Traditionally, collections of texts are digitally represented as a set of documents containing the text along with some kind of markup to define extra information, like metadata, annotations, etc. We propose a different ...