Website Evaluation System: Collaboratively Discovering what makes a Website Good
Jon Dron, Phil Siviter, Chris Boyne, Richard Mitchell, University of Brighton, United Kingdom
WebNet World Conference on the WWW and Internet, in San Antonio, Texas Publisher: Association for the Advancement of Computing in Education (AACE), Chesapeake, VA
This short paper examines a method for the collaborative discovery of metadata, one of several uses of CoFIND (Collaborative Filter in n dimensions) first presented at WebNet 99 (Dron et al 1999). We will discuss an implementation of the system customised to collaboratively evaluate a limited range of websites. In its full version ,CoFIND is developing into a self-organising learning environment, but in the iteration discussed here it has been cut down to only provide a means of rating a static list of web pages. Ratings may be in the form of comments and discussions using a built-in discussion mechanism. However, the main means of rating is through the use of qualities. Qualities are the things that users find valuable about web pages, and are entered by the users of the system themselves. These provide the n dimensions referred to in CoFIND's name, providing more dimensions of evaluation than the two-dimensional scales found in typical collaborative filters. To prevent immoderate growth of qualities they are subject to evolutionary forces. Successful qualities are those that are used frequently to rate and/or to seek web pages. Success is rewarded by a higher position on the list of selectable qualities, whilst failure means slipping down that list and eventually falling off the end if out-competed by other qualities. A feedback loop is formed between the list of resources and the qualities used to assess them. If a quality provides a list or resources of use to the user then it is more likely to be selected again. We performed this experiment to explore the potential for using CoFIND to identify the issues that are important to users when visiting web sites. The results should not be considered as an objective assessment of the qualities users seek to find or avoid but more as a piece of action research to find out how it should be done.
Dron, J., Siviter, P., Boyne, C. & Mitchell, R. (2000). Website Evaluation System: Collaboratively Discovering what makes a Website Good. In Proceedings of WebNet World Conference on the WWW and Internet 2000 (pp. 685-686). San Antonio, Texas: Association for the Advancement of Computing in Education (AACE).
© 2000 Association for the Advancement of Computing in Education (AACE)