I have just read an amazing book “Shibumi” by Trevanian (a.k.a. Rodney William Whitaker) probably the best spy novel I have read so far. In the book, a big data computer (called Fat Boy) is operated by a “data scientist” (although is not called that way). I enjoyed very much the following paragraph, an analogy of the emptiness of big data without insight and also a musing about how difficult is to find relationships from activity data (the kind of research we do!)
“Order the list emotionally. Go for indices indicating love, friendship, trust—this sort of thing. Go from closest to most distant.”
The First Assistant’s eyes shone as he took a deep breath and lightly rubbed his fingers together. This was a fine challenge demanding console virtuosity. Love, friendship, trust—these imprecisions and shadows could not be located through approaches resembling the Schliemann Back-bit and Non-bit Theory. No computer, not even Fat Boy, can respond to such rubrics directly. Questions have to be phrased in terms of nonfrequency counts and non sequitur exchange relationships. In its simplest form, actions performed for no measurable reason, or contrary to linear logic, might indicate such underlying motives as love or friendship or trust. But great care had to be exercised, because identical actions could derive from hate, insanity, or blackmail. Moreover, in the case of love, the nature of the action seldom helps to identify its motivational impulse. Particularly difficult is separating love from blackmail.
It was a delicious assignment, infinitely complicated. As he began to insert the first probes into Fat Boy, the First Assistant’s shoulders twisted back and forth, as though he were guiding a pinball with body-english.