Context sharing in a ‘real world’ ubicomp deployment

https://doi.org/10.1016/j.pmcj.2010.04.004Get rights and content

Abstract

While the application of ubicomp systems to explore context sharing has received a large amount of interest, only a very small number of studies have been carried out which involve “real world” use outside of the lab. This article presents an in-depth analysis of context sharing behaviours that built up around use of the Hermes interactive office door display system received during deployment. The Hermes system provided a groupware application supporting asynchronous messaging facilities, analogous to a digital form of Post-it notes, in order to explore the use of situated display systems to support awareness and coordination in an office environment. From this analysis we distil a set of issues relating to context sharing ranging from privacy concerns to ease of use; each supported through qualitative data from user interviews and questionnaires.

Introduction

Context sharing is an established topic within ubiquitous computing research. However, context sharing is underexplored in real world longitudinal ubicomp deployments where behaviours have built up around new technologies integrated into everyday life. Many examples of ubicomp prototypes exist where the motivation has been to detect and automatically share new or previously unavailable types of context, such as a person’s current location [1] or even the state of their coffee cup [2]. However, Mark Wieser’s ideas for ubiquitous computing [3] were not simply a new direction for the application of contemporary and emerging technologies. Weiser first presented his vision of ubiquitous computing (ubicomp) in the early 1990’s when he outlined how this new computing paradigm raises new challenges for human computer interaction, as computers would now be “interwoven in to the fabric of everyday life”. Weiser’s vision included seamless integration of computing technology into an environment and more natural interactions with technology. Once the technology becomes a familiar part of everyday life it is then possible to empirically study its impact in detail. For ubicomp prototypes and their new functionalities to be interwoven into the fabric of everyday life it is crucial for adoption and appropriation behaviours to take place. One way to enable this is through long-term deployment of prototypes. This article presents a qualitative analysis of the usage behaviours which emerged through the use of a prototypical ubicomp system (Hermes [4]) supporting context sharing and deployed over an extended period.

Hermes provided a groupware application to support context sharing through asynchronous messaging facilities. The aim of the work was to explore the use of situated displays systems to support awareness and coordination in an office environment. The system consisted of interactive digital door displays deployed outside offices in the Computing Department at Lancaster University and available for use by the office occupant and any visitor to the office or passer-by. Occupants of offices equipped with Hermes displays (display owners) were able to manually share personal context in the form of short textual messages (e.g. “Gone for coffee” or “Running 15 min late”) or images. One particularly novel aspect of the work on Hermes was longitudinal deployment outside of the lab in order to allow integration of the system into existing daily routines. Evaluation of usage in Hermes, both quantitative and qualitative, enabled a range of issues relating to context sharing to be identified (detailed quantitative analysis of usage logs collected during the Hermes deployment can be found in [4]). While previous publications discussing the Hermes system have largely focussed on user-centred design or technical issues, the focus of this article is empirical analysis of themes around context sharing using qualitative data obtained from users of the system over a longitudinal period.

The remainder of this article is structured as follows. Section 2 presents related work. Section 3 introduces and describes the Hermes system, including a description of the design and deployment methodology, technical overview and evaluation techniques used. Section 4 provides background on the deployment domain and key uses of Hermes (for context sharing), focussing on the adoption of Hermes into the working routines of two of the ten display owners in particular. Section 5 distils a set of issues relating to context sharing which emerged from qualitative feedback (primarily questionnaires and post-deployment interviews) from all ten door display owners. Section 6 provides a broader discussion of context sharing in the Hermes system, and the paper is concluded in Section 7.

Section snippets

Related work

A range of display-based ubicomp prototypes exist with varying display size and functionality. However, for the majority of these systems the goal was exploration of technical feasibility and the system remained a ‘lab’ prototype with little or no documented use. This section discusses display-based prototypes of similar design or development approach to Hermes.

The “dynamic” door display project at Georgia Tech [5] is similar to the Hermes system in terms of enabling awareness and providing

Hermes

Hermes was designed to support asynchronous messaging and has been used to investigate how such a system, allowing explicit sharing of information, can be used to help support awareness and coordination between staff, and between staff and students in the Computing Department at Lancaster University. Ten Hermes displays were deployed for approximately 27 months from May 2002 to August 2004. The system was also used to explore the affordances of a digital approach over traditional paper-based

Adoption and use of Hermes

While previous publications have considered categorisation of types of awareness in a sample of messages shared using Hermes [17], here we explore how users’ behaviour changed as they integrated Hermes into their daily routines and utilised it for sharing personal context. The definition of context used in this work is “… any information that can be used to characterize the situation of an entity. An entity is a person, place, or object that is considered relevant to the interaction between a

Emerging issues

This section presents a set of pertinent issues related to context sharing, distilled from qualitative analysis of feedback (questionnaires and post-deployment interview) from all ten Hermes door display owners. Where users A and B are discussed this is mentioned explicitly.

Discussion

In Hermes personal context (e.g. an owner’s current location) was shared by manually setting a message, which contrasts with the popularity of investigating the technical feasibility automatically capturing and sharing personal context (e.g. via an active badge [1] or sensor-rich coffee mug [2]). Hermes gave an owner complete control over context shared, but the trade-off was the need to consciously set a message every time this was necessary (either at the door display or remotely), which

Concluding remarks

Weiser’s view of ubicomp technology is that it “disappears” or “weaves itself in” and this is what was experienced with Hermes. Quantitative analysis of usage logs showed that nearly half of door display owners (four out of ten) shared an average of two or more messages per day clearly indicating that for these owners Hermes became part of their working routines. From analysis of owner interviews and questionnaires is it clear that a user-centred design approach was central to the success of

References (21)

  • R. Harper, Looking at ourselves: an examination of the social organization of two research laboratories, in:...
  • H. Gellersen, M. Beigl, H. Krull, The MediaCup: awareness technology embedded in a everyday object, in: Proceedings of...
  • M. Weiser

    The Computer for the 21st century

    Scientific American

    (1991)
  • D. Fitton, Exploring the design, deployment and use of hermes: a system of digital interactive office door displays....
  • D. Nguyen, J. Tullio, T. Drewes, E.D. Mynatt, Dynamic door displays, GVU Technical Report GIT-GVU-00-30, Georgia Tech,...
  • W. Trumler et al.

    Smart doorplate

    Personal and Ubiquitous Computing

    (2003)
  • K. O’Hara, M. Perry, S. Lewis, Social coordination around a situated display appliance, in: Proceedings of the SIGCHI...
  • K. O’Hara, R. Harper, A. Unger, J. Wilkes, B. Sharpe, M. Jansen, TxtBoard: from text-to-person to text-to-home, in:...
  • J.F. McCarthy, T.J. Costa, E.S. Liongosari, UniCast, outcast & groupcast: three steps toward ubiquitous, peripheral...
  • N. Segawa, Y. Murayama, Y. Nakamoto, H. Gondo, M. Miyazaki, A message board on WWW for on-door communication, in:...
There are more references available in the full text version of this article.

Cited by (0)

1

Tel.: +44 0 1524 5 10312; fax: +44 0 1524 5 10492.

View full text