Facebook announced its new Graph Search on 01/15/2013:
With Graph Search, Facebook becomes a little more like a DATABASE for its users (many dimensions, time is less relevant) - as opposed to the present massive 1-dimensional data stream (present -> past):
This is how Facebook engineers who developed Graph search also expained the project to Wired: "“It’s like Facebook is this big database and you’re doing a lookup on the results that match."
For me, it is a new example of how database/narrative opposition in digital media (which accompanied it from the start) works now in web apps and services. Database is "spatial" - you can search and use other operations to get records using all fields, with time stamp being just one dimension among others.
A narrative (and a timeline) strongly mark a single dimension: time. Things become related if they occur close to each other in time. (In a timeline, other dimensions of similarity and possible links between far away points are not visible). Posts and replies on FB is one example (do you often comment on posts from a month before?); film editing is another example- a sequence of shots made to work together visually. (Of course, filmmakers and other artists also play with the functions and limits of our memory, making connections between events that can be far apart in a narrative stream.)
P.S. Connections to data visualization:
From this perspective, a scatter plot is more close to a database logic - especially with interactive options (and extended to graph matrix, etc.) ; a line graph is more like a narrative, with moods, anticipation or other analog qualities going up and down.
For an example that visualizes the linear/spatial opposition, see this classic project by Martin Wattenberg (2001): http://www.turbulence.org/Works/song/gallery/gallery.html
For one possible way to visualize a film combining narrative and spatial representations to help us study the connections between shots, sequences and themes, see these visualizations from my new article "Visualizing Vertov"