Distinguish a cut from an event
An editorial cut changes the view without necessarily changing the underlying activity. Conversely, a new event can begin within one continuous shot. State whether your labels follow visual edits, real-world actions, or both. Treating every cut as a new event can undermine a temporal task whose boundaries depend on behavior rather than production style.
Choose acceptable editorial patterns
List hard cuts, fades, replay sequences, split screens, slow motion, and inserted graphics separately. A blanket no-edit requirement may rule out useful footage without explaining the real concern. Describe which patterns remove necessary continuity or introduce misleading repetition. Include source examples where possible so reviewers can identify the intended distinction during a pilot rather than guessing from terminology.
Track relationships across versions
A published edit may derive from a longer recording and include reordered or repeated material. Ask for version identifiers and source offsets where available. If complete source relationships are unknown, document that limit. Keep alternate edits grouped when evaluation independence requires it, and avoid counting multiple cuts of the same activity as unrelated source coverage simply because their files differ.
Review transitions deliberately
Inspect samples around cuts, introductions, outros, and montage sections. Check whether temporal labels, transcripts, and descriptions remain valid across those transitions. Decide how exceptions are flagged or excluded before expanding acquisition. Fiund reviews whether proposed footage fits these conditions and the intended use; the presence of a playable video file is not evidence that its editorial structure supports your task.
Review checklist
- Separate editorial boundaries from action boundaries.
- List acceptable and excluded editing patterns.
- Track source relationships between alternate edits.
- Inspect annotations around transitions in the pilot.