The Intellectual Challenge of CSCW: The Gap Between Social Requirements and Technical Feasibility

Citation: Mark Ackerman (2000) The Intellectual Challenge of CSCW: The Gap Between Social Requirements and Technical Feasibility. Human Computer Interaction (RSS)
DOI (original publisher): 10.1207/S15327051HCI1523_5
Semantic Scholar (metadata): 10.1207/S15327051HCI1523_5
Sci-Hub (fulltext): 10.1207/S15327051HCI1523_5
Internet Archive Scholar (search for fulltext): The Intellectual Challenge of CSCW: The Gap Between Social Requirements and Technical Feasibility
Download: http://dx.doi.org/10.1207/S15327051HCI1523 5t
Tagged: Computer Science (RSS) essay overview social computing theory (RSS)

Summary

Ackerman defines the "social-technical gap ... the divide between what we know we must support socially and what we can support technically."

This gap is one of the central problems for HCI.

He builds his argument by reviewing a large and organized set of findings from research in the area of computer supported cooperative work (CSCW). First is that people have very nuanced behavior when it comes to social activity in work. He references Goffman's observation that decisions about information sharing are nuanced, complex and require care, yet access controls are typically coarse. He also references Interpretivists like Kling, Star, and Suchman emphasize the challenges of meaning making in heterogeneous, potentially conflicting groups, without shared histories, the use of boundary objects, and the fluidity of roles.

Other key findings are:

• The importance awareness of coworkers or managers in a shared space and how sharing shapes incentives.
• Users negotiated and renegotiated norms, and use back channels
• CSCW systems often require a critical mass of users
• People use systems in ways designers don't expect

P3P

Ackerman presents an account of the P3P proposed web privacy protocol as illustration of a design proposal with a significant social-technical gap. The ambition of the project was to have user's browsers be able to automatically negotiate the exchange of personal information with sites according to the user's wishes. Think of Facebook's expansive privacy settings, but for every site on the Internet. This design problem is intractable because the parameter space is nearly infinite and properties that might be desirable socially (e.g. allow sites to only keep my data for a limited amount of time) are enforceable technically or politically. A real world design problem of this kind cannot afford an ideal solution, it must be further constrained. "This is the social-technical gap."

He breaks down the social technical gap inherent to P3P into 3 issues:

1. Systems do not capture the nuanced distinctions that people make.
2. People implicitly switch roles but systems have explicit states.
3. Systems are have discrete states, but people operate ambiguously.

Section 4

He opens this section by engages with the book The Sciences of the Artificial (Simon 1969). CSCW is "science of the artificial" in the sense that it studies not the natural world, but an an artifice (i.e. human constructed technology and society). He criticizes Simon for ignoring the social technical gap, but sees CSCW as both an "engineering discipline attempting to construct suitable systems for groups organizations and other collectivities, and at the same time, CSCW is a social science attempting to understand the basis for that construction in the social world."

Next he describes three categories of interventions that might improve awareness and to address current problems. These are ideological initiatives which require design methods that "prioritize the needs of people using the systems," political solutions like mandating trade union participation, and incorporating CSCW into the computer science curriculum.

He subsequently turns to the question of how to build CSCW into a science / intellectually coherent research area. Drawing the analogy of "first order approximations," he describes a set of work around that don't bridge the gap, but "edge around it in ways that are not extremely odious and ... do so with known effects." FOAs include email, chat systems. These are not a solution, but they work because they allow people to work out norms, roles and how they will exchange information for themselves. A cool illustration of a novel FOA is Hudson and Smith (1996) who create a system that transmits muddles A/V feeds that produce awareness without explicit interaction or privacy disruption. Another FOA is like collaborative filtering, "architectures that neither require action nor delegate it."

As a FOA solution for P3P he suggests an architecture of "privacy critics" that warn users about sites that are not trustworthy or may not protect privacy. Think of privacy badger or an ad blocker.

In conclusion he emphasizes that the social-technical gap is a major intellectual contribution that CSCW can make. We aren't here to create cool toys, we are here to understand and produce usable systems for "groups, organizations, communities, and other forms of collective life."

Technical research in CSCW

In this section Ackerman begins to argue that CSCW is transitioning from a field where designers may have been limited by their poor understandings of the social world to a field with a good understanding of the requirements of the social world but lacking in design and technical solutions to social problems. These solutions are not likely to be easy to find and are not a problem of technical competency. Instead, "the gap is likely to endure." He spends a paragraph speculating that these problems are not specific to a particular computer architecture. Even advanced AI / neural networks are unlikely to solve the problem.

Another way you might think the gap could go away is if society changes to close the gap. Two ways this might happen are through a co-evolutionary process in which society and technology adapt to one another gradually and a "Neo-Taylorist" process in which humans adapt to computational rationality. Some coevolution is likely to happen, and CSCW are on the front lines of this process in a sense. Ackerman argues that "technological trajectories ... may also be responsive to intellectual direction" in order to guide the process of technological change.

Theoretical and Practical Relevance

Ackerman's essay is a highly influential article that helps position the CSCW field in the gap between social science and engineering. This is important because it helps CSCW researchers define their collective goals problems. Seeing CSCW as a science of the artificial orients us to the importance of design and technology development. The notion of First Order Approximations is a useful guide toward how to connect social science and design.