Agile Response and Collaborative Agile Workflows

Introduction

Crisis response is a domain of work that tests the limits of collaborative information technologies. It reveals significant beneficial potential, such as enhanced capability for distributed collaboration and coordination, but it also highlights many challenges. The scale, complexity, mobility, dynamic contexts, and time-critical nature of crisis response and the need for reliable technologies demand synchronised innovation in best practice, ubiquitous and mobile computing, autonomous, agent-based systems, social organizational, legal and ethical aspects. To realise the potential of collaborative information technologies for crisis responders, it is critical to recognize the interdependence of human and non-human agency.

Connectivity breakdowns frequently highlight the extent of such interdependence. The widespread loss of communications among first-responders, emergency managers and the general public in the United States of America after hurricane Sandy, for example, affected people’s ability to coordinate response efforts. In contrast, when power and networks are live, technologies extend human agency. Affected members of the public and formal and non-governmental response agencies utilize a variety of digital technologies to collect, share and make sense of information, coordinate intra- and inter-agency collaboration and communication with other relevant actors (local authorities, insurances or community support groups) and wider publics.

Augmented with the right kinds of technologies, human ability to communicate, collaborate and coordinate emergency response can be improved. And augmenting these capabilities can enable new forms of ‘agile response’ (Harrald, 2006; Harrald, 2009). Despite sitting uncomfortably with the 1980’s ‘revolution in military affairs’ through new ‘agile’ surveillance, communication and targeting technologies (Cockburn, 2012) the concept of agility can be useful to re-specify what might constitute ‘agile response’ for disaster response. Resonant with agile software design methods, key features include adaptive planning, rapid and flexible coordination, sensitive to context. Agility as we use it here describes enhanced abilities to combine knowledge, skills, resources from diverse human and non-human actors (colleagues, the public, sensors, software agents) on the fly. It suggests that distributed, but closely coupled, diagnostic and remedial work should be supported, that scope for ‘emergent interoperability’ is needed amongst changing ‘adhocracies’ of actors (Mendonça et al., 2007), and support for improvisation within clearly structured response management, combining agility and discipline (Harrald, 2006).

This ideal of agile response is hard to realise. Both military and civic attempts to leverage the potential of IT have led to spectacular, costly failures. For example, in 1992 a new London Ambulance Dispatch System was implemented, intended to enhance the efficiency of the service. However, in practice it incapacitated operators’ local knowledge, leading to such significant delays that it was removed after only one day of operation, at considerable cost (Shapiro, 2005). More recently, the UK FireControl project, which sought to increase efficiency through IT supported centralised control for the fire brigade, was abandoned at a cost of nearly half a billion pounds, because it did not support the actual practices of coordinating emergency response (National Audit Office, 2011). These examples of IT failures make concrete what is at stake in developing support for agile emergency response. Designers that seek to enhance human-technology ‘collaboration’ should practice carefully radical and radically careful innovation (Latour, 2008).