Paper Prototypes and Beyond

Article excerpt

ABSTRACT

Although approaches to user centered software development have existed for almost twenty years a rift still exists between theory and practice. In practice, many software projects are designed at the code level to the detriment of the endusers. Good Usability Engineering combines complex back-end functionalities with attractive, effective and efficient user interfaces. Successful interfaces minimize cognitive load and help users to achieve their goals. Goals can be defined in terms of intended outcomes which act as benchmarks for developing and testing functionality through prototypes. Paper-based prototyping bypasses the time and effort required to create a working, coded user interface. Instead, it relies on very simple tools like paper, scissors and stickers. However, to be a reliable guide, paper mock-ups need to model accurately the site's functionality and convey the right information.

This paper describes the challenges presented by a complex online information design project, an online research resource of over 45,000 records based on the catalogs of Exhibitions of the Royal Photographic Society 1870-1915. It describes how paper prototyping (used successfully previously) was used to address these challenges and reflects on the problems that came close to derailing the project this time and their impact on the design and the design process. It concludes by considering a digital alternative to paper prototyping that offers similar ease of use and low cost, combined with the ability to quickly generate interactive mock-ups that overcome some of the limitations of paper prototypes.

INTRODUCTION

This paper is about usability engineering and rapid prototyping in Communication Design. Although approaches to user-centered software development have existed for about twenty years a rift still exists between theory and practice (Holzinger, 2005). Many software projects are designed at code level, to the detriment of end users. "Generally the last thing you want to do when beginning to design an interactive system is write code" (Buxton, 2007, 240). Good usability engineering combines complex back-office functionality with attractive, effective and efficient user interfaces. Successful interfaces clearly signal the affordances they offer (Norman, 2002, 87-104; Krippendorff, 2006, 111-114). Affordances are the perceived and actual properties of something that determine how it could possibly be used (Gibson, 1979, 127-135). Interfaces with clear explicit affordances minimize cognitive load and help users to achieve their goals (Krug, 2000, Uff; Holzinger et al, 2008). Goals can be defined in terms of intended outcomes, and can be used as benchmarks or baselines (Brown, 2003) against which functionality of the design can be tested. If the intended outcomes of the design are clearly articulated, then the performance of the design can be tested to ascertain how well it supports those outcomes. In effect, all designs are tested after the sales launch. Sometimes they fail in use. An old but dramatic example was the Comet passenger jet aircraft. Launched in 1952 as the world's first jet airliner, the aircraft was withdrawn from service in 1953 because of serious structural cracks caused by metal fatigue (Dempster, 1960). Four years later when the redesigned aircraft was re-launched, the Boeing 707 had reached the market and went on to dominate domestic passenger jet aircraft sales. Less well-known but more recent examples are the circular mouse for the original iMac and the Power Mac G4 Cube (Buxton, 2007, 44). These examples show that performance failures don't have to be as dramatic as air crashes to matter. Norman (2002) describes a catalog of products including doorways, telephones, projectors, washing machines, refrigerators, radios, cookers, taps and hi-fi systems that frustrate their users. Also, designs don't have to fail in use to be unsuccessful. Well-known examples from the history of design and advertising are the Ford Edsel (Bonsall, 1992) and Strand cigarettes (Hackley, 2003). …