> that scales with the number of videos, not their length. That seems pausible, but if so, the entire calculation would have to be redone from scratch, with qu

> that scales with the number of videos, not their length. That seems pausible, but if so, the entire calculation would have to be redone from scratch, with qualitatively (not quantitatively; different units, not just different values) different numbers, so bringing "1.6 million" into it is still a misleading non sequitor.