There is a significant shift from the model where ‘all data is salient,’ for example, each entry on a calendar is a relevant appointment, to a model of being able to recognize different kinds of data and appropriate actions related to specific data types. The focus level upshifts to the correlation, trend, and anomaly level of big data abstractions rather than on the unitary level of the data flows themselves.
Daily quantified self-tracking data for example may be useful from a longitudinal perspective and might not need to be reviewed unless there is an anomaly. Another example is that the relevant action might be looking for correlations across multiple data streams. There could be potential linkage between coffee consumption, social interaction, and mood per as this Sen.se multiviz project investigates, finding some correlation between social interaction and mood.
Discussed at greater length in: Swan, M. Sensor Mania! The Internet of Things, Wearable Computing, Objective Metrics, and the Quantified Self 2.0. J Sens Actuator Netw 2012, 1(3), 217-253.