A stream needs a clear start, or a load-more edge

What it is

A timeline or a feed can run off into what cannot be seen. People must be able to tell: this is the head, and more can still be fetched — a start, and a load-more edge. Without a head, it is unclear whether the story is complete. Without a “there is more” edge, the current window is taken as the whole set, or sliding never answers whether the end has come. This is how a stream’s structure shows its boundary. It is not whether infinite scroll loses the footer — that is a different paging problem.

Why it happens

A stream looks like a river without banks. Banks are boundaries: the start says “the story begins here”; the end or “load more” says “there is more, if you ask.” Missing a start, open lands in a middle stretch, sequence has no anchor, and earlier events are assumed not to exist. Missing a load edge, the bottom of the window looks like a natural end, and distant events are taken as absent. Auto-appending the next batch near the bottom, with no “fetching / this is the end” edge, makes two endings indistinguishable: a temporary window end, and a true empty. The edge should also speak in the axis’s language — “earliest record,” “show earlier” — not only as a spinner. A stream may have no absolute end (live monitoring), but it must have an edge for the range currently loaded, or the range is being withheld.

Studying it

Present the same history with an explicit start and a “earlier” control, with auto-append and no copy, and with a silent truncation. Tasks: “find the earliest” and “is there anything earlier.” Independent variables: start mark, copy on the load edge, auto-append or not. Dependent variables: false “that’s all,” failure to find the start, time spent sliding with no end. Edgeless auto-append should produce the most “are we there yet” uncertainty.

Where it stops holding

A short timeline that fits on one screen already shows head and tail in the viewport. A live monitoring stream’s head is “now”; the tail may never come, but it should say this is live, not a closed history. A search-hit stream’s edge is the query, not a time-head; mixing the two makes people expect earlier hits in time that the query does not contain. Offline, “load more” must refuse and say why, not sit as a dead edge that does nothing when tapped.

Applying it

• Put “this is the earliest record” or an equivalent at the time-head; if the head is not loaded, put “show earlier.”
• Even with auto-append, the bottom needs two different edges: “loading” and “no more.”
• If open lands mid-stream, offer a jump to the head or to now.
• How to check: ask “is there anything earlier.” The answer must be readable from an edge on the surface, not guessed from whether this flick was the last.

Related

• Within the group: E4.21.1 A timeline lays out discrete events in time order · E4.21.2 When event density is uneven, ticks should not be equal linear intervals · E4.21.4 Key nodes must look different from ordinary events
• Adjacent: E5.07 Infinite scroll · E5.08 Load more buttons
• Search terms: feed edge · load more · timeline start