Improving Document Navigation Using Space-Filling Thumbnails
Jason Alexander
Department of Computer Science
University of Canterbury
Abstract
Space-Filling Thumbnails (SFT) is a new technique for document navigation that disposes of traditional scrollbars. SFT provides a matrix of thumbnail images, one for every page of the document, on a single screen. Pages can be selected for viewing by using a mouse.
A prototype Space-Filling Thumbnails interface has been designed, developed and implemented. Six further document navigation systems have been implemented to enable comparative evaluations.
Three formal experiments have been carried out, all of which have shown Space-Filling Thumbnails to be more efficient than other research and mainstream document navigation systems. The first experiment (n = 13) showed SFT to be significantly faster than all other document navigation interfaces for both visual search and spatial re-acquisition tasks. Experiment Two (n = 32) showed that SFT is robust and faster than Thumbnail-Enhanced Scrollbars (TES) for visual and spatial search tasks in documents of differing lengths. Lastly, in Experiment Three (n = 12), the possible “Achilles’ Heel” of SFT was tested—long documents. SFT proved to be more efficient than TES even in its “worst-case scenario” situation—a 300 page book. Subjective measures from these experiments showed favour for SFT.
We have shown that Space-Filling Thumbnails outperforms traditional document navigation systems through the exploitation of users’ spatial memory. The swiftness of visual and spatial searches in SFT have proved that this system has the potential to revolutionise document navigation, as a valuable addition to, if not replacement for, scrollbars. Acknowledgements The author would like to thank his supervisor Andy Cockburn for his inspiration, enthusiasm and guidance throughout this project. He would also like to thank his family, staff members and fellow post-graduate students who have shown interest and offered suppor. All of the participants involved in the experiments also deserve thanks, for giving up their time and providing valuable feedback.
