Robust annotation positioning in digital documents

Reviewer: Maxine Cohen

This paper reports the results of two studies examining user expectations of robust annotation positioning in modified documents. The authors present an interesting case for one of the problems encountered by users as we move our world ever deeper into the electronic management and revision of documents. The authors clearly define the terms needed to better understand the annotation positioning problem. The first issue they address is how digital annotations work. The second issue is how documents are modified. Starting with a discussion of the way paper based annotations work, the authors then relate it to the digital annotation arena. They introduce the concept of an anchor and of a context related to a content area, both of which are needed to create a robust annotation. They then discuss the issue of modification, and present a modification classification scheme describing three types of modifications: deletes, rewords, and moves. They describe what it means when an annotation is “orphaned” (when an online document changes and the link to the annotation is lost). The authors include a good summary of the current research and challenges in the area, including reporting on several studies of users using an electronic annotation system. They acknowledge many of the commercial systems already performing electronic annotations, and describe how they work. They provide a good set of references. Their study extends the work of others further by delving into peoples’ expectations in dealing with annotations, using those expectations to further fine-tune the annotation algorithms. The article contains several interesting and excellent tables that further explain the authors’ results and findings. Two trends seem to emerge from their studies: the greater the percentage of the annotation anchor text found, the more satisfied people are; and the more drastic the modifications to the anchor text, the less satisfied people were when the annotation anchor was found. This second finding was a little surprising. It seems that with significant changes, participants were all right if the annotation was orphaned. Keywords, proper names, and quotations are important factors to figure into the annotation algorithm. Online Computing Reviews Service