Scholarly Introduction

Orlando's approach to humanities computing provides one answer to the question of how computing can assist in the creation and use of a literary history. Its technologies were developed in the context of on-going literary research. Its markup or tagging system structures a body of knowledge which is characteristically interpretive and evaluative. Although literary history is heavily dependent on quantifiable and factual statements, it is most interested in the unquantifiable. Biographical accounts of authors consist not only of birth, death and publication dates, but also of information (and sometimes speculation) about such matters as political allegiances, religious beliefs, race, class, and sexuality. Critical or expository accounts of texts depend on complex as well as relatively simple concepts, and the application or interpretation of the former may differ sharply among different critics or groups. Orlando's tagsets represent and make searchable both relatively straightforward and highly interpreted information. It is possible to search such fairly simple concepts as Anthologization, Contract, Copyright, or Earnings, and also ones, such as Motifs, Techniques, or Narrative Voice, on which the views of its readers, like those of its subjects, are bound to be various.

Orlando began at a moment when some, like Rosanne G. Potter, were complaining that those working at computer analysis of texts, more and more absorbed in scientific methods, had "forgotten they were seeking information about literature" and that such endeavours needed to pursued in dialogue with "the kind of analysis that only the unaided human mind can apply." Humanities computing is haunted by information-processing. Pressure towards 'precision' in tag 'meanings' and applications creates the risk of undermining the ability of traditional literary scholarship to handle the multivalent, ambiguous, and unclassifiable. What are needed, according to many digital humanists, are more sophisticated ways of dealing with text electronically. This is a vast field of inquiry that invites many different approaches. Given the divergent methods of literary history, no single system will serve every need, nor be sufficiently uncontroversial to be both definitive and interesting.

Orlando has sought to produce a system of electronic text markup that is both extensive and sufficiently wide-ranging to address material from several centuries and the entire range of literary genres. It aims to represent a literary history grounded in the texts and contexts studied, offering as part of its representational structure a flexible set of interpretive rubrics that avoids the liabilities associated with rigid taxonomies, and making its readers active partners in its literary historical endeavours.

Orlando is built using SGML (Standard Generalized Markup Language), a 'metalanguage' used to demarcate text in documents according to consistent principles in such a way that they can be processed by computers. Orlando's tagging language, designed to encode the literary history of women, is composed of some tags common to other projects together with many new tags, unique here. It includes 205 tags, 114 attributes, and 635 attribute values.

The project chose SGML because it was an international standard, because it can run on various programmes and is therefore unlikely to become obsolete, and because it was emerging (and is now established) as the standard tool for archive-quality encoding of scholarly texts in the humanities. The Text Encoding Initiative has established a set of standard protocols for the editing of primary texts. Orlando's aim to encode critical materials with subject-specific markup (rather than the formal features and variants of existing primary texts) meant that it made sense to depart from the TEI standard as it existed when the project was specifying its tagset. But Orlando has departed from the TEI only when it is necessary to give added, subject-specific value to the tagging scheme. Its retention of key TEI tags provides the potential for interlinkages between Orlando and other bodies of SGML-encoded texts, such as the pioneering Women Writers Project. While Orlando was underway, XML or Extensible Markup Language, a streamlined and simplified version of SGML emerged: it facilitated delivery of the textbase via the World Wide Web.