Messaging Rules

DS Colony · Last updated 1 October 2026

[Beta draft — founder-authored, pending a counsel sanity pass. Flagged for review.]

1. What messaging is for

Direct messages on DS Colony are for professional and educational use: study help, project collaboration, discussing someone's evidence or work, research questions, and career conversation connected to Data Science & AI. If a message could not plausibly begin "as one DS/AI practitioner to another", it probably doesn't belong here.

This is deliberately not a general-purpose chat platform. There are no group chats, no broadcast lists, no media attachments, and no message discovery — conversations are 1-to-1, text-only, and start from real touchpoints (a post, a portfolio, a study group, a recruiter invite).

2. Prohibited in messages

Everything in the Community Guidelines prohibited list applies to messages, with emphasis on:

• Harassment, threats, or bullying — including persistent unwanted contact after being asked to stop.
• Spam and bulk solicitation — mass-DMing users with offers, pitches, or invites is a violation (rate limits exist; working around them is also a violation).
• Off-platform scam solicitation, phishing links, or anything deceptive.
• Sexual content of any kind. Child-safety (CSAM) reports hide the message immediately on receipt and are escalated to I4C with evidence preserved.
• Sharing pirated material or others' private data (see the Copyright Policy).

3. Your protections

• Report any message — the flag control on any message sends it to the human moderation queue. Reporter identity is never revealed.
• Block any person — blocking stops both of you from messaging; your existing history stays visible to you but the channel closes. Blocks are silent; the other side gets no notification.
• Message privacy — only the two of you can read the conversation. The only exception: a reported message can be reviewed by the human moderation queue; that path is role-gated, scoped to the reported message, and logged. Messages are never pre-screened, filtered, or analyzed.
• Limits — sending is rate-limited (roughly 30 messages per 15 minutes, 10 new conversations per hour) to bound unsolicited contact. These are abuse brakes, not quotas to exploit.

4. Enforcement

Reported messages follow the Moderation & Reporting policy: human review, hide (reversible), warn, or suspend. Message content involved in moderation actions is retained as part of the audit trail; otherwise messages live until either party deletes their account (which removes the conversation for both).