Doing Enterprise-Architecture A Maturity-Model Guide to Architecture Development (Tom Graves, Slade Beard) (z-library.sk, 1lib.sk, z-lib.sk)

【One-Line Pitch】 A practical field guide for enterprise architects who need to turn architecture from a diagramming exercise into a repeatable, dependency-driven capability that delivers provable business value—read this if you want a step-by-step maturity model rather than another framework catalogue. 【Book Arc】 - **Opening (~0%–10%)**: Frames the core problem—what enterprise architecture should actually do, in what sequence, and for what business purpose—and introduces the dependency-driven maturity model as the organizing spine of the whole book. - **Early (~10%–30%)**: Builds the foundations of change: how to make sense of any kind of change (planned or unplanned), the "art and science" of working with complexity and uncertainty, and the CSPAR pattern (Context, Scope, Plan, Action, Review) as a context-neutral change pattern. - **Middle (~30%–55%)**: Deepens the Change-Layers model into practical sub-layers—from abstract interaction models through logical, physical, deployment, and action models—and introduces the Change-Cycle and fractal "same-and-different" thinking needed to connect across domains. - **Late (~55%–85%)**: (Excerpts do not cover this range in detail.) Presumably develops the maturity-model stages and the skills/leadership needed to progress through them. - **Ending (~85%–100%)**: (Excerpts do not cover this range.) Likely addresses proving business value and sustaining architecture practice over time. 【Key Takeaways】 - **Architecture must start anywhere and cover everything** (Opening): The book insists on a literal "architecture of the enterprise" spanning people, machines, IT, and more—methods must work at every scope, scale, timescale, and stage of change, not just for IT projects. - **Dependencies drive the maturity roadmap** (Opening): Rather than adopting a maturity model wholesale, you derive your own sequence from dependencies—starting with a consistent capability for change, then building incrementally so each step is a genuine step-change in maturity. - **CSPAR is the universal change pattern** (Early): Context, Scope, Plan, Action, Review works as both a vertical completeness checklist (Change-Layers) and a cyclical process (Change-Cycle), applicable from a multi-year transformation down to checking a boarding pass. - **Different modes suit different phases of work** (Early): The Analyst mode (facts, measurement, logic, time-to-think) is for planning and verification, not real-time action; the book warns against blurring modes—e.g., emotional attachment to "facts" signals a discipline breakdown. - **Fractal "same-and-different" frameworks are non-negotiable** (Middle): Context-neutral sameness connects everything; context-specific difference handles industry needs; without fractal thinking, whole-enterprise architecture collapses into fragmented mini-frameworks. - **Plan work progresses through layered models** (Middle): From abstract Interaction-Model (P.1) to implementation-neutral logical model (P.2), physical/development model (P.3), deployment model (P.4), and action model (P.5)—each layer adding detail while staying aligned with Context and Scope. - **Shortcuts in stakeholder work cost more later** (Early): Rushing interviews, ignoring conflicts, or limiting stakeholders to "important people" misses information that returns to bite you—allow the work the time it needs. - **Real-world urgency validates the method** (Early): The COVID travel-agency example—a new CRM designed, built, and running in ten days—shows the approach is meant for fast response to emergencies, not just planned change. 【Reading Tips】 - **Deep-read the foundations (Chapters 1–4)**: The CSPAR pattern and the modes of change are the conceptual load-bearing walls; everything later builds on them. - **Use CSPAR as a working checklist immediately**: Pick a live project task and run it through Context/Scope/Plan/Action/Review—the book explicitly invites this practice. - **Skim the layered-model detail (P.1–P.5) on first pass**: Return to it when you actually need to produce interaction, logical, or deployment models; it is reference material as much as narrative. - **Watch for the fractal-thinking sections**: They are the hardest conceptual shift and the most commonly skipped—budget extra time there. - **Keep it at your desk**: The authors designed it as a compact reference to revisit over years, not a one-time read. 【Coverage Limits】 The excerpts cover the opening through roughly the middle of the book (foundations, CSPAR, Change-Layers, Change-Cycle, fractals); the later maturity-model stages, skills/leadership content, and business-value proof are not represented, so this guide cannot assess them.