[jstogin]

Some graduate students will likely spend a substantial part of their PhD understanding this paper. They will learn a lot in the process, and then they can contribute by either extending the result or finding a way to simplify a part of the proof. Or if they have a (very) related interest, they may be able to adapt some of the techniques to the problem they're interested in. Slowly over time, through this process, the knowledge might diffuse to other less related areas.