1 Drag-and-drop and Perceptible Affordances issues within Personal Web Portals: A iGoogle Case Study Jacques Chueke jacques.chueke.1@city.ac.uk George Buchanan george.buchanan.1@city.ac.uk Stephanie Wilson stephanie.wilson.1@city.ac.uk City University London - Centre for HCI Design Northampton Square, London, UK, EC1V 0HB ABSTRACT In this research we investigate Perceptible Affordances and their role in drag-and-drop interactions within the iGoogle Personal Web Portal. A usability evaluation with Eye Tracking was conducted in order to better understand how this website signals the possibility for Drag-and-drop interaction with its content. Our protocol for analysis also included participants thinking aloud, so we could better grasp their strategies for identifying potential interactions. We verified that expertise influences how able a user is to identify and understand cues for interaction within an interface. In the case of personal web portals, a general lack of familiarity with this type of website hinders the effective use of these systems by new users. Clearer and stronger cues for perceptible affordance are needed to ensure that users can quickly adopt and adapt these systems. Author Keywords Eye Tracking, Personal Web Portal, Drag-and-drop, Perceptible Affordances, Dashboard, Widget. INTRODUCTION In this paper, we investigate the role of Perceptible Affordances in users’ interactions with web pages. It has been established that perceptible affordances shape the functions that users anticipate a system may have, and subsequently influence which features are used, and how. Poor perceptible affordance misleads users, resulting in them failing to identify some features, or suffering problems in using specific tools. In Gaver’s (1991) words, “…Perceptible affordances are inter-referential: the attributes of the object relevant for action are available for perception. What is perceived is what is acted upon. From this point of view, interfaces may offer perceptible affordances because they can offer information about objects which may be acted upon, we can understand displays in terms of the subset of normally available perceptual information various media make available for various actions.” Drag and drop, a very common interaction within modern graphic user interfaces, is known to suffer from poor perceptible affordances. For example, Nielsen (2008) comments “…Drag-and-drop designs are often the worst offenders when it's not apparent that something can be dragged or where something can be dropped”. Spool (2005) adds, “The problem with drag-and-drop is that it doesn’t have any affordances. You can’t tell when you’ve encountered an element that is dragable. The result is that this powerful capability is often only known by developers”. The problems first encountered in desktop usability are now recurring on the web (See Lunn: 2010). Personal Web Portals (e.g. iGoogle, My Yahoo! and Netvibes) provide users with a single ‘dashboard’ web page that acts as an access point for a rich variety of information (e.g. the time, weather, news, email, etc.). These sites have been identified (e.g. by Norman: 1999) as suffering a range of usability problems, including drag-and-drop. To better understand how (bad) perceptible affordance impacts how users adopt a new and unfamiliar tool, we have taken personal web portals as a test case, and investigated them using a protocol that includes both eye tracking and think-aloud methods of data capture. Eye tracking technology was selected because it provides quantitative information on what people observed during a test session. By analysing this data set it is possible to evaluate and theorize about how a group of users obtained the gist of possible interactions within a specific website – in this research we were specifically concerned about control features and drag-and-drop interactions. Nielsen (2009: 05) argues that when eye tracking is used for usability testing “…we learn finer points about what draws attention and why. And by studying what people look at and do not look at, we are able to glean more detailed insights about how the Web user works his or her way through usable and unusable designs”. We anticipated that expertise would influence a user’s ability to identify the signals given by the interface as to what features it includes, and how these could be accessed and/or used. From our study, we have arrived at some initial findings that we hope to generalize into a more generic and wider understanding of perceptible affordances. USABILITY EVALUATION Ten (10) subjects were tested on iGoogle website on May 5, 2011. There was a pre-task demographics questionnaire, asking about age, expertise with computers and Internet frequency of use. iGoogle was observed by three (03) beginner, four (04) intermediate and three (03) advanced subjects with Internet and computers expertise. In order to avoid biased interaction the selected participants were unaware of the website and its features. The selected browser was Mozilla Firefox version 4.0.1 with screen resolution of 1280 x 1024 in a PC running Windows 7 Professional Edition. The Internet connection was City University LAN Network (speed of 100Mb/sec) and Tobii x60 was the Eye Tracking device used during the experiment. The software for analysis used was Tobii Studio v. 2.3.0.0