Video and audio

Required

• Run media tools in a sandbox with no network

  MED-14

  Run every tool that decodes uploaded media, such as FFmpeg or ImageMagick, in a sandbox. It has no network, sees only its input and output folders, and stops at a time and memory limit.

Recommended

• Transcode every video into an adaptive stream

  MED-13

  Transcode every uploaded video in a background job into an HLS or DASH stream with three or more renditions, and play that stream. Never play the uploaded original.

• Probe each media file and refuse it past its limits

  MED-15

  Before transcoding, read each file's duration, resolution and frame rate with a probe such as ffprobe. Refuse the file when any value passes the limit of its upload kind.

• Make a poster and a thumbnail from every video

  MED-16

  Make a poster image and a thumbnail for every video in the transcoding job, and let the uploader pick another frame. Keep the thumbnail address the same when the video is encoded again.

• Keep captions as WebVTT files with each video

  MED-17

  Store each video's captions as WebVTT files, one per language, and list them in its player and its stream. Let uploaders add and correct them, and mark machine captions as automatic until a person checks them.

Optional

• Even out the loudness of uploaded audio

  MED-18

  Measure each uploaded audio track's loudness by ITU-R BS.1770 and adjust it to one target, such as −23 LUFS with peaks under −1 dBTP.

Rules on video and audio in other playbooks

• Send video that starts fast and costs little sets the poster, preload and range requests for <video>.
• Load embedded videos and maps only after a click keeps third-party players from tracking visitors.
• Give each video a watch page gets videos into search results.
• Let large downloads resume with range requests lets players seek and resume.
• Report child sexual abuse material and keep the evidence applies to video uploads too.