Overview

The principle of metadata filtering suggests theimportance of reconceptualising current metadatainterfaces from a single expert-filtered andconsistently displayed metadata content descriptionto a contextually relevant, reconfigurable anduser-driven interface. This post-doc filtering isthus both user-led and user-focused, thus betteraddressing users’ diverse needs and requirementsthan traditional (expert-led/single display)interfaces. The principle offers flexibility andfacilitates the serendipitous discovery ofinformation resources. However, in doing so, itshould also be noted that this does not precludefiltering metadata so that only currentstandards-based, objectivist metadata is used andpresented in a standard OPAC, as this may be whatthe users need. Multi-field, stand-alone andnon-web-based Online Public Access Catalogues(OPACs) are complex and outdated (Ballard andBlaine, 2011).

Users prefer web-based and straightforward, integratedinterfaces with facet filters (Ballard and Blaine,2011). As Alemu and Stevens (2015) indicate, ‘Theusers’ preponderance to traverse across informationsources (mediums), supported by triangulatedinformation from disparate sources, necessitatesthat the focus is placed on interlinking ideascontained in various sources such as databases,e-books and e-journals. The freedom of informationfrom its constraining medium through seamlessinterlinking should help users to connect tointer-disciplinary and multi-disciplinary domains ofknowledge. However, this requires a new conceptualapproach to information management.’

User information search behaviours

Bates (1989) argues that users’ search and browsingbehaviours are non-linear, characterised by herberry picking model. Instead, it is iterative,dynamic and evolving. They employ various techniquessuch as searching, browsing, footnote chasing,citation searching, journal run, area scanning,subject searches in bibliographic databases andauthor searching (Bates, 1989). Furthermore, Batesargues that ‘as the sizes and variety of databasesgrow and the power of search interfaces increases,users will more and more expect to be able to searchautomated information stores in ways that arecomfortable and familiar to them’. Wilson'sinformation model (2016) also agrees with thisassessment. Thus, search, and discovery systems needto consider user behaviours and mental models.

Users prefer seamless access to full-text articles andimmediate access to electronic resources (Alemu,2014). In addition, the convenience of access(place), immediacy and currency (up-to-dateness) arevital to them.