The moderation queue
Reading what arrives, deciding whether it is published, and the other things a moderator can do to a report.
Nothing reaches the public board unless a moderator approves it. There is no automatic publishing and there must not be. What a moderator approves and declines is a judgement, and it belongs to whoever runs the site, who publishes their standard on the site's moderation standard page.
Moderators and administrators work here, under Moderation › Queue, after accepting the moderator agreement. An administrator publishes new versions of the agreement under Moderation › Agreement, with a second-factor check, and sees who accepted which version.
The queue#
Reports waiting for a decision come first, oldest first. Search reaches every report in every state, and the filters you set stay in the address, so a view can be bookmarked or shared with another moderator. Decide here opens the decision form under a row, without leaving the queue.
Deciding#
Every act below needs a written reason, except marking a flagged report as read. The reasons are kept with the report's history, which only staff see.
- Approve. The report is published, if its reporter allowed publication. A moderator can publish a report's text and hold back its files, never the reverse.
- Decline. Choose one of the fixed reasons below. If the reporter's address is still readable, they are emailed the reason with a link to the site's terms.
- Unclear. Neither, for now.
The reasons for declining:
- It is not a first-hand account of something that happened.
- It is outside what this logbook records.
- It identifies a private person in a way we cannot publish.
- It contains content that is unlawful or that we cannot host.
- It is abusive, or it is spam or an advertisement.
- It repeats a report already filed.
- We could not understand it well enough to assess it.
Declined reports are counted by reason on the board's activity page, never shown.
Approving many at once. Several waiting reports can be approved together, up to 100. One approval is recorded per report, and no reason is asked. Reports that cannot be approved (already decided, say) are skipped and named. Declining many at once exists only on the flood screen.
Other things a moderator can do#
- Change an earlier decision.
- Withdraw a published report from the board, or restore one a moderator withdrew. A reporter's own withdrawal is never undone by a moderator.
- Severity: a rating with a reason, for staff. Never sealed, never public.
- Incident: link reports that describe the same incident, so they can be read together.
- Second read: reports that contain words from the review-flag list are marked for a second read, and a moderator confirms they have been read.
- Editorial note: a published note beside a report, with citations, versioned. Turned on per logbook under Moderation settings. An administrator can mark a note under review, withdraw it, or republish it.
- Sources: links a moderator attaches to a report. Seen by staff and vetted journalists only, never by the public.
- Threat flag: raised on a report that describes a threat to somebody's life or safety, and emails the owner at once.
- Corroboration: how many published accounts there are of the same incident, counted from the incident links moderators make. Moderators also see, across the board, how many published reports are corroborated, how many stand alone, and how many of each severity.
- Files and evidence: decided per file. See Files and evidence.
- Reporter details: opened only with a reason. See Reporter details.
Corrections#
A reporter can correct their report, as a new version in the same thread. The new version comes to the queue like any report. A correction can take back permission to publish, never give it.
Moderation settings#
Under Moderation › Moderation settings, administrators set:
- whether the site publishes a public board at all, and whether the public may download the published reports as a spreadsheet and each logbook's published chain file;
- the review-flag words: the shipped list, your own, or none;
- whether each logbook takes editorial notes;
- which logbooks are published as counts only, and their small-area thresholds. See Maps and numbers.