Take this for example:

import rerun as rr

rr.init("rerun_example_points3d", spawn=True)

rr.log("/", rr.Clear(recursive=True), static=True)

rr.set_time_sequence("frame", 1)
rr.log("/points", rr.Points3D([[0, 0, 0]], radii=0.5))

rr.set_time_sequence("frame", 2)
rr.log("/points", rr.Points3D([[0, 1, 0]], radii=0.5))

rr.set_time_sequence("frame", 3)
rr.log("/points", rr.Points3D([[0, 2, 0]], radii=0.5))

What would you expect to see in the viewer at frame=3?

If you guessed "nothing", you're right (although that doesn't seem very useful...):

What about frame=0..3?

If you guessed "3 points", you're right (although at this point you should be frantically scratching your head already):

Now what if I asked you about Dataframe(0..3)? What about Dataframe(0..3, SparseFill::LatestAtGlobal)?! 😱

Anyway, it's extremely niche, so definitely not the highest prio right now.