We introduce and define ‘heterotic computing’ as a combination of two or more computational systems such that they provide an advantage over either substrate used separately. This first requires a definition of physical computation. We take the framework in Horsman et al. (Proc. R. Soc. A 470, 20140182. doi:10.1098/rspa.2014.0182), now known as abstract-representation theory, then outline how to compose such computational systems. We use examples to illustrate the ubiquity of heterotic computing, and to discuss the issues raised when one or more of the substrates is not a conventional silicon-based computer. We briefly outline the requirements for a proper theoretical treatment of heterotic computational systems, and the advantages such a theory would provide.
@article(Kendon++2015, author = "Viv Kendon and Angelika Sebald and Susan Stepney", title = "Heterotic computing: past, present and future", journal = "Phil. Trans. R. Soc. A", volume = 373, pages = "20140225", doi = "10.1098/rsta.2014.0225", year = 2015 )