Abstract
Some fundamental questions about the nature of texts have been asked by members of the encoding community as they have struggled to define what it is that they are encoding. This ought not to be surprising. The act of encoding involves a blatant intervention in text-files of a kind that scholarly editors in the print domain are normally shielded from, however heroic their emendation of corrupted wording may be. They have typically treated many aspects of the physical presentation of text as irrelevant to their pursuit. However, this self-preserving instinct finds itself in a tighter corner in the electronic domain where complete specification is crucial for computer processing.