University of Southampton Research Repository ePrints

FACULTY OF ENGINEERING AND APPLIED SCIENCE DEPARTMENT OF ELECTRONICS AND COMPUTER SCIENCE A mini-thesis submitted for transfer from MPhil to PhD by Timothy J. Miles-Board Early hypertext visionaries proposed entire online archives of the world’s literature, with everything associatively linked to everything else. Today, the most widespread hypertext system is the World-Wide Web (WWW), a publicly accessible and globally distributed medium. However, the WWW is not living up to the promise of hypertext associativity the majority of hypertext linking on theWWW is estimated to be intended for navigational purposes only. WWW authors typically have new ideas to contribute, and assert particular relationships between these and existing ideas already published in order to demonstrate both the reliability of the conceptual foundation being built on, and the innovation and significance of the new ideas. However, these associations are rarely rendered as associative links which seamlessly link the new material into the global context. This research investigates the possibility of capturing these implicit inter-document associations through annotation, and then using these annotations to assist the hyperauthoring process. The hypothesis of this work is therefore that by capturing inter-document associations through annotation, a better hyperauthoring process will result, both in terms of the quality and coverage of the new writing, and in terms of the seamless (associative) integration with the global context, helping the WWW evolve to achieve all of its potential hypertextual richness. The Annotation LInking ENvironment (ALIEN) has been implemented to demonstrate techniques for capturing inter-document associations made by an author whilst reading, using free form annotations. Further work proposed includes the re-purposing of these captured associations to assist the authoring and linking processes through dynamic visualisation of the association structures “as-you-type”, and automatic associative linking.