Exc.1: Kling and Scacchi (1982) The Web of Computing: Computer Technology as Social Organization

Kling, Rob and Walt Scacchi (1982). The Web of Computing: Computer Technology as Social Organization. Advances in Computers. 21:2-60. Available online

• Detailed comparison of discrete-entity model and web model of design on pages 6-11

• Case study of two data system introductions on pages 12-16

• "In contrast to the discrete-entity models, which gain simplicity by ignoring the social context of computing developments, web models make explicit the salient connections between a focal technology and its social and political contexts."(p.3)

• "When an analyst uses a discrete-entity model to understand the computing capabilities of an organization he usually begins by asking, What kind of equipment and facilities do they have?" (p.9)

• "In contrast, analysts using a web model begin by asking: What kinds of things do people do here?" (p.9)

• "Analysts who use a discrete-entity model explain the occurrence of difficulties in realizing the direct benefits of new computing technologies and the occurrence of unexpected negative side effects or outright failures by attributing blame to a set of discretely identifiable causes: the operating managers were inadequate; the computer used was too small or cumbersome; promised data were not available; users were disinterested; there was insufficient support from top managers; etc." (p.3)

• "The formal goals of an organization are a fair-to-poor guide for learning what it does. The formal procedures of an organization provide a fair-to-poor guide to the way things are done. The formal features of a computer system are a fair-to-poor guide to what it can do and how it is used." (p. 8).

• "Discrete-entity models indicate what may happen under the best of circumstances. The web models are best suited to ask what is likely."

• "computing innovations have to fit the career contingencies and work style of local participants as well as system configurations and organizational procedures."

• From a 1970s case study:

  • "Automated data validation emerged as a necessary condition for ensuring the proper operation of production systems. Data had to be checked by each program to determine if they were appropriate and whether they should be treated as regular or exceptional. The major portion of most of the modem production systems at INSURE had been designed and implemented specifically to handle exceptional or errant cases. As one software systems manager noted, 90% of our code is devoted to handling 5% of the cases."
  • "Even within a firm that has major organizational and financial commitments to data base management (as well as computing), there still are not enough skilled people, much less, sufficient time, inclination, and budget to convert old production systems, maintain currently operational systems, and develop new systems as desired."

• Five core propositions (with our brief description):

  • The Provision of Computing Services Is Specialized: Requirements and challenges for data management depend on the specific situation at each organization (types of tasks, types of data, existing systems and processes, staff)
  • The History of Commitments Constrains Present Choices: Updates or changes to an organization's established information systems usually need to be implemented while existing staff have to keep the original tools going (i.e., try to minimize disruptions in the delivery of data products) , so they tend to be incremental add-ons that match existing skill sets.
  • Narrow Incentives and Opportunities Motivate Choices: Key individuals contributing to an organization's data management sequence will be motivated to tackle transition challenges if they can see a very direct benefit to themselves (e.g., streamline their specific tasks, gain valuable skills that would be an asset when moving to the next job, likely to impress key managers)
  • Macrostructural Patterns Influence the Character of Local Computing: Development and uptake of data system improvements will progress when there is a link to career options and professional advancement of local contributors. However, if they see few such oppportunities, "then innovation, as well as the ease with which local computing systems will be used and kept operational, will decline, and the systems will become less well understood, more demanding to use, and more costly to sustain."
  • Computing Systems Evolve through Fitting and Packaging: "Computing work is an ensemble of computing tasks. Existing task chains can become tangled and knotted when they interlink. This disordering occurs when participants alter, plug up, or bypass established workflows. Administrative discretion in handling an "exceptional situation" might entail skipping regular events or states in a task chain. A task chain can be unclogged through congenial, covert, or coercive negotiation between the affected participants. Another important way to unclog recurrently congested work flows is through computing innovation, be it with a new system development or an enhancement to some existing system.*" *.