Web History Tools and Revisitation Support: A Survey of Existing Approaches and Directions

Abstract

Millions of web pages are visited, and revisited every day. On average, every second page loaded was already visited before by the same user — individual means for recurrence rates range between 20% and 72% (cf. p. 24). People revisit pages within a session or between parallel ones, they reuse web-based tools habitually, monitor specific content or resume interrupted sessions, and they want to re-find content after longer periods of time. Current history tools that support such revisits show unique and severe shortcomings. Often, revisits are cumbersome, more than necessary. This survey summarizes existing knowledge about revisitations on the web, and surveys the potential of graphic-based web history tools. A taxonomy of revisit-types distinguishes between short-, medium-, and long-term revisits, but also intra-and inter-session revisits. Assisted by a clear nomenclature this provides more clarity to the current discussion. The potential use of graphic-based tools is analyzed and discussed with respect to the found categories. The value of the current, mainly non-graphical history tools, such as back button, bookmarks, history list, search engines, and search bars is examined and related to the potential offered by graphic-based tools. The survey provides summaries of key studies and bodies of research for those who are interested in improving the web users' experience by simplifying the processes of going back to resources visited seconds, minutes, hours, weeks, or even months ago. It is meant for developers and researchers, browser and search engine producers, web usability professionals, and those who feel an irresistible urge to creatively innovate the web. The time has come to design and offer more appropriate history support. This survey aims at providing a foundation, as well as valuable ideas for doing so.