Enterprise Data Governance Reference Master Data Management, Semantic Modeling (Pierre Bonnet(auth.))（Z-Library）

【One-Line Pitch】 A management-level guide to treating reference and master data as a strategic asset: it explains how governance, semantic modeling, and MDM architecture can restore data quality and regulatory traceability. Best for executives, IT managers, and architects who must align business rules with an increasingly regulated information system. 【Book Arc】 - **Opening (~0%–10%)**: Frames the problem — digital-economy regulation (Sarbanes Oxley, Basel II, Solvency II, IAS-IFRS) demands transparency and traceability, while legacy IT systems are complex, badly adapted, and fragile. Introduces MDM and semantic modeling as the forward-looking remedy. - **Early (~10%–35%)**: Defines the company-and-data foundation: why scattered integration layers lack real models, what a "rich data model" is, and how reference/master data differ from transactional data. Distinguishes MDM from CDI, PIM/PLM, and LDAP repositories. - **Middle (~35%–55%)**: Moves to strategy and economics — regulatory pressure on boards, the three-stage transformation (master data → business rules → business processes), and ROI/intangible-asset accounting for data and software. - **Late (~55%–80%)**: Covers MDM maturity levels (virtual → static → semantic, model-driven), governance functions (versioning, validation rules, approval, access rights, hierarchies), and organizational aspects for enterprise-wide adoption. - **Ending (~80%–100%)**: Shifts to the IT department view — semantic modeling framework, logical data modeling, organization modeling, and technical integration of the MDM system. 【Key Takeaways】 - **Data quality is now a governance obligation, not an IT preference** (Opening): regulations force transparency and traceability onto the board, making data a corporate-governance matter rather than a back-office concern. - **A rich data model beats a data dictionary** (Early): meaning comes from relationships between data elements; those relationships reveal validation rules and integrity constraints that a flat dictionary cannot express. - **MDM must precede rules and process repositories** (Middle): the book argues for regaining control of reference/master data first, then business rules (BRMS), then processes (BPM) — reversing this order is described as a considerable IT failure. - **Maturity progresses from virtual to semantic MDM** (Late): virtual and static MDM carry risks; only model-driven, semantic MDM aligns the solution directly with the semantic model to reach "total quality." - **Governance functions are business-facing tools** (Late): version management, use-context data entry, validation rules, approval workflows, access rights, and hierarchy management are designed for business users, not just IT consoles. - **Software packages create silos that MDM can neutralize** (Middle): PIM/PLM handle products only and hard-code validation rules; an enterprise MDM on a neutral platform can synchronize across PLM, ERP, and supply chain without replacing them. - **Data and rules are intangible assets to be capitalized** (Middle): ROI should account for intangible IT assets, with performance indicators jointly specified by business and IT. - **Context changes validation** (Early): the same data may be validated differently depending on whether it is used by headquarters or affiliates, so use-context management is essential. 【Reading Tips】 - Read Part 1 (Chapters 1–4) carefully if you need the business case and ROI argument; skim if you already accept the premise. - Deep-read the maturity-level and governance-function chapters — they are the practical core for planning an MDM program. - Treat the semantic modeling and logical data modeling chapters as the technical hard spot; they reward slow reading with architects or modelers. - Use the three-stage transformation logic (data → rules → processes) as your implementation checklist before starting any MDM project. - Note the case study on semiconductor product configuration as a concrete example of MDM coexisting with ERP/PLM rather than replacing them. 【Coverage Limits】 The excerpts cover the book's structure, opening argument, strategic framing, and several middle chapters, but do not provide full detail on later technical chapters (semantic modeling procedures, logical data modeling, organization modeling, technical integration). Specific figures, ROI calculations, and complete chapter contents are not fully represented.