Determining gene function by using transgenesis in the mouse has been immensely popular amongst researchers since the introduction of the techniques in the 1980s. However, we are still a long way from knowing the function of the majority of the genes in the genome. It is becoming increasingly accepted that there needs to be a coherent programme of systematic mutation of every protein-coding gene in the mouse genome. The target is therefore to generate approximately 20,000-25,000 gene knockout mouse models. This is an ambitious aim but one which is vital to enhance our understanding of physiology, development and disease. Recently a number of programmes have been established to meet this aim. A particular feature of these large co-ordinated methods for generating genetically modified mice is that they may obviate the need for individual researchers to perform the onerous process of generating the mutant mouse of their gene of interest - in the future we may be able to order our mouse model of interest with the same ease with which we order laboratory reagents and supplies today.