The question should not be whether the technology is foreseeable; rather the question should be how difficult is it to identify embodiments and write a claim that covers them? Under our refinement theory, the DOE is inappropriate when refinement costs are low vis-เ-vis later-developed technology. It must seem counterintuitive to non-patent lawyers that refinement costs could be low, but often patent applicants can capture unforeseen embodiments through greater conceptual effort and the use of various claim drafting strategies.32 Application of the DOE to later-developed technology can be justified only as a means of discouraging excessive investment in refinement. In other words, the doctrine should be used to block socially wasteful expenditures when inventors find it profitable to exert considerable effort identifying and claiming later-developed embodiments sim-ply to preempt entry by a competitor. Surprisingly, this rules out application of the DOE to startlingly new equivalents, because these types of equivalents are conceptually unattainable no matter the amount of time and money spent on refinement efforts. Patent applicants would not refine their claims to cover these equivalents, and inventors’ incentives are not much affected by a minute probability of loss of effective patent protection.