Resource discovery services for digital library have evolved significantly. There is an increasing use of dynamic user interface. New ways of user interactions are also emerging. Faceted searching for example provides a “navigational metaphor” for boolean search operations. It also results in greater user satisfaction (Olson 2007). AquaBrowser is a leading library product which provides faceted searching and new resource discovery features based on emerging interaction design patterns ('fresh, modern interface' according SerialsSolutions). These features appear to support better ways of searching: fast drill-down of results, enhanced search context.
An interaction design pattern we explored earlier this year, through a UX2.0 study (heuristic inspection) is the "Word Cloud" feature which according to SerialsSolution provides context-sensitive exploration. The unique UI is essentially a composite design pattern based on word (tag) cloud and spatial navigation - a term coined by the Nielsen Norman Group. AquaBrowser's tag cloud is unique as it does not utilise the typical size effect and rely on colours showing different types of word association. It also uses spatial navigation in combinant - tags move or reorganising during user interactions. Despite its popularity, the use of tag cloud even in its most basic form, remains contentious and nascent at least from usability perspectives. A recent statement from the Nielsen Norman Usability Week has urged sites to use tag cloud with caution and certainly not to use it for mainstream purposes. Furthermore this technique may either become a standard design or expire in a few years. On the contrary, recent studies on AquaBrowser have confounded the usual dismissive views of word cloud, suggesting potential for serendipitous discovery even for domain experts (Olson 2007). Moreover, word cloud has been an active subject in digital library research and development, e.g. as a visualisation device to gauge content relevance 'as a glance' (Gottron 2009).
Word Cloud is only one of many aspects of AquaBrowser explored in the heuristics study. My colleague Lorraine who performed the heuristic inspection, has blogged about the AquaBrowser inspection findings - also see the full report of the study.
Word Cloud in the Main Library, Edinburgh University
User Study for AquaBrowser and UX2.0
A consistent message arising from the heuristic inspection has been the need for elaborated user studies to gain insights and richer understandings surrounding the use of new interface features such as the AquaBrowser Word Cloud. In addition to usability, other evaluative aspects such as the usefulness of UIs and functions also merit further investigation. This is certainly true for the UX2.0 project. The Olson study (ibid) of AquaBrowser in particular, has several limitations. First, the participants were all domain experts (PhD researchers) with no experience of using AquaBrowser (novelty bias). Second, the test environment did not involve users performing tasks in real-use contexts.
To address some of these issues, I have submitted a proposal entitled AquaBrowser User Experience (AquaBrowserUX) to the JISC Enhancing Library Management Systems programme. The proposal has been funded. We are beginning a user study, from 19th April 2010 for a duration of six months. The main aim of the study is:
To evaluate the user experience of AquaBrowser at the University of Edinburgh.
The study involves holistic evaluation centred on both fundamental and practical use aspects of resource discovery. It addresses the need to better understand user needs and workflows - a requirement identified from the recent JISC/SCONUL Library Management Systems study. The project has three objectives:
- To undertake user research and persona development
- To evaluate the usefulness of AquaBrowser
- To evaluate the usability of AquaBrowser
The above work will be carried out by a team of usability consultants (university and external staff) working both online and on-site at the university library. It is a collaboration between the National e-Science Centre (NeSC) and Information Services of the university, managed under the UX2.0 initiative.
The user study dovetails with a similar user study of UX2.0. It will be extended (additional and alternative survey questions) to address a need relating to user-centred design (UCD) development. Part of UX2.0's work involves low-fi/content-rich UI prototypes evaluation. This makes a basic form of user research necessary, in particular the derivation digital library personas (see below) for UI prototypes development. Developing new user experience is also a key objective of UX2.0. The development goes beyond the technical aspects of repository. The project evaluates human experience through usability and usefulness studies of the broader remits of digital libraries, in addition to the electronic bookshelf, 'search and retrieve' services. Therefore, the user research will also provide valuable empirical data and rich insights in terms of incorporating broader form of user services such as social networking.
Qualitative User Research and Usefulness Evaluation (Objectives 1, 2)
User research and usefulness evaluation are currently being undertaken. Ethnographic study of library users is the key mechanism for gathering various forms of empirical data. The work primarily involves user interviews at the Edinburgh University Library, Main Library in George Square (see below). The study is directed at a broad base of library users of different roles, academic background, system exposure and information seeking behaviours. The interviews will survey the general usage behaviour of library services as well as elicit user perspectives and recommendations for AquaBrowser, gauging its usefulness in terms of function and fitness for purpose (cf. library use contexts). In addition, supplementary online surveys will be conducted for those users who are less inclined to library visits but use library services online.
In addition to user interviews, field studies at the Main Library are being devised. The study has two scopes: 1) general observation relating to the general usage behaviours of library services for the investigation of digital libraries as user and social environments, 2) the use of AquaBrowser in library, focusing on the patterns/behaviour of users at library terminals, involving real-use contexts in which users supplying their own search tasks.
Usability Testing (Objective 3)
Using the personas derived from the user research, users (per persona) will be recruited for usability testing of the AquaBrowser interface. The test will be informed by UX2.0 heuristic inspection. Both faceted searching and the word cloud functions are expected to be a key focus of this test.
The user study will yield empirical data required for determining the usability and usefulness of AquaBrowser, as well as provide rich insights for developing broader digital library services (environments). The data will be analysed and synthesised for a key output - a summative 'AquaBrowser User Experience' case study. In addition, a set of digital library personas will be produced. Persona is a hypothetical individual representing a discernible character facet of real users. It is a form of user modelling in the human-computer interaction (HCI) field. Compared to other methods (e.g. ‘stick figure’ use cases), persona is more user-centric, realistic and entails rich archetypal information: background, behaviour pattern and personal motivation. This information enables library stakeholders both at the university and the JISC Community, to better judge user goals and needs, and to overcome incorrect system assumptions. Persona will also be useful for system development and UI prototyping.
Keeping up to date and in touch
Progress of the project will be disseminated through a series of forthcoming blogs, Twitter and a summative event to report on the project results. The unique project tag (hashtag for Twitter): aquabrowserux. A project wiki is also available: http://bit.ly/aquabrowserux. Contribution on all aspects of this project are welcome. Please feel free to leave a comment and get in touch.