Photoware Interface Design for Better Photo Management

Abstract

Digital photography has largely replaced traditional film-based photography. Low cost and easy creation of digital files encourage people to take more photographs (photos) than before. Photo management involves selecting photos for placement into albums for future review and publishing them through different media in order to share with others. Photoware is designed to help people manage a large amount of digital photographs. To manage digital photos more efficiently, not only does software performance need to be improved, but also the usability of the software design needs to be refined. Well designed interfaces are one of the solutions to expedite the photo management tasks. The purpose of the study was to design a new photoware interface that would allow users to execute photo management tasks more efficiently. A three-phase study was designed to assess and compare Adobe Photoshop Album with Google Picasa. Phase one identified two major features that were problematic: 1) visibility of images with the photo tray features and 2) the non-existence of export templates that allow photo format settings to be placed into memory for future use. In Phase two, these two features were designed and implemented through paper and dynamic prototypes. Phase three consisted of usability testing of the prototypes. Thirty university students participated in a time-on-task study to measure the performance of the new interface. Also, user preferences were recorded through a post-task questionnaire. The users performed more quickly on the interface design with the two new features, as well as preferred them over the existing products when using mean values. However, the statistical correlation (Spearman - rank order correlation) between the two usability variables was not significant. The lack of correlation was supported by Frøkjær et al.’s view that usability measures should be considered independently. In other words, task time performance alone cannot be used to predict general user preference, because subjective preference could be affected by a range of other factors.