I design structure before writing code.
My work bridges the gap between business requirements and engineering reality —
through asset-based architecture, requirement traceability, and the
SOM framework I developed over two decades of practice.
A practitioner's answer to the semantic gap in software quality management — inspired by manufacturing's Bill of Materials.
Modern software projects suffer from a persistent "semantic gap" — requirements written in natural language, code artifacts in formal structures, and quality documentation spread across disconnected tools. This paper proposes SOM (Software Object Model/Material), a framework that assigns a persistent identifier (SOM ID) to each functional unit (Program) at the planning stage, connecting requirements, code, test cases, and documentation under a single traceable anchor — analogous to a part number in manufacturing's Bill of Materials.
A serialized essay explaining the SOM framework from first principles. Written in Korean with English subtitles — 5 episodes published, 10 planned.
Across hundreds of software delivery projects — from government systems to commercial products — I kept encountering the same failure pattern: planners, developers, and QA teams could not share a common language for the same software unit.
SOM is my answer. Not a tool, not a process — a framework for assigning identity to software units from the planning stage, so that every artifact across the project lifecycle can be traced back to a single, stable anchor.
The intellectual lineage of SOM draws from manufacturing (BOM), information architecture, and — perhaps unexpectedly — the biographical-annals style (Jijeonche) of Sima Qian's Shiji, in which a subject-centric structure enables cross-referencing across domains.
Each product is independently planned, architected, and shipped — not a demo, a deployable engine.
If you are working on requirements traceability, software quality, or ALM —
I would be glad to hear from you.