Copy-number variation (CNV)--the presence of additional or missing segments of chromosomes in some individuals--has been found to be abundant in humans and adds another dimension of variation to the genome. Copy-number variants have already been associated with some diseases and disease susceptibilities and are likely to prove as significant as sequence polymorphisms in this respect. Changes in copy number of parts of the genome are known to be a feature of many cancers, and their analysis is expected to reveal genes involved in carcinogenesis. This article will present a somewhat biased and occasionally speculative discussion of the current and future significance of CNV with a particular focus on the potential of molecular copy-number counting in the analysis of small, damaged or heterogeneous samples.