Analisando sistemas de classificação na web sob a perspectiva da interação social em comunidades de prática

  • Authors:

  • Fernando M. Figueira Filho;Paulo Lício de Geus;João Porto de Albuquerque

  • Affiliations:

  • Instituto de Computação - UNICAMP;Instituto de Computação - UNICAMP;Escola de Artes, Ciências e Humanidades - USP

  • Venue:
  • Proceedings of the VIII Brazilian Symposium on Human Factors in Computing Systems
  • Year:
  • 2008

Quantified Score

Hi-index 0.00

Visualization

Abstract

Social interaction mediated by information technology has been studied from several perspectives over the past years. This paper draws attention to the relevance of classification systems and categories as important mediation artifacts in human-computer interaction. In this sense, classification schemes provide meaning to information and mutual understanding between the parts in an interaction. However, studies in distributed cognition show us that classification schemes have a flexible nature and depend on the context in everyday human activity. Under this perspective, this paper analyzes three approaches to classification in the web: the semantic web, collaborative annotation and hybrid approaches. The analyses are based on the concept of communities of practice and on the perspective of context as a problem of social interaction in these communities. Based on these theoretical works, this paper derives two design principles for web classification systems: firstly, collaboration allows for participating users to be active agents in content classification. Secondly, multidimensionality permits the co-existence and management of multiple classification perspectives, which is of fundamental importance in complex, heterogeneous and large-scale infrastructures such as the web.