Academic journal article JITTA : Journal of Information Technology Theory and Application

People, Places and Things: Leveraging Insights from Distributed Cognition Theory to Enhance the User-Centered Design of Meteorological Information Systems

Academic journal article JITTA : Journal of Information Technology Theory and Application

People, Places and Things: Leveraging Insights from Distributed Cognition Theory to Enhance the User-Centered Design of Meteorological Information Systems

Article excerpt

ABSTRACT

There are many challenges in developing information systems to support information intensive collaborative work such as weather forecasting. The Australian Bureau of Meteorology has instituted the forecast streamlining and enhancement project (FSEP) for its next generation of meteorological information systems (MetIS) and significantly, has recognized the critical importance of grounding new MetIS in a thorough understanding of the weather forecasting process. This poses a major challenge for researchers due to the forecasters' very busy 24/7 deadline-driven working environment and from the fact that critical information requirements arise from the situated, embodied and distributed nature of cognitive interactions between forecasters.

This paper explores the utility of distributed cognition (Dcog) theory as one approach to overcome these research challenges and generate insights for the design of the Bureau's next generation of MetIS. At the theoretical level, Dcog theory allows for the capture and validation of design insights through observing cognitive behavior viewed as a system of individuals interacting within their material environment. At the methodological level, the data collection techniques deployed captures the complex socio-technical nature of forecasters' information sharing without interrupting their work. This paper highlights the utility of Dcog theory in sensitizing designers to the cognitive implications of changes to information systems and/or work processes and how the use of Dcog can empower user centered design methodologies.

INTRODUCTION

The Australian Bureau of Meteorology is a federal government funded national agency providing weather information in a range of formats to a wide range of clients. The Bureau's activities take place 24 hours a day, 365 days a year with observational and forecast weather data collected and analyzed at regional offices based in the capital cities of each state and the Northern Territory.

In recent years, the Bureau has faced increasing pressures due to the imposition of resource constraints concurrently with increasing customer demand for weather services. In response, the Bureau has commenced the forecast streamlining and enhancement project (FSEP) in an effort to guide the development, design and implementation of the next generation of information systems for supporting meteorological staff (forecasters) in the forecast process. In implementing the FSEP project the Bureau faced a number of challenges including how to acquire a detailed understanding of the weather forecasting process. Two aspects of this were: firstly, forecasters were frustrated by the support provided by the existing information systems, skeptical that FSEP would deliver better systems, and suspicious that the new FSEP paradigm might compromise their professionalism by automating aspects of the forecast process. secondly, initial attempts to understand the forecast process have increased burdens on forecasters. The Bureau has adopted a modified form of extreme programming (http://www.extremeprogramming.org/) as its software development methodology for FSEP. It enables incremental and iterative development, but relies heavily on end-user feedback. This has not been completely successful, and exhibited similar problems outlined by Gasson (2003).

As a result, there was the need to generate a research approach that simultaneously retained the context of the forecasting activity, captured the processes and interactions involved but that did not add to the burden on forecasters or take them away from their forecasting activities.

In this context, this paper explores the utility of distributed cognition (Dcog) theory as one approach to overcome these research challenges and generate insights for the design of the Bureau's next generation of meteorological information systems. Dcog theory rejects the laboratory as the appropriate context for understanding and argues for studying cognition as it occurs in its natural setting (Hutchins, 1995a). …

Search by... Author
Show... All Results Primary Sources Peer-reviewed

Oops!

An unknown error has occurred. Please click the button below to reload the page. If the problem persists, please try again in a little while.