Beyond the "Blob": Mapping Information Systems through Institutional Ethnography

TL;DR

Information Systems (IS) are social phenomena, yet we often lack the tools to see how they are actually "organized." This paper introduces Institutional Ethnography (IE), a method that moves beyond abstract theories to map how everyday work is coordinated by texts and "ruling relations." By looking at clinical software and law firms, the authors show that technology failure is often a predictable outcome of how institutional paperwork governs human behavior.

Background: Why "Experience" Isn't Enough

In the world of IS research, we often talk about "resistance to change" or "organizational culture" as if they are invisible forces acting upon people. Institutional Ethnography (IE), founded by Dorothy Smith, rejects this "blob-ontology." Instead of assuming power is an abstract concept, IE argues that power is a set of ruling relations—actual, material activities coordinated through texts (documents, digital records, forms).

The paper’s core insight is that to understand why a system fails, you must start from the standpoint of the person using it and follow the paper trail (or digital trail) that dictates their "doings."

The IE Framework: Four Pillars of Inquiry

The authors break down the IE methodology into four actionable concepts:

1. The Problematic: Not a research question, but a "disjuncture"—a puzzle where the lived experience of a user contradicts the official institutional narrative (e.g., "Why does this 'participatory' process feel like rubber-stamping?").
2. Standpoint: Starting the research from the actual embodied experience of the expert (the nurse, the lawyer) rather than the manager's view.
3. Textual Mediation: Identifying the "texts" (forms, software interfaces) that connect local work to extra-local institutional standards.
4. Ruling Relations: Mapping the complex web of corporations, bureaucracies, and professional discourses that coordinate these texts.

Case Study 1: The "Rubber Stamp" in Clinical Design

In a study of a Vancouver hospital implementing a Clinical Information System (CIS), the hospital intended for clinicians to participate in the design. However, nurses felt like "puppets on a string."

Using an IE lens, the authors identified the Decision Request Form (DRF) as the culprit.

• The Mediation: The DRF was created by a project team outside the clinical meeting.
• The Result: Because the DRF summarized work already "vetted" elsewhere, it effectively curtailed discussion. If a clinician disagreed, the DRF was simply escalated to a steering committee where the decision was overturned anyway.
• Insight: Participation wasn't a failure of communication (as Habermas might suggest); it was a structural outcome of how the DRF text coordinated the "ruling relation" between clinical safety and project efficiency.

Case Study 2: Subverting the Knowledge Management System (KMS)

Why would a group of tech-savvy patent lawyers subvert a system they helped bring into their firm?

• The Standpoint: From the lawyer's desk, drafting a patent requires "precision" that the KMS didn't support. Lawyers needed seven screens or a giant desk, not a database.
• The Ruling Relation: The legal industry is governed by "Time-Based Billing." Entering data into a KMS is an administrative task that doesn't generate billable hours.
• The Outcome: Adoption became "performative." Senior lawyers delegated KMS entry to juniors just to maintain the perception of use, while the real value-generating work happened on paper and in conversations.

Critical Insights & Takeaways

The authors argue that IE allows IS researchers to link the Local (someone's desk) with the Extra-local (global standards of capital or bureaucracy).

Essential Lessons:

• Information Technology is a "Superpower" of Textual Mediation: Software isn't just a tool; it is a replicable text that coordinates activity across time and space.
• Generalizability comes from the "Ruling": You don't need a massive sample size if you can show how a standard form (like a DRF) coordinates behavior across any hospital using that process.
• Practical Utility: For practitioners, this means looking beyond "User Acceptance Testing" and asking: How does this system interact with the existing "texts" and "billing models" that already govern the user's life?

Conclusion

Institutional Ethnography reminds us that the "social" in socio-technical systems is not an abstract mystery. It is a map of documents, forms, and professional obligations. By mapping these "ruling relations," we can design systems that actually support the work people do, rather than the "blobs" we imagine them to be.