A theoretical framework for designing and evaluating semi-structured document triage interfaces

Abstract

The authors of this chapter (Fernando Loizides and Aekaterini Mavri) consider several topics that will be of interest to readers. A notable feature of it is that they back up their recommendations with references to research evidence. Instead of the usual 'how to' guidelines that can be found in manuals of information architecture, this chapter explains why designers should pay attention to certain elements of information architecture, for example, and refers to the research evidence. As the authors themselves emphasise, 'user studies' is a huge area of research, and it is impossible to refer to all the underpinning evidence. Another problem is that 'user studies' means different things to different disciplines with an interest in the information user. The dialogue project (Dervin and Reinhard, 2006; Dervin, Reinhard and Shen, 2006) examined convergences and divergences in how three fields (library and information science, human-computer interaction and communication and media studies) looked at users and each other. The authors concluded that proper shared dialogue was desirable, but rare. The structures for communication across research and practice, and between the disciplines, did not make for mutual understanding, easy discussion, or - frankly - the understanding of users or audiences that the researchers (and practitioners) truly want. Power, prestige and a dependence on 'authority' get in the way. With that in mind, and without taking away from the excellent presentation of research evidence in this chapter, you should remember that you may need to approach the evidence in this chapter (and other chapters in this book) with the sense-making questions set out in the introduction. What are my reactions to this evidence? What do I agree with? What do I disagree with? What else do I need to know or experience to make further progress with this topic? There are several themes in this chapter. First, there is the idea of the semi-structured document. The authors' emphasis is on journal articles, patents and similar research study formats, and how these may be presented to readers to help them evaluate their contents quickly and successfully. There is considerable research on the difference in comprehension of reading text (usually linear text) on screen versus paper (e.g. Mangen, Walgermo and Bronnick, 2013). On-screen reading is different as it seems to be more difficult for readers to navigate some types of documents, to check whether they are moving around effectively, and whether they have a good spatial mental representation when reading on screen. Semi-structured documents such as journal articles usually have a standard format - introduction, methods, results and discussion - which can support navigation. Readers know what to expect in particular sections, and the order in which they expect to find things. Nevertheless, readers may have all sorts of different purposes when trying to find and extract information from an online semi-structured document and this chapter discusses the importance of providing a table of contents alongside the full text, and the need to deal with captions, images, tables and figures. It provides an overview of the research evidence and covers information architecture, external factors such as accessibility, and assistive tools.