounts show a locked page with the final dispute.
  • Accounts can send a dispute once per year.

Community Rules

• Specific Rules:
  • No soliciting, no ghosting, no sexting on the app, no falsifying reporting.

Picture Rejection

• Rejection Criteria:
  • Under 18 y.o., child porn, celebrities (copyright infringement), no people, group photos, violence, drugs, death, weapons, offensive gestures, different gender to profile, personal details.