Put the worst user reports first in the moderation queue

Fields

• Where the report came from (report button, notice form, email to legal)
• Reason the reporter picked, or the law the notice names
• Reporter's comment, or the full notice
• Reported message, post, listing or profile text
• Earlier messages around it, oldest first (empty if none)

Questions

1. Which community rule, if any, does the reported content most clearly break? Choice

   Judge the reported content, using the thread for meaning. Pick the rule it breaks most clearly, whatever reason the reporter picked. When it breaks more than one, pick the one that could do the most harm. Do not decide what action to take.

   • No rule broken Rude, silly or unwelcome at most, but breaks none of the rules.
   • Spam or scam Repeated promotion, fake offers, or requests for money, account details or a trade outside the marketplace.
   • Harassment or bullying Insults, bullying or repeated unwanted contact aimed at a person.
   • Hate Attacks people for a protected characteristic such as race, religion, disability, sex or sexual orientation.
   • Violent threat Threatens or encourages violence against a person or a group.
   • Unwanted sexual content Unwanted sexual messages between adults, or sharing or threatening to share intimate images without consent.
   • Child safety Any sexual interest in a minor, attempts to get a minor to share images or keep contact secret, or exploitation of a minor.
   • Suicide or self-harm States an intent to self-harm, or encourages suicide or self-harm.
   • Violent extremism Supports or promotes a violent extremist group or act.
   • Illegal goods or services Offers drugs, weapons, stolen accounts or other illegal goods or services.
   • Impersonation Pretends to be another person, a staff member or an organisation.
   • Private details exposed Posts someone's address, phone number, school or other private details without consent.
   • Copyright or trademark Copies protected work or uses a trademark without permission.

2. Is this a user report, a notice that the content is illegal, or an order from an authority? Choice

   Judge from how the report is written as well as where it came from. A report-button comment that says the content breaks a named law and asks for it to be taken down is a notice.

   • User report Flags content as breaking our rules, without claiming a law makes it illegal.
   • Notice of illegal content A person or organisation says the content is illegal and names the law or the kind of offence, such as a Digital Services Act notice or a takedown request from a rights holder.
   • Order or request from an authority Presents itself as coming from a police force, a court, a regulator or another public authority, ordering action or asking for information.

3. Does the reported content fit the reason the reporter gave? Yes / No

   Compare the reported content with the reason picked or the law named. Yes: The content fits the reason given. No: The content fits a different rule or no rule, or the reason given does not describe what the content does.

4. Is the reported content aimed at a specific, identifiable person? Yes / No

   Count content sent to a person, or content that names, tags or clearly points to one. Yes: The content is aimed at a specific person. No: The content is aimed at a group, at no one, or at the public in general.

5. Does the reported content show a risk of serious physical harm soon? Yes / No

   Count a threat of violence with a time, a place or a named target, a plan to meet someone in order to harm them, or a stated plan or intent to self-harm soon. Signs that a child may be at risk, without a plan to meet, are marked by the child safety question instead. Yes: The content or thread shows such a threat or plan. No: No threat or plan of serious harm soon appears.

6. Does anything in the report suggest a child may be at risk? Yes / No

   This flag sends the report to trained people first; it does not decide that abuse happened. Read the reporter's text, the reported content and the thread. Count: sexual talk with or about someone who says or appears to be under 18; requests for images; requests to keep contact secret from parents; offers of gifts, money or game currency in return for something; questions about where a child lives, goes to school or when they are alone; pushing a child to another app; threats to share images. Yes: At least one such sign appears. No: No such sign appears.

7. Does anyone in the reported content or thread push to continue on another app, a private channel or an outside payment method? Yes / No

   Count a named or disguised app, a phone number, a username on another service, or a payment outside the marketplace. Yes: Someone pushes to move the contact or a payment elsewhere. No: No such push appears.

8. How serious is the reported content? Scale

   Judge the content and thread as a whole, whatever reason the reporter picked.

   • Low Rude or low quality, with no clear breach of a rule.
   • Medium A clear breach of a rule, with no danger to anyone and no money at stake.
   • High Targeted abuse, a scam with money or accounts at stake, or content that may be illegal.
   • Critical A danger to someone's life or safety, or to a child.

9. Does the notice explain why the content is illegal clearly enough for a reviewer to assess it? Choice

   Judge only the explanation, not whether the notifier is right. Code checks the link, the name, the email and the good faith statement.

   • Explained The notice says what the content is, which law or right it breaks, and why, so a reviewer can assess it without research.
   • Not explained The notice claims the content is illegal but gives no reason, or too little for a reviewer to assess.
   • Not a notice The report makes no claim that the content is illegal: a user report under our rules.

10. Is there enough text in the report to judge it? Yes / No

    Answer No when the report depends on an image, a video, a link or earlier messages that are not included. Yes: A careful reader could judge the report from the text given. No: The text depends on something not included.

Lens columns

violation, violation_probability, notice_type, notice_type_probability, reason_matches, reason_matches_probability, targets_a_person, targets_a_person_probability, imminent_harm, imminent_harm_probability, child_safety_concern, child_safety_concern_probability, moves_off_platform, moves_off_platform_probability, severity, severity_average, notice_explanation, notice_explanation_probability, enough_context, enough_context_probability