Towards Basic Categories for Describing Properties of Texts in a Corpus

The paper discusses the basic principles for describing properties of texts to be stored in a corpus and suggests the standard that is used in the majority of corpora developed at the University of Leeds and can be potentially employed for describing texts in any corpus collecting activity. The standard defines the minimal subset of tags and attributes that are necessary for describing texts stored in a corpus. The proposed text typology helps to position a corpus under development with respect to a reference corpus covering all possible features by explicit selection of a subset of features to be considered in the study.