Fizi: Created page with "{{DISPLAYTITLE:Module 7 — IRC Moderation}}
DarkWorld IRC Staff Training
Module 7: IRC Moderation
{| class="wikitable" style="width:100%;" |- ! Program | DarkWorld IRC Staff Training Program |- ! Module | 7 of 10 |- ! Difficulty | Intermediate |- ! Estima..."

2026-08-08T17:50:34Z

Created page with "{{DISPLAYTITLE:Module 7 — IRC Moderation}} <div style="background:#151515; border-left:5px solid #8b5cf6; color:#eeeeee; padding:16px; margin-bottom:20px;"> <span style="font-size:170%; font-weight:bold;">DarkWorld IRC Staff Training</span><br> <span style="font-size:125%;">Module 7: IRC Moderation</span> </div> {| class="wikitable" style="width:100%;" |- ! Program | DarkWorld IRC Staff Training Program |- ! Module | 7 of 10 |- ! Difficulty | Intermediate |- ! Estima..."

New page

{{DISPLAYTITLE:Module 7 — IRC Moderation}}

<div style="background:#151515; border-left:5px solid #8b5cf6; color:#eeeeee; padding:16px; margin-bottom:20px;">
<span style="font-size:170%; font-weight:bold;">DarkWorld IRC Staff Training</span><br>
<span style="font-size:125%;">Module 7: IRC Moderation</span>
</div>

{| class="wikitable" style="width:100%;"
|-
! Program
| DarkWorld IRC Staff Training Program
|-
! Module
| 7 of 10
|-
! Difficulty
| Intermediate
|-
! Estimated study time
| 4–5 hours
|-
! Assessment
| Knowledge check, moderation simulations, and written case analysis
|-
! Prerequisite
| [[DWIRC:Staff Training/User Support|Module 6 — User Support and Communication]]
|}

= Module Overview =

IRC moderation is the controlled use of channel or network authority to stop harmful or disruptive behavior.

Effective moderation is not measured by the number of users warned, kicked, or banned. It is measured by whether staff:

* Protect users and channels.
* Restore normal communication.
* Apply policies consistently.
* Use proportionate action.
* Minimize harm to innocent users.
* Explain important actions clearly.
* Preserve relevant evidence.
* Escalate serious incidents.
* Review temporary restrictions.
* Remain accountable.

This module focuses primarily on channel moderation and the judgment required before using elevated IRC authority.

Network-wide operator enforcement will be covered in:

'''[[DWIRC:Staff Training/IRC Operator Fundamentals|Module 9 — IRC Operator Fundamentals]]'''

= Learning Objectives =

After completing this module, the candidate should be able to:

# Distinguish channel moderation from network enforcement.
# Determine when staff intervention is appropriate.
# Use warnings, voice restrictions, kicks, and bans proportionately.
# Select a reasonably precise ban mask.
# Handle spam, flooding, harassment, and ban evasion.
# Respond to bots, raids, and relayed abuse.
# Avoid punishing innocent users.
# Preserve evidence before it disappears.
# Write clear moderation reasons.
# Record significant actions.
# Review temporary restrictions.
# Escalate cases outside their authority.

= 1. Purpose of Moderation =

The purpose of moderation is to:

* Stop disruptive behavior.
* Protect users.
* Preserve the purpose of a channel.
* Restore normal conversation.
* Enforce applicable rules.
* Prevent repeated abuse.
* Provide a fair and reviewable response.

Moderation must not be used to:

* Win arguments.
* Silence reasonable criticism.
* Punish someone for disliking staff.
* Promote a staff member’s personal beliefs.
* Favor friends.
* Take over a channel.
* Demonstrate power.
* Retaliate against complaints.
* Bypass an investigation.
* Conceal staff mistakes.

= 2. Channel Authority and Network Authority =

== Channel Authority ==

Channel founders and authorized channel staff normally manage:

* Channel topics.
* Participation rules.
* Voice and operator status.
* Kicks.
* Channel bans.
* Channel access lists.
* Channel bots.
* Channel-specific moderation.

== Network Authority ==

DarkWorld IRC network staff may handle:

* Network-wide policy violations.
* Spam across multiple channels.
* Network advertising and recruitment.
* Severe harassment.
* Ban evasion across channels.
* Attacks against users or infrastructure.
* Malicious bots.
* Services abuse.
* Staff impersonation.
* Unauthorized relays.
* Incidents affecting multiple servers or channels.

A channel operator should not pretend to have network authority.

An IRC operator should not interfere with legitimate channel management without a valid network-level reason.

= 3. When Should Staff Intervene? =

Staff intervention may be appropriate when:

* A network rule is being violated.
* Harm is ongoing.
* A channel requests authorized assistance.
* Normal channel management is unavailable.
* Multiple channels are affected.
* A bot or relay creates network-wide disruption.
* A serious security or safety issue exists.
* Management assigns the matter.

Staff may decide that no action is required when:

* The issue is a minor personal disagreement.
* A legitimate channel rule was enforced.
* Evidence is insufficient.
* The reported conduct is not prohibited.
* The user is asking staff to retaliate.
* The issue belongs to another project.
* The situation has already ended and requires review rather than immediate intervention.

Choosing not to take enforcement action can be the correct professional decision.

= 4. Moderation Decision Framework =

Before acting, ask:

# What conduct is occurring?
# Is it still happening?
# Which rule applies?
# Who is affected?
# What evidence is available?
# Has the user already been warned?
# Is immediate action necessary?
# What is the least severe effective response?
# Could the response affect innocent users?
# Do I have authority?
# Must the action be recorded?
# Does the case require escalation?

During an emergency, staff may need to act quickly. The action should still be reviewed and documented afterward.

= 5. The Moderation Response Ladder =

A general response ladder is:

# No action.
# Guidance or informal correction.
# Formal warning.
# Targeted speak restriction or removal of voice.
# Kick.
# Temporary channel ban.
# Longer channel restriction.
# Escalation for network-level action.
# Permanent action approved through the proper process.

This is a framework, not a mandatory sequence.

Serious incidents may require immediate protective action without a warning, including:

* Active flooding.
* Malware distribution.
* Credential phishing.
* Credible threats.
* Doxxing.
* Coordinated raids.
* Active ban evasion.
* Attacks against network infrastructure.

= 6. Guidance and Informal Correction =

Guidance is appropriate when:

* The violation is minor.
* The user may not understand the rule.
* The conduct caused limited harm.
* The user is cooperative.
* An immediate sanction is unnecessary.

Example:

<pre>
Please stop repeating the same message. It is disrupting the channel.
</pre>

Guidance should not sound like a threat.

= 7. Formal Warnings =

A formal warning should include:

* The conduct that must stop.
* The applicable rule or policy.
* The required corrective action.
* A possible consequence if it continues.
* An official policy link where relevant.

Example:

<pre>
Formal warning: Stop sending unsolicited external-network invitations. This may violate the DarkWorld IRC advertising and recruitment policy. Continued activity may result in further restrictions.
</pre>

A warning is useful only if:

* It reaches the correct user.
* It is understandable.
* The user has a reasonable opportunity to comply.
* It is documented when significant.

A warning may be skipped when delay would permit serious ongoing harm.

= 8. Speak Restrictions and Moderated Mode =

Sometimes it is sufficient to stop a user from speaking without removing them.

Possible options, depending on current server configuration, include:

* Removing voice.
* Applying a targeted quiet or speak restriction.
* Setting the channel to moderated mode.
* Restricting unidentified users from speaking.
* Restricting notices or CTCP messages.

A targeted restriction is usually less disruptive than moderating an entire channel.

Before using `+m`, consider:

* How many legitimate users will be silenced.
* Whether voiced users can continue.
* Whether staff can manage voice requests.
* Whether a targeted restriction is available.
* How long the mode will remain.
* Who will restore normal operation.

= 9. Kicks =

A kick removes a user from a channel but does not necessarily prevent them from returning.

Syntax:

<pre>
/KICK #channel nickname reason
</pre>

Example:

<pre>
/KICK #Training ExampleUser Repeated flooding after warning
</pre>

A kick may be appropriate when:

* The disruptive conduct must stop immediately.
* A warning was ignored.
* The user needs to be removed from the current discussion.
* A channel rule authorizes removal.
* A ban is not yet necessary.

A professional kick reason should be:

* Short.
* Accurate.
* Neutral.
* Related to the conduct.
* Free from insults.
* Free from unnecessary confidential details.

Poor reasons include:

<pre>
Annoying
Go away
Because I said so
Idiot
Staff decision
</pre>

= 10. Channel Bans =

A channel ban normally prevents a matching user from joining or participating.

Add a ban:

<pre>
/MODE #channel +b mask
</pre>

Remove a ban:

<pre>
/MODE #channel -b mask
</pre>

View bans:

<pre>
/MODE #channel +b
</pre>

A ban may be appropriate when:

* A kicked user immediately returns and continues.
* The conduct is serious.
* Ban evasion is occurring.
* A bot repeatedly disrupts the channel.
* The user presents an ongoing risk.
* The channel’s established rules support the restriction.

= 11. Choosing a Ban Mask =

A full user mask appears as:

<pre>
nickname!ident@hostname
</pre>

Possible masks include:

{| class="wikitable" style="width:100%;"
! Mask
! General match
! Limitation
|-
| `ExampleNick!*@*`
| Current nickname
| Easy to evade by changing nickname
|-
| `*!example@*`
| Ident
| Idents may be shared or changed
|-
| `*!*@Example.User.DarkWorld.Network`
| Displayed host
| Host may be shared or changed
|-
| `*!*@*.provider.example`
| Wider provider range
| May affect many innocent users
|-
| `*!*@*`
| Everyone
| Normally dangerously broad
|}

Before selecting a mask, consider:

* Registered account.
* Displayed host or cloak.
* Ident.
* Whether the host is shared.
* Whether a bouncer, VPN, gateway, or relay is involved.
* Whether an extended account-based restriction is available.
* Whether the ban is temporary.
* How many innocent users may match.

The goal is the narrowest reliable restriction that stops the harmful conduct.

= 12. Ban Then Kick =

When both a ban and kick are necessary, the normal sequence is:

# Set the ban.
# Kick the user.
# Confirm the ban entry.
# Record the action.
# Review the ban later.

Example:

<pre>
/MODE #channel +b *!*@Example.User.DarkWorld.Network
/KICK #channel ExampleNick Repeated disruption after warning
</pre>

Setting the ban first helps prevent immediate re-entry between commands.

Staff should verify the exact target and mask before sending either command.

= 13. Temporary and Permanent Restrictions =

Most ordinary moderation bans should be reviewable.

A moderation record should identify:

* Whether the restriction is temporary or indefinite.
* When it was applied.
* Why it was applied.
* Who applied it.
* When it should be reviewed.
* Which evidence supports it.
* Whether the user was warned.
* Whether innocent users may be affected.

An indefinite restriction should not mean “set and forgotten.”

Permanent or long-term network restrictions should require appropriate authorization and review.

= 14. Ban Exceptions =

A ban exception may allow a user to bypass a matching channel ban.

View exceptions:

<pre>
/MODE #channel +e
</pre>

Add:

<pre>
/MODE #channel +e mask
</pre>

Remove:

<pre>
/MODE #channel -e mask
</pre>

Exceptions can help users affected by a shared-host ban, but they can also make the ban structure difficult to understand.

Where possible, replace an unnecessarily broad ban with a more precise restriction instead of creating many exceptions.

= 15. Spam Moderation =

Spam may include:

* Repeated messages.
* Mass private messages.
* Repeated links.
* Automated promotions.
* Cross-channel advertising.
* Repetitive irrelevant text.
* Messages distributed through multiple accounts.

Response factors include:

* Number of messages.
* Number of recipients.
* Number of channels.
* Whether automation is involved.
* Whether the user was warned.
* Whether the messages contain malicious links.
* Whether the conduct is continuing.

A single accidental repeated message usually does not require the same response as coordinated mass spam.

= 16. Flood Moderation =

Flooding may include:

* Text flooding.
* Join and part flooding.
* Nickname-change flooding.
* CTCP flooding.
* Notice flooding.
* Private-message flooding.
* Repeated formatting.
* Bot-command flooding.

== Single-User Flood ==

Possible response:

# Warn if practical.
# Apply a targeted restriction.
# Kick if necessary.
# Ban if the behavior continues.
# Record repeated or serious conduct.

== Multi-User Raid ==

Possible response:

# Apply emergency channel modes if authorized.
# Preserve evidence.
# Restrict the active sources.
# Avoid uncontrolled broad bans.
# Request assistance.
# Escalate to network operators.
# Restore normal channel modes afterward.
# Document the complete incident.

= 17. Harassment Moderation =

Harassment cases require attention to:

* Repetition.
* Targeting.
* Whether contact is unwanted.
* Attempts to evade blocks or bans.
* Threats.
* Use of multiple accounts.
* Private-message conduct.
* Exposure of personal information.
* Conduct across several channels.

Staff should not require a targeted user to continue engaging with the harasser.

A normal response may include:

* Advising the target to stop responding.
* Helping the target use client ignore features.
* Preserving evidence.
* Applying channel restrictions.
* Escalating repeated or network-wide conduct.
* Taking immediate protective action for threats or doxxing.

Client-side ignore is useful, but it should not be treated as the only response to serious network abuse.

= 18. Ban Evasion =

Possible indicators of ban evasion include:

* Immediate return under a different nickname.
* Repeated conduct from another connection.
* Use of a different account after restriction.
* Admission of evasion.
* Use of a relay, bot, or another user to continue.
* Strongly matching timing and behavior.

However:

* Similar behavior does not prove identity.
* A shared host does not prove identity.
* A VPN exit may be shared.
* A bouncer may host multiple users.
* A relay may display remote users through one bot.

Staff should combine available evidence and escalate broader enforcement decisions.

= 19. Bots and Automated Clients =

Bots may be legitimate, misconfigured, compromised, or malicious.

Before acting, determine:

* What the bot does.
* Who operates it.
* Whether the channel authorized it.
* Whether it is an official or registered service.
* Whether it is a relay.
* Whether it responds to commands from anyone.
* Whether the behavior is accidental.
* Whether other channels are affected.

An actively flooding bot may need immediate removal even if the cause is accidental.

After containment:

# Notify the operator where appropriate.
# Explain what behavior must be corrected.
# Require testing before return where necessary.
# Escalate malicious or repeated behavior.
# Record significant incidents.

= 20. Relay-Related Moderation =

A relay bot may display messages originating from users outside DarkWorld IRC.

Staff should identify:

* The DarkWorld channel.
* Relay bot nickname.
* External network or platform.
* Remote channel.
* Registration or approval status.
* Whether prohibited advertising is being relayed.
* Whether the relay operator can filter the content.
* Whether the abuse comes from one remote user or the relay system generally.

Do not assume the relay bot’s operator personally wrote every relayed message.

However, relay operators remain responsible for compliance and moderation controls.

Unauthorized or non-compliant relays should be handled through the approved relay-compliance process.

= 21. Advertising and Recruitment Moderation =

When possible advertising is observed:

# Preserve the message and context.
# Determine whether it was requested.
# Check frequency and distribution.
# Determine whether it promotes or recruits for an external network or project.
# Check for an applicable approval.
# Compare the conduct with the approved scope.
# Stop ongoing mass abuse where authorized.
# Escalate administrative decisions.
# Document the result.

A normal relevant link in conversation is not automatically the same as repeated unsolicited recruitment.

= 22. Malicious Links and Phishing =

Possible malicious conduct includes:

* Fake NickServ links.
* Fake login pages.
* Credential requests.
* Malware downloads.
* Shortened links designed to conceal a destination.
* Staff impersonation.
* Messages claiming that users must reveal passwords.

If malicious content is active:

# Stop further distribution.
# Preserve the destination safely without encouraging users to open it.
# Warn affected users.
# Do not publicly repeat sensitive tokens or credentials.
# Escalate immediately.
# Identify potentially compromised accounts.
# Record where the content was distributed.

Trainees should not open suspicious files or links on personal or production systems.

= 23. Channel Takeover Situations =

A channel takeover claim may involve:

* A compromised founder account.
* Unauthorized access-list changes.
* A netsplit or Services outage.
* Former staff retaining access.
* An external project dispute.
* A user misunderstanding live operator status.
* Abuse of IRC operator override.

Staff should:

# Preserve current information.
# Avoid irreversible changes.
# Confirm registered ownership.
# Separate live modes from ChanServ access.
# Protect the channel from immediate harm.
# Escalate founder or ownership changes.
# Record every administrative intervention.

A trainee must not resolve a takeover claim by simply giving founder or owner access to the person who complains first.

= 24. Evidence During Moderation =

Relevant evidence may disappear after:

* A user quits.
* A message scrolls away.
* A bot disconnects.
* A channel mode changes.
* Services resynchronize.
* A netsplit ends.

Where practical, preserve:

* Exact messages.
* Timestamps.
* Channel name.
* Nickname and visible mask.
* Registered account information, where authorized.
* Relevant mode changes.
* Kick or ban reasons.
* User warnings.
* Staff actions.
* Relay or bot information.

Do not delay immediate protection solely to collect perfect evidence.

= 25. Moderation Records =

Use a record such as:

<pre>
Case reference:
Date and time:
Channel:
Moderating staff:
User nickname:
Registered account, if relevant:
Visible user mask:
Conduct observed:
Rule or policy:
Warning issued:
Action taken:
Ban or restriction:
Reason:
Duration:
Other users affected:
Evidence location:
Escalated to:
Review date:
Final outcome:
</pre>

Avoid unnecessary personal information.

= 26. Moderation Reasons =

Good reasons describe conduct:

<pre>
Repeated message flooding after warning
Unsolicited advertising across multiple channels
Continued harassment and ban evasion
Malicious link distribution
Unauthorized relay activity pending compliance review
</pre>

Bad reasons attack the person:

<pre>
Bad person
Annoying
Idiot
Nobody likes you
Staff says so
</pre>

Moderate conduct, not personality.

= 27. Correcting Moderation Mistakes =

If staff make a mistake:

# Stop any continuing harm.
# Remove or correct the incorrect restriction.
# Restore affected access where appropriate.
# Inform the affected user honestly.
# Notify senior staff when required.
# Preserve the original action record.
# Add the correction and reason.
# Identify how to prevent recurrence.

Do not delete or falsify records to hide an error.

An appropriate acknowledgement may be:

<pre>
The restriction was applied to an incorrect mask and has now been removed. We apologize for the disruption. The incident record has been corrected and referred for review.
</pre>

= 28. Appeals =

A user may disagree with a moderation decision.

Staff should:

* Explain the available review process.
* Avoid public arguments.
* Preserve the original record.
* Refer the matter to an authorized reviewer.
* Disclose conflicts of interest.
* Avoid retaliating against the appellant.
* Implement the review outcome.
* Correct confirmed errors.

Where independent review is required, the person who took the disputed action should not be the only person deciding the appeal.

= 29. Staff Disagreement During an Incident =

If staff disagree:

# Protect users from immediate harm.
# Avoid arguing publicly.
# Preserve the relevant facts.
# Follow the current incident leader or senior authority.
# Move the disagreement to an approved staff channel.
# Record major conflicting actions.
# Request review after the incident.

Staff must not repeatedly reverse each other’s modes in public while an incident is active.

= 30. Practical Exercises =

These exercises must be performed only in an approved training channel using test identities.

== Exercise 1: Warning ==

Write and deliver a professional warning for repeated text flooding.

The warning must state:

* What must stop.
* Why it is disruptive.
* What may happen if it continues.

== Exercise 2: Kick ==

With trainer authorization:

<pre>
/KICK #ApprovedTrainingChannel TestUser Repeated flooding after warning
</pre>

The trainer should assess the accuracy and professionalism of the reason.

== Exercise 3: Ban and Removal ==

Using a trainer-provided test identity:

# Inspect the visible user mask.
# Select a precise mask.
# Add the ban.
# Confirm the ban list.
# Kick the test user.
# Remove the exact ban.
# Verify removal.
# Record the exercise.

== Exercise 4: Shared Host ==

The trainer provides three test users sharing a host.

The candidate must propose a restriction targeting only the disruptive identity, using currently supported methods.

No broad ban should be applied without trainer approval.

== Exercise 5: Channel Raid ==

The trainer simulates several flooding users.

The candidate must:

* Identify the raid.
* Apply approved temporary modes.
* Request assistance.
* Preserve evidence.
* Restore the original modes.
* Complete an incident record.

== Exercise 6: Appeal ==

The candidate receives a simulated complaint about their own ban.

They must:

* Avoid arguing.
* Explain the review process.
* Preserve the record.
* Disclose their involvement.
* Refer the decision appropriately.

= 31. Practical Scenarios =

== Scenario 1: First Minor Flood ==

A normally cooperative user accidentally pastes several lines rapidly and immediately apologizes.

'''Recommended response:'''

Guidance may be sufficient. Consider context, harm, and cooperation before using a sanction.

== Scenario 2: Repeated Flood After Warning ==

A user continues flooding after a clear warning.

'''Recommended response:'''

A kick or targeted temporary restriction may be proportionate. Apply a ban if needed to prevent immediate return and continued disruption.

== Scenario 3: Shared VPN Address ==

One abusive user and several innocent users appear through the same VPN host.

'''Recommended response:'''

Do not automatically ban the entire host. Consider account-based or more precise restrictions and escalate if network-level action is needed.

== Scenario 4: Coordinated Raid ==

Twenty new connections join and flood several official channels.

'''Recommended response:'''

Apply authorized emergency protection, request network-operator assistance, preserve evidence, and treat it as a network incident rather than an ordinary single-channel violation.

== Scenario 5: Critical Staff User ==

A user respectfully criticizes a staff decision without flooding or harassment.

'''Recommended response:'''

Do not punish the user for criticism. Direct them to the review process if appropriate.

== Scenario 6: Relay Advertisement ==

An approved relay repeatedly imports invitations to an external network.

'''Recommended response:'''

Contain ongoing disruption if necessary, record the relayed content, notify the relay operator, and escalate through relay compliance. Existing approval does not authorize policy violations.

== Scenario 7: Staff Friend ==

A friend of a staff member repeatedly violates channel rules, and the staff member asks others to ignore it.

'''Recommended response:'''

Apply policy neutrally, document the conflict, and escalate interference or favoritism.

== Scenario 8: Incorrect Ban ==

A ban intended for one user removes several legitimate users.

'''Recommended response:'''

Correct the restriction promptly, restore affected access, acknowledge the error, record the correction, and select a more precise method.

= 32. Knowledge Check =

Answer the following questions in your own words:

# What is the purpose of IRC moderation?
# What must moderation never be used for?
# What is the difference between channel and network authority?
# When may no staff action be the correct decision?
# What questions should be asked before moderating?
# What is the moderation response ladder?
# When may a warning reasonably be skipped?
# What should a formal warning contain?
# When might a targeted speak restriction be preferable to `+m`?
# What is the purpose of a kick?
# Does a kick prevent a user from returning?
# When may a channel ban be appropriate?
# Why should a ban be set before a kick when both are needed?
# What factors determine a suitable ban mask?
# Why are broad provider bans risky?
# What information should a temporary ban record contain?
# What is a ban exception?
# How does single-user flooding differ from a coordinated raid?
# Why is client-side ignore not always sufficient for serious harassment?
# What evidence may indicate ban evasion?
# Why does a shared VPN or bouncer host not prove identity?
# How should a malfunctioning bot be handled?
# What information should be collected about an abusive relay?
# How should possible advertising be evaluated?
# What should staff do about active phishing?
# Why must channel-takeover claims be escalated?
# What evidence should be preserved during moderation?
# What makes a good moderation reason?
# What should staff do after applying an incorrect restriction?
# How should an appeal be handled?
# What should staff do when they disagree during an active incident?
# Why must temporary emergency modes be reviewed?

= 33. Written Assignment =

Write approximately '''800–1,100 words''' analyzing this incident:

<blockquote>
During a busy event in an official DarkWorld channel, ten new users join within one minute. Four begin flooding repeated advertisements for another IRC network. Two use the same VPN hostname as three legitimate users already in the channel. An approved relay also begins importing similar advertisements. A channel operator sets `+m`, bans the entire VPN host, kicks all newly joined users, and leaves the channel without recording the actions.
</blockquote>

Your analysis must explain:

* Which behavior required immediate intervention.
* Whether `+m` was proportionate.
* Why the broad VPN ban requires review.
* Why all newly joined users should not automatically be treated as attackers.
* How the relay should be handled.
* What evidence should be preserved.
* Which actions belong to channel moderation.
* Which issues require network or compliance escalation.
* How innocent users should be restored.
* What should be documented.
* When emergency modes should be removed.
* How the operator’s actions should be reviewed.

= 34. Module Completion Requirements =

To complete this module, the candidate must:

* Read the complete lesson.
* Complete all six practical exercises.
* Correctly answer at least '''24 of the 32''' knowledge-check questions.
* Complete the written assignment.
* Demonstrate accurate warning, kick, and ban procedures.
* Correctly restore all training modes and bans.
* Complete an acceptable moderation record.
* Pass the raid and appeal simulations.
* Receive trainer approval.

= 35. Trainer Evaluation =

{| class="wikitable" style="width:100%;"
! Evaluation area
! Maximum points
|-
| Moderation judgment
| 20
|-
| Warnings and communication
| 10
|-
| Kick and ban accuracy
| 15
|-
| Ban-mask precision
| 15
|-
| Flood and raid response
| 15
|-
| Evidence and documentation
| 10
|-
| Appeals and accountability
| 10
|-
| Authority boundaries
| 5
|-
! Total
! 100
|}

Recommended passing score: '''75 points'''.

A candidate should not pass if they:

* Use moderation for personal retaliation.
* Repeatedly apply dangerously broad restrictions.
* Cannot restore their own mode changes.
* Ignore harm to innocent users.
* Falsify or conceal actions.
* Punish users for legitimate complaints.
* Refuse required escalation.
* Use elevated commands without authorization.

= 36. Quick Moderation Checklist =

<pre>
1. What behavior is occurring?
2. Is the behavior still active?
3. Which rule applies?
4. What evidence is available?
5. Who is affected?
6. Was a warning given or reasonably skipped?
7. What is the smallest effective action?
8. Could innocent users be affected?
9. Do I have authority?
10. Is network-level assistance required?
11. What must be documented?
12. When will the restriction be reviewed?
13. Who will restore temporary modes?
14. Is an independent review required?
</pre>

= 37. Next Module =

After passing this module, continue to:

'''[[DWIRC:Staff Training/Incident Handling|Module 8 — Abuse and Incident Handling]]'''

----

'''Previous:''' [[DWIRC:Staff Training/User Support|Module 6 — User Support and Communication]]
'''Program:''' [[DWIRC:Staff Training Program|DarkWorld IRC Staff Training Program]]
'''Next:''' [[DWIRC:Staff Training/Incident Handling|Module 8 — Abuse and Incident Handling]]

[[Category:DarkWorld IRC Staff Training]]
[[Category:DWIRC]]
[[Category:IRC Moderation]]
[[Category:Staff Documentation]]

DWIRC:Staff Training/Moderation - Revision history

Fizi: Protected "DWIRC:Staff Training/Moderation" ([Edit=Allow only administrators] (indefinite) [Move=Allow only administrators] (indefinite))

Fizi: Created page with "{{DISPLAYTITLE:Module 7 — IRC Moderation}}
DarkWorld IRC Staff Training
Module 7: IRC Moderation
{| class="wikitable" style="width:100%;" |- ! Program | DarkWorld IRC Staff Training Program |- ! Module | 7 of 10 |- ! Difficulty | Intermediate |- ! Estima..."

← Older revision		Revision as of 22:50, 8 August 2026
(No difference)

DWIRC:Staff Training/Moderation - Revision history

Fizi: Protected "DWIRC:Staff Training/Moderation" ([Edit=Allow only administrators] (indefinite) [Move=Allow only administrators] (indefinite))

Fizi: Created page with "{{DISPLAYTITLE:Module 7 — IRC Moderation}} DarkWorld IRC Staff Training Module 7: IRC Moderation {| class="wikitable" style="width:100%;" |- ! Program | DarkWorld IRC Staff Training Program |- ! Module | 7 of 10 |- ! Difficulty | Intermediate |- ! Estima..."

Fizi: Created page with "{{DISPLAYTITLE:Module 7 — IRC Moderation}}
DarkWorld IRC Staff Training
Module 7: IRC Moderation
{| class="wikitable" style="width:100%;" |- ! Program | DarkWorld IRC Staff Training Program |- ! Module | 7 of 10 |- ! Difficulty | Intermediate |- ! Estima..."