Old recordings of superseded content are a hazard, not an archive

Retention decisions made by default tend to keep exactly the wrong material.

Storage is cheap enough that nobody has to decide, so nobody decides, and the collection grows. The result is a library in which a current recording and a badly out of date one look identical: same format, same appearance, same apparent authority. A student searching for an explanation has no reliable way to tell that the requirements described in one of them changed two revisions ago. They watch it, they believe it, and they act on it, which is a genuine risk and not merely an untidiness.

The problem is worse in areas where content is tied to changing rules, standards or procedures, because superseded material there is not just stale, it is wrong in ways that matter. An old recording that confidently describes a process that no longer applies is more damaging than no recording at all, because it removes the uncertainty that would otherwise prompt someone to check. Retention obligations exist and differ by context, and they should be established locally, but they set a floor rather than a ceiling and they rarely require indefinite public availability.

What helps is deciding at the point of creation how long the thing should live and what should trigger its review, then acting on that. Marking material clearly with the version of the content it reflects. Withdrawing or flagging recordings when the underlying material changes. Keeping what must be kept for evidentiary reasons without leaving it in front of students as though it were current. Distinguishing between retained and published is most of the work.