Why this is in my collection

From the publisher:

Here you'll find one key to the development of a successful information system: Clearly capture and communicate both the abstract and concrete building blocks of data that describe your organization. In 1995, David Hay published Data Model Patterns: Conventions of Thought - the groundbreaking book on how to use standard data models to describe the standard business situations. Enterprise Model Patterns: Describing the World builds on the concepts presented there, adds 15 years of practical experience, and presents a more comprehensive view. You will learn how to apply both the abstract and con

Highlights

  • The world decomposes into a small set of recurring structures — Hay shows that people/organizations, geography, assets, activities, and agreements recur in every enterprise; this finite pattern vocabulary is the intellectual bedrock under any claim that data models can be standardized enough to generate from.
  • Abstraction levels are a deliberate design dial — his layered approach (from highly abstract party/role patterns down to concrete industry variants) teaches when to generalize and when to specialize; the exact judgment call in every hub-and-satellite or ensemble design session.
  • Patterns are conventions of thought, not just diagrams — a shared pattern language changes how modelers and business people converse about the enterprise; the modeling world's ubiquitous language, and the reason a concept library beats a glossary.
  • A pattern encodes decades of failed alternatives — each canonical structure survives because naive versions collapsed under real requirements; using patterns is inheriting tested rules rather than rediscovering them per project, curation as a moat.
  • The party/role pattern quietly solves half of banking — separating who someone is from the roles they play dissolves the customer/employee/supplier duplication every institution suffers from; a pattern that banking data work stands on directly.
  • Standard patterns make deviation detectable — once the expected structure is named, a nonstandard model becomes a flag to investigate rather than invisible drift; patterns functioning as a rules layer over modeling itself.

Highlights on this page are generated with the help of AI.