‘When I use a word,’ Humpty Dumpty said in rather a scornful tone, ‘it means just what I choose it to mean — neither more nor less.’ One wonders what Humpty might have said about compound words, such as ‘expression profiling’, for example. Only a few years ago, life was simple: rapid progress was being made towards defining the identities of specific genes, and the development of high-throughput methods for simultaneously analyzing genes in parallel led to the development of the concept of expression profiles which could uniquely represent the individual cell types and cellular states found within tissues and organs (Hughes et al. 2000). However, more recently, the primary concepts underpinning our understanding of the meaning of gene expression have blurred somewhat. Defining the physical bounds for a gene, for example, is no longer a question simply of determining mileposts up and downstream from the point of initiation of transcription, since these may be modulated and influenced by chromosomal context and chromatin state. Again, defining expression no longer solely reflects the concept of translation of the encoded information content of an mRNA into its cognate protein; we must include in our understanding the concepts of RNA processing, transport, sequestration, and degradation, and the roles of genes producing transcripts that are non-coding. Finally, it is becoming clear that specific sequences of DNA can interact via a number of mechanisms that influence expression, including an-tisense transcription, the occurrence of alternative splicing, and the production of small RNA molecules. The complexities can be maddening, as Alice in Wonderland observed: ‘But I don't want to go among mad people!’ ‘Oh, you can't help that,’ said the cat. ‘We're all mad here!’.