Link services or link agents?

Abstract

A general link service for the WWW has been used within an Electronic Libraries' project. Experience using it shows that as the links become increasingly interesting to the user, processing them becomes increasingly expensive. Eventually textual analysis, ontological services and remote database lookups conflict with the goal of prompt delivery of documents. This paper summarizes the history of the Link Service software behind the Open Journal project together with the kind of links that it has been used to produce. Building on this work it then discusses how the paradigm, architecture and user interface of the DLS have been newly modified both in response to user feedback and also to allow more linking facilities to be added to the WWW environment. We then introduce AgentDLS, an agent-style system that offers suggestions to help the user's browsing and information discovery activities.

The impact of Web self-efficacy, age, and Web experience on bookmark manipulation

“…The same trend was observed in the present study, where Web self-efficacy and bookmark manipulation had a positive correlation of coefficient of 0.45, indicating that the higher the level of ability, the more variety of behavior the users engage in to organize their bookmarks. Similarly, previous studies found a strong link between experience and bookmark organization [9], which the present study also observed as a significant positive direct link (0.18), as well as mediated through Web self-efficacy. Again, the positive correlation suggested that the more experienced the users are, the more bookmark manipulations were performed.…”

Hypermedia on the Web

“…This architectural choice is extremely flexible and powerful, making many currently unavailable functionalities possible. For instance, this enables different link sets over the same content for different audiences and tasks [Carr 1998], [Lewis 1999]. Third-party individuals can provide links and create guided tours over documents owned by others.…”

The evolution of hypertext link services

“…Hyper-G [Andrews 1995a] provided a solution for the problem of broken links in the Web's open environment in the form of a safe but restricted environment with guaranteed consistency for link editing. By contrast, the Distributed Link Service consisted of an unenclosed environment providing a navigational overlay to Web pages based on within-node text analysis to identify implicit link opportunities such as key phrases, personal names and bibliographic citations [Carr 1998]. …”