The interesting thing about this experiment is not the fact that the bacteria developed adaptatory mutations in a given gene to deal with increased heat, as much as the fact that when the experiment was repeated, the same mutations were observed in the same gene in the DOMINANT bacterial strains at various time points.

Think about it.

It suggests that in a given species, there seems to be an optimal adaptation to a given set of environmental conditions, that is determined by the genetic make up of the population prior to attaining those conditions, and the path that the changes in those conditions take through time.

While that's just a paraphrasing of the definition of evolution, I don't believe anyone has shown that a given evolutionary path is experimentally reproducible.

The title of the aritcle is misleading though. Evolution is not being predicted. But if you've seen it once, it may be predictable.