Some unanswered questions:

1. Is the nucleation area (area perhaps showing dilatancy) any bigger for an M8 than an M1?
2. Does dilatancy start years before an earthquake or millseconds?
3. Is dilatancy a perceptible effect on a fault with a weak, perhaps over-pressured layer of gouge?

Dilatancy as summarized by Bolt is so unconfirmed by data that most scientists have moved on to other, more-likely but still long-shot precursors.