? Dr Carsten Sørensen, LSE
IS414 Designing Information Services
Social Technology Information and Communication Technology (ICT) is one of the most pervasive and forceful
phenomena emerging the past 50 years. Intricate bundles of hardware and software, take part
of our lives wherever we go and whatever we do, and are steadily conquering new grounds.
The application of ICT to support and automate work activities has come a long way from the
world’s first business computer Leo supporting basic logistics and payment functions (Caminer et al., 1998), to an e-world of globalised virtual supply chains, the World Wide Web and mobile
In this week we will explore the evolution of software in terms of the relationships between the
social and the technical. It will be argued that one of the essential aspects of the mobilisation of
interaction is the shifts in the use of software in terms of what aspects of the world are
represented and modelled by the software — the problem domain — and in terms of the types
of work activities are supported through using the technology — the application domain1
It will be argued that the increased mobilisation of interaction has resulted in extending the
relationships between collaborative aspects of work and the representation of these aspects in
What is ICT Used For?
How can we characterise the historical development of ICT use from the early years of scientific
programming of individual bespoke mainframe computers to the use of standard applications on
mobile phones relying on global infrastructures?
Mathiassen (1998) focuses on software development and argues that the history of ICT can be
characterized in three Eras. Era I from the early sixties to the mid seventies gave us mainframes
with simple terminals running software with the purpose of increasing productivity and efficiency
through automation. Era II, from the mid seventies to the late eighties, saw the rise of Personal
Computers (PCs) in local area networks being used for the improvement of individual and group
effectiveness. Era III, from the late eighties until today, is characterized by the emergence of
global networks, mobile computing and standardised software. Here the emphasis is on
providing strategic systems supporting process integration and large-scale collaboration.
Dahlbom (1996) characterizes in a similar fashion the history of computing in terms of four
stages of software use: Stage 1 saw computing machines, such as the mainframe applied for
transaction purposes. Stage 2 characterise personal computing, which in Stage 3 transformed
into collaborative computing. Stage 4 saw the emergence of ICTs such as the Internet
infrastructure and interactive video. Dahlbom (1996) furthermore argues that the diversity,
volume and volatility of areas seeking and molding software must lead us to re-appraise
established practices of bracketing the discussion of software design from software use.
Both Mathiassen’s and Dahlbom’s characterisation illustrates the industrialisation of ICT from
the tailoring of bespoke systems to configuring ERP systems. Although this simple view of the
development is flawed given the reports on the difficulties of managing and indeed controlling
software development and implementation processes (Ciborra and Associates, 2000), we can
clearly characterize the innovation of software artifacts as an increasing degree of
1 The problem domain concept is here used in a similar way the definition of "field-of-work" and application
domain similarly as "cooperative work arrangement" by Schmidt, K. (1993): Modes and Mechanisms of
Interaction in Cooperative Work. In Computational Mechanisms of Interaction for CSCW, ed. C. Simone
and K. Schmidt. Lancaster, England: University of Lancaster, pp. 21-104. [COMIC Deliverable 3.1]. ,
Schmidt, K. & C. Simone (1996): Coordination mechanisms: An approach to CSCW systems design.
Computer Supported Cooperative Work: An International Journal, vol. 5, no. 2-3, pp. 155-200.
? Dr Carsten Sørensen, LSE
industrialization where standard software products substitute the pre-industrial craft of building
bespoke systems (Quintas, 1994). However, these divisions of computer use and software
development into distinct historical eras do not serve our purpose of characterising the diversity
of software artefact use where all of the technologies mentioned above could be applied.
Contemporary software use shows a heterogeneous picture of multiple types of software
artefacts coexisting, for example large-scale mainframe processing interfacing with web-
enabled front-end systems.
This significant reconfiguration of the relationships between software development and software
use, implies the need for new conceptualisations of software artefacts themselves. Much similar
to the available conceptual frameworks for understanding and describing the analysis and
design of information processing in organizations there now is a need for conceptual schemas
for a variety of human conduct involving the use of computer technology – software situated in a
context of use (Dahlbom, 1996).
Orlikowski & Iacono’s (2001) suggest that a tool view can view ICT as a labour substitution tool
in terms of productivity, information processing, or social relations. Dahlbom (1996) argues that
our understanding of ICT often emphasizes one type of technology or technology use, and he
provides four examples: 1) Technology identified with tools, small machines or instruments
facilitating work or entertaining; 2) Technology as complex large scale industrial systems and
infrastructures; 3) Technology as media connecting people; and 4) Technology as the human-
made interface in the foreground.
Table 3.1 synthesises the different perspectives on ICT as drawn from (Orlikowski and Iacono,
2001), (Dahlbom, 1996), and (Mathiassen, 1998) in terms of the four perspectives:
? Transaction, where ICT support the production, distribution, and general management
of organisational information;
? Interpretation, where ICT provide the individual user with a tool supporting their
interpretation, navigation and production of both structured and ill-structured information;
? Connection, where software artefacts support the establishment and maintenance of
inter-personal connections; and
? Collaboration, suggests software support for interdependent collaborative processes
through mutual awareness support, shared workspaces and in particular through the
codification and embedding of collaborative processes.
Independently of these four dimensions we can characterise the extent to which ICT will
informate or automate in specific situations (Zuboff, 1988). For example, the same order
processing system can mobilise and automate the interaction between workers completing
orders as well as supporting all associated roles, including the management function in
obtaining an up-to-date overview of the state-of-affairs. Extending the use of ICT from
transaction and interpretation to also include connections and collaboration has significantly
pushed design into various social science fields.
We will later analyse the impact of highly mobilised interaction on the horisontal and vertical
organising of work activities in more detail. In the remainder we will explore one of the
fundamental aspects of the mobilisation of interaction, namely the representations of social
aspects in technical arrangements. This is both a concern when the technology specifically is
mobile ICT. However, more generally ICTs will embed social arrangements as well.
Transaction Interpretation Connection Collaboration
Description Transaction Individual use of Collective use of Collective use of
? Dr Carsten Sørensen, LSE
systems PC-based tools networked networked
automating supporting technologies technologies
organisational individual supporting supporting
procedures or productivity connections collaboration
ICT Labour Information Productivity tool Social relations Collaboration Substitution Tool processing tool tool tool View
ICT Development Era I: From Era I: From Era II: From Era I: Era Mainframes Mainframes to PCs to Era III: Mainframes to (Mathiassen, Era II: PC’s Global Networks Era II: PC’s 1998)
Technology View Technology as Technology as Technology as Technology as (Dahlbom, 1996) system and tool system and system and
infrastructure infrastructure & infrastructure &
Technology as Technology as
medium medium ICT Use Stages Stage 1: Stage 2: Stage 4: Media Stage 3: (Dahlbom, 1996) Transaction Personal Infrastructures Collaborative
Table 3.1: The synthesis of perspectives into four mutually interdependent discussions of software
The Technological Invasion of the Social
Latours (1991) book chapter title "Technology is society made durable" is of course an obvious starting point in this analysis of the relationships between the social and the technical. Indeed,
from the point of view of an Actor-Network Theory (ANT) perspective, it could be argued that the
separation of the technical and the social is a mistake, as these mutually constitute each other.
Indeed, in his book "Reassembling the Social" Latour (2005) offers an extensive destruction of
this separation. As he argues on page 7:
"Organizations do not have to placed in a wider social frame since they themselves
give a very practical meaning to what is meant to be nested in into a wider set of
affairs. After all, which air traveller would know the gate to go to without looking
anxiously and repeatedly at the number printed on her boarding pass and circled
red by the airline attendant?"
Here, the boarding pass, its inscriptions of essential information for both the passenger and the
airways personnel denotes the coming together of human and non-human actors, in ANT terms.
The main point of this is not the ANT argument as such but rather that it is essential to
understand the intricate relations between human and technological actions. Here in particular
how non-human actors in the form of ICT increasingly infiltrate the human actors through
mobilised interaction both representing aspects and modelling behaviour of human actors and
their activities. Latours simple hotel-key example where a series of refinements from a simple
key (which is often not returned when the hotel guest checks out) to a very heavy key plus signs
reminding the hotel guests to return their keys illustrates how human and non-human actors
jointly inscribe the intended behaviour (Latour, 1991). This chapter is exactly concerned with the
representation of human actors in digital non-human ones. The hotel key and the airplane
boarding pass are neat examples in that we can explore the straightforward processes of
human actors constructs the actions through combinations of micro- and local-mobility (Luff and
? Dr Carsten Sørensen, LSE
Heath, 1998) carries the artefacts. However, when this interaction between human actors as well as between human and non-human actors are both mobilised and conflated in the sense that the non-human actor may represent and even stipulate the actions of the human actor. Applying a Latourian argument, the problem is here one far from the traditional manufacturing setting where the human and non-human actors by and large optimise their activities through explicitly representing and modelling the behaviour of the non-human actors. For example, the advances in supply-chain management can be seen as a result of representing the physical elements of this supply-chain as well as the steps they go through from individual components to sub-assemblies to finished products. The human actors implicitly both accommodate this arrangement as well as directly shape it in response to contingencies. The latter is for example illustrated in the Kanban card example where the stipulated process of a forklift driver responding directly to the release of a Kanban card is overridden due to known contingencies elsewhere in the supply-chain (Schmidt, 1993). This process of automatically orchestrating the movement of parts so the car-engine, tires, chassis, body, seats, dashboard and other parts are seamlessly summoned from around the globe through mobilised interaction and they arrive at one place as by magic in an orderly fashion according to the MRP-II (Material Resource Planning) planners hidden scripts. Largely, this principle works well and has produced remarkable advances in the way products are manufactured. However, the underlying rationale here is that the parts assembled into an automobile do not object to being represented in this mobilisation of interaction between systems and humans.
The essential resource in most manufacturing factories is the machinery representing large capital investments. It is important to manage these resources to ensure maximum utilisation. Supply-chain technology such as Kanban and MRP-II exactly serves this purpose. Enterprise Resource Planning (ERP) technology extends this to other resources in the organisation. The machinery and material of the factories has to be managed as capital investment. When employees represent the most essential resource of the organisation, then the management of the human resources becomes most essential (Grant, 1996).
If we now consider the representation in non-human actors (ICT) of human actors and their behaviour, then this is an altogether different matter. People generally mind what is being said about them in databases and other systems. We are so much more important to ourselves than the car-tyre is to itself that we in many situations will complain bitterly if we feel misrepresented by systems, and in other situations actively will seek to be represented for our own purposes. Therefore, the aspect of mobilising interaction in terms of representing human aspects in technological systems evokes a set of complex problems and issues. The mobilisation of interaction, for example through the use of mICT, offers dramatically new means by which aspects of human actors can be represented in artefacts and instantly carried across great distances. The combination of a parcel delivery vehicle equipped with a wireless computer continuously updating a central server when parcels are signed picked up or signed off can in essence also function as a remote surveillance system of the single driver uniquely associated with this device.
Modelling Problem and Application Domains
It is at this point perhaps helpful to introduce the distinction between the problem domain and the application domain (Mathiassen et al., 2000). The problem domain denotes the part of a context that a computerised system is used to administrate, monitor or control. The application domain is an organisation that administers, monitors or controls a problem domain. We can then explain the manufacturing example above in terms of the automotive parts and the processes involved in assembling them into a car as the problem domain, where the work associated with managing this process being situated in the application domain of manufacturing management and control.
How can we then characterise the representation of human actors and their actions in technological systems? This can be conceptualised analytically in two ways; either as the representation in technology of the application domain or as a representation of a problem domain in which there are human actors. In the case of a hospital system supporting the
? Dr Carsten Sørensen, LSE
management of beds and patients, both patients and beds are by the system treated as non-human actors all part of the assemblage of the problem domain to be managed. Nurses and doctors probably do not have second thoughts about how they represent patients in their systems — even if we as patients often wish to gain insight into what is written about us. As
patients we are simply not part of the application domain, but merely, from the point of view of the system functions, part of the problem domain to be represented, modelled and managed. Nurses and doctors will, however, most likely take a much deeper interest in the representation of the application domain of medical care in computer-based systems. This is, amongst others, because they form the application domain and a system explicitly mobilising the interaction of nurses and doctors through representing these in the technology will have immediate consequences for their work conditions. For example, the implicit modelling of doctor carrying a personal pager alters the ways in which this doctor will engage with his or her surroundings as others at the touch of a button can page the doctor. Of course the working conditions of doctors and nurses can be greatly altered by a bed-management system modelling the problem domain of beds and patients. However, if the system itself does inscribes aspects of the application domain then it does not directly offer any possibilities for mobilising information about doctors and nurses, neither does it directly facilitate the automation of their actions through stipulating these via the system.
We can characterise the vertical aspects of organising through the concepts of problem- and application domain. The problem domain and application domain can be recursively related in that one persons application domain can be another person’s problem domain. This is generally how we functionally can characterise the role of management where the application domain of the office worker exactly is the application domain of the office manager, who in turn, along with the office workers may form the problem domain of the executives engaged in the oversight of strategic and tactical concerns. Weeks 5 and 6 will explore the horisontal and vertical aspects of organising activities in more detail.
The representation and modelling of the problem domain generally supports the management of this domain and can to some extent support the actors collaborating through informating about the current state-of-affairs and through automating processes in the problem domain. Actors can also signal their actions through others observing the changes made to the state of the elements in the problem domain. For example, a head nurse may obtain an overview of what her staff has done through observing the current utilisation of beds and through the observable actions when beds are moved and occupied. However, it is only in very simple cases possible to co-ordinate collaborative activities through interacting and observing the state-of-affairs of the problem domain (Schmidt, 1993). The representation of the application domain and its actions are essential for the support of distributed activities as we will see in much more detail in Week 5. The explicit representation of the application domain exactly enables the mobilisation of interaction concerning the co-ordination of distributed activities. Representing the application domain is essential for the support of actors co-ordinating distributed activities.
? Dr Carsten Sørensen, LSE
Table 3.2: Listing of “objects of articulation work” from (Schmidt and Simone, 1996)
As an example, early Computer-Aided Software Engineering (CASE) tools would support the
production of software through various graphical modelling and data-repository tools. The
underlying repository would simply represent the design decisions made throughout the
collaboration. The repository would often support remote access and sharing of models across
networks and as such at least nominally support collaboration amongst software developers
through distributed access to a representation of the problem domain for software development,
namely the formalised decisions reached concerning the system designed in the form of lines of
code, diagrams, documentation etc. However, these early CASE tool repositories would not
include representations of the application domain, or co-operative arrangement (Schmidt and
Simone, 1996), for the software development, such as the software developers, their roles,
? Dr Carsten Sørensen, LSE
tasks, plans and actual activities and communications. As such, any attempt to exclusively rely on the CASE repository as the resource for mobilising interaction in distributed collaboration would break down as soon as two developers would need to mesh work products, negotiate access to the same part of the model stored in the repository or engage in planning of who should do what work. In order to solve these problems of co-ordinating work developers would then need to enrol additional technologies in effect mobilising the interaction between developers, such as telephones, email, project plans, memos etc.
In a similar fashion, the traditional Bill-of-Materials (BoM) employed in manufacturing to manage the flow of parts formalise and model the problem domain in terms of components making up sub-assemblies and final products but not directly model and support the collaborative work surrounding their use (Carstensen and Sørensen, 1996). To enable remote coordination of work associated with the BoM additional, representations of both the static and dynamic aspects of the application domain can be included. This can then support the collaborative work, for example, concerning joint decisions regarding the BoM (Carstensen and Sørensen, 1996). In Table 3.2 reproduced from (Schmidt and Simone, 1996, p. 190) we see an example of a taxonomy characterising the types of objects of co-ordination work that can be represented in collaborative systems. These are divided up according to whether they are nominal (planned) or actual, and the distinction between co-operative work arrangement and field of work is isomorphic to that of application domain and problem domain.
Issues of Concern
Asymmetrical social relations and technological assumptions of symmetry: Most mobile
services, such as the networking services (Mathiassen and Sørensen, Forthcoming) offered by mobile phones embed technological assumptions about the symmetry of establishing connections. The dramatically dropping costs and increased ease of establishing inter-personal connections can lead to the individual experiencing that this technological assumption of symmetry in social connections clashes with the inherent asymmetrical nature of human interaction (Goffman, 1959; Ljungberg and Sørensen, 2000; Kakihara et al., 2004). In terms of control, the much used network-metaphor, emphasising relation, may not be sufficient as it carries with it the assumption of individual control in the establishment and maintenance of connections. Perhaps Mol & Law's fluid metaphor is more appropriate as its emphasis on variation and transformation (Mol and Law, 1994; Kakihara, 2003).
The illusion of ubiquity: The technical opportunities offered by the development of ubiquitous
information environments has largely so far sold the illusion that ubiquity implies disappearance of our everyday concerns for technology. However, the more mobile communication technology is supposed to disappear, the more it ends up occupying our attention (Sørensen and Gibson, 2004). The tightening of the connection between day-to-day activities through the use of mobile services requires extensive care (Ciborra, 1996), the batteries need charging, the settings must be changed so the phone does not disturb in meetings, alarms are set, constant SMS messages replied to etc.
The pressure of availability: The everyday rhythms of interaction with mobile technologies
create social expectations of instant availability - one aspect of the "anytime, anywhere" society. However, the social consequences of instant availability can be a pressure to render oneself available as well as offering easy instant access.
The pressure of asynchronicity: Related to the implicit or explicit pressure to be available for
interaction, asynchronous technologies such as mobile email and SMS can result in increased demands for timely interaction. These technologies offer convenient flexible means of making asynchronous connections but the ongoing practices of experimenting with their interpretive flexibility can easily lead to the sense of temporality in the situation pushing the social understanding of asynchronicity. If a person constantly checks and answers email, then others will assume their email interaction with this person to be instant.
Drifting media obligations: Related to the pressure of availability are the pragmatic
consequences of choosing a particular medium for interaction in certain contexts. The textual
? Dr Carsten Sørensen, LSE
replacement of voice when an SMS is sent instead of a voice call can be viewed as an attempt to avoid intense interactivity. If the camera on 3G video-calls is deliberately turned off, the receiver of the resulting voice-only call will naturally ask questions concerning the motives for this choice as part of the essential social process of exploring others' opinions of us (Goffman, 1959).
Caminer, D., J. Aris, P. Hermon, & F. Land (1998): L.E.O. - The Incredible Story of the World's First Business Computer. London: McGraw-Hill Education. Carstensen, P. & C. Sørensen (1996): From the Social to the Systematic: Mechanisms Supporting Coordination in Design. Journal of Computer Supported Cooperative Work, vol. 5, no. 4, December, pp. 387-413. Ciborra and Associates, C. U. (2000): From Control to Drift: The Dynamics of Corporate Information Infrastructures. Oxford: Oxford University Press. Ciborra, C., ed. (1996): Groupware and Teamwork. Chichester, United Kingdom: John Wiley & Sons. Dahlbom, B. (1996): The New Informatics. Scandinavian Journal of Information Systems, vol. Vol. 8, no. 2, pp. 29-47. Goffman, E. (1959): The Presentation of Self in Everyday Life. New York, NY: Bantam. Grant, R. (1996): Towards a knowledge based theory of the firm. Strategic Management Journal, vol. 17, no. Winter, pp. 109-122. Kakihara, M. (2003): Emerging Work Practices of ICT-Enabled Mobile Professionals. PhD Dissertation, The London School of Economics and Political Science. http://is.lse.ac.uk/research/theses/ Kakihara, M., C. Sørensen, & M. Wiberg (2004): Negotiating the fluidity of mobile work. In The Interaction Society: Practice, Theories, & Supportive Technologies, ed. M. WibergIdea Group Inc., pp. Chapter 7. Latour, B. (1991): Technology is Society Made Durable. In A Socology of Monsters: Essays on Power, technology and domination, ed. J. LawRoutledge, pp. 103-131. Latour, B. (2005): Reassembling the Social: An Introduction to Actor-Network Theory. Oxford: Oxford University Press. 0-19-925604-7. Ljungberg, F. & C. Sørensen (2000): Overload: From transaction to interaction. In Planet Internet, ed. K. Braa, C. Sørensen, and B. Dahlbom. Lund, Sweden: Studentlitteratur, pp. 113-136. Luff, P. & C. Heath (1998): Mobility in Collaboration. In Proceedings of ACM 1998 Conference on Computer Supported Cooperative Work. ACM Press. Mathiassen, L. (1998): Reflective Systems Development. Mathiassen, L., A. Munk-Madsen, P. A. Nielsen, & J. Stage (2000): Object Oriented Analysis and Design. Aalborg: Marko Publishing. Mathiassen, L. & C. Sørensen (Forthcoming): A Theory of Organizational Information Services. Under Review for International Journal. Mol, A. & J. Law (1994): Regions, Networks and Fluids: Anaemia and Social Topology. Social Studies of Science, vol. 24, pp. 641-71. Orlikowski, W. & C. S. Iacono (2001): Research Commentary: Desperately Seeking the ―IT ‖in IT Research —A Call to Theorizing the IT Artifact. Information Systems Research, vol. 12, no. 2, pp. 121 –134. Quintas, P. (1994): A product-process model of innovation in software development. Journal of Information Technology, vol. 9, no. 1, pp. 3-17. Schmidt, K. (1993): Modes and Mechanisms of Interaction in Cooperative Work. In Computational Mechanisms of Interaction for CSCW, ed. C. Simone and K. Schmidt. Lancaster, England: University of Lancaster, pp. 21-104. [COMIC Deliverable 3.1]. Schmidt, K. & C. Simone (1996): Coordination mechanisms: An approach to CSCW systems design. Computer Supported Cooperative Work: An International Journal, vol. 5, no. 2-3, pp. 155-200. Sørensen, C. & D. Gibson (2004): Ubiquitous Visions and Opaque Realities: Professionals Talking About Mobile Technologies. INFO: The Journal of Policy, Regulation and Strategy for Telecommunication, Information and Media, vol. 6, no. 3, pp. 188-196. Zuboff, S. (1988): In the Age of the Smart Machine. New York: Basic Books.