Making Failure a Scientific Virtue

This is an idea that's time is well passed due:

Even if a scientist’s data don’t deliver the hoped-for results, they still have uses. This is especially the case today, when some of the most compelling discoveries have come from meta-studies, in which statisticians sift the data from several previous papers to reach their conclusions. “…Your dead end may be another scientist’s missing link, the elusive chunk of data they needed,” says Mr. Goetz.