There have been at least three significant attempts in the last fifty years to comprehend what exactly is this text thing that we scholarly editors and textual critics work with. The initial wave was the Greg-Bowers New Bibliography which tried conscientiously to use all surviving witnesses as forensic evidence to reconstruct the author's intention. The text according to this view was ultimately a product of volition, and the task of the textual critic was a recuperative psycho-historico-linguistic one. The second attempt was marked by Continental inclusiveness and semiotic despair at identifying a single stable authoritative version. This despair produced the view that text was constituted by all recoverable manifestations of it. The third attempt came with the wide accessibility of digital recording of texts and has had two branches. The first, spawned by poststructuralist literary theory (and perhaps partly by childhood Lego-deprivation) has seen electronic text as a field of liberationist politics in which readers of the interactive documents emerge at last from their slavery to author(itarianism), while the second (which, paradoxically, is disciplinarian rather than liberationist) has seen text as a conceptual structure, an "ordered hierarchy of content objects" or OHCO What these two apparently disparate views have in common is an approach to the nature of text which focuses on the potential of the expressive medium. The views are driven not by what an author might have tried to say (New Bibliography), nor what all the witnesses record (théorie de texte), but rather what the medium makes it possible to say.