Internal Case Library¶
Artifact — instantiates Collective Learning System
Preserves lessons as whole, context-rich narrative cases — not fielded records — so future readers can reason by analogy and judge for themselves what transfers.
An Internal Case Library is a curated collection of rich, whole narrative cases — each one a preserved account of a real situation with its context, constraints, decisions, and outcome intact. Its defining move is a deliberate choice of form: where other memory mechanisms distill a lesson into a rule, a tag, or a checklist, the case library refuses to distill, because the judgment a future reader will need — when the lesson applies, what made it work, where it would break — lives in the surrounding context that distillation strips away. It exists for lessons that would be falsified by reduction to a checklist. A reader does not look up "the rule" here; they find the most analogous case and reason from it to their own situation, doing the transferability judgment themselves using the context the library took care to keep. It is the archetype's vessel for judgment-heavy, analogy-transferable learning.
Example¶
A litigation firm keeps a library of past matters, each written up as a full case: the client's situation, the strategic choices the partners considered, why they took the line they took, how opposing counsel responded, and how it resolved. When a new dispute comes in that smells familiar — a supplier contract with an unusual indemnity clause — a junior associate does not consult a "rules for indemnity disputes" checklist, because no such checklist could carry the judgment that matters. Instead they pull the three most analogous past matters and read them whole: one where an aggressive early motion paid off, one where it backfired because the judge disliked it, one where a settlement posture worked. From the differences between those cases and the case at hand — this judge, this counterparty, this clause — the associate reasons out which precedent's logic transfers and which does not. The library's value is precisely that it kept the context a checklist would have thrown away.
How it works¶
- Preserve the whole, not the takeaway. Each case keeps situation, decision, reasoning, context, and outcome together, because the reusable judgment is in their relationship, not in any extractable rule.
- Curate for richness, not coverage. Cases are selected and written up for how much future reasoning they can support, not to log every event; a small set of deep cases beats a large set of thin ones.
- Index for analogy. Cases are made findable by the situational features a reader reasons from — domain, constraints, the shape of the problem — so the most similar case surfaces.
- Leave the transfer judgment to the reader. The library does not rule on where a case applies; it hands the reader the context to make that call, which is the point.
Tuning parameters¶
- Distillation restraint — how much a case is trimmed toward a summary. More trimming eases reading but erodes exactly the context that makes analogical reasoning possible; the library's whole reason to exist is to trim less than a database would.
- Curation bar — how selective the collection is. A high bar keeps every case worth reading but risks omitting the odd case a future reader needs; a low bar risks a sprawl no one mines.
- Indexing scheme — by outcome, by domain, or by problem-shape. Problem-shape indexing best supports analogy but is the hardest to maintain consistently.
- Narrative voice — neutral reconstruction versus authored point-of-view. A point-of-view case teaches faster but embeds one interpretation the reader must then discount.
When it helps, and when it misleads¶
Its strength is that it carries the judgment that no rule can, which is why fields built on situated expertise — law, medicine, design, diplomacy — reason from cases in the first place; case-based reasoning is the formal name for exactly this move.[1] It is the right vessel whenever reducing a lesson to a checklist would delete the discernment a future reader needs.
Its characteristic failure mode is false analogy — a reader latches onto the surface similarity of a case and imports its conclusion while missing the context that made that conclusion valid, transferring the answer without the reasoning. The classic misuse is treating the library as a precedent to be applied rather than reasoned from, so a vivid old case ossifies into an unexamined default. The guarding discipline is to read for the differences between the stored case and the situation at hand as carefully as the similarities, and to keep each case's boundary conditions explicit so the reader is reminded what made it work.
How it implements the components¶
codification_or_contextualization_process— it sits firmly at the contextualization pole of this component: its deliberate choice is to keep each lesson as a whole, context-rich case rather than convert it into a rule or checklist.transferability_check— it does not run a formal test; it equips the reader to make the transferability judgment by preserving each case's context and boundary conditions, so a reader can reason out what transfers to their own situation.
It keeps the form and the context; it is not the searchable, fielded store those cases might live in (institutional_memory_store — that's Lessons-Learned Database, its nearest twin: the database reduces a lesson to tagged fields for retrieval, the case library keeps it whole for reasoning). It also does not route cases to adopters (diffusion_channel — Best-Practice Diffusion Protocol).
Related¶
- Instantiates: Collective Learning System — the contextualizing vessel for judgment-heavy lessons that resist reduction to rules.
- Consumes: Cross-Team Retrospective and After-Action Review supply the raw episodes that become cases.
- Sibling mechanisms: Lessons-Learned Database · Cross-Team Retrospective · After-Action Review · Community of Practice · Best-Practice Diffusion Protocol · Learning Review Cadence · Organizational Memory System
Editorial Notes¶
Form Classification¶
Form family: Record, Log & Register
Rationale: The library durably accumulates context-rich narrative cases so future readers can recover past situations and reason from their provenance by analogy.
Nearest alternative: Representation, Specification & Plan — Each case is a narrative artifact, but the library's primary value depends on the accumulating historical collection.
Review outcome: Adjudicated after independent review; high confidence.
Origin Attribution¶
Primary origin: Organizational & Management Science
Origin pattern: Convergent development
Present-day reach: Multi-domain
Rationale: Organizational knowledge management preserves experience as reusable cases, matching the internal library's purpose. Cognitive-science case-based reasoning formalizes retrieval and adaptation; information science and case teaching shape preservation and interpretation.
Related originating lineages:
- Cognitive Science — Case-based reasoning materially explains retrieval and adaptation by analogy rather than rule application.
- Education & Pedagogy — Case-method teaching preserves contextual narratives for judgment and transfer.
- Library & Information Science — Cataloging and retrieval materially determine whether relevant whole cases can be found at decision time.
Review resolution: Organizational knowledge management preserves experience as reusable cases, matching the internal library's purpose. Cognitive-science case-based reasoning formalizes retrieval and adaptation; information science and case teaching shape preservation and interpretation. The retained alternate domains identify documented formative or independently established origins, not downstream applicability alone. domain_reach=multi_domain because the operating pattern has established use in several fields. The entry generalizes an established mechanism without inventing a new cross-domain composite.
Review outcome: Researched adjudication after independent review; high confidence.
Sources consulted:
- https://www.dfki.de/en/web/research/projects-and-publications/publication/9610 — DFKI publication on case-based reasoning as a methodological foundation for organizational knowledge management.
Notes¶
The case library and the Lessons-Learned Database are the two poles of one spectrum: the database codifies (reduce to fields, optimize for retrieval), the library contextualizes (keep whole, optimize for reasoning). A mature learning system usually wants both — the database to answer "has this happened before?" fast, the library to answer "how should I think about it?" well — and the mistake is asking either to do the other's job.
References¶
[1] Kolodner, J. L. Case-Based Reasoning. Morgan Kaufmann (1993). Defines case-based reasoning as retrieving similar stored situations and using them to solve new problems. registry ↩