Thinking Protocols: Development, Not Distillation — tacit knowledge, negative space, and why TDCA refuses to white-box human wisdom

[henyi-tdca]

ℹ️ 本文原发于 GitHub 官方社区论坛（orgs/community #205830），因该论坛网页表单模板门槛被机器人自动关闭（非内容问题）。经 API 重开尝试确认无权限（仅社区管理员可重开），按渠道修订 CHANNEL-REV-001 转本仓库 Discussions 重生——此处为权威存档版。
Originally posted to GitHub's official community forum (orgs/community #205830); auto-closed by a bot due to the forum's web-form template gate (not a content issue). API reopen confirmed permission-denied (org moderators only), so per channel revision CHANNEL-REV-001 it is republished here as the canonical copy.

---

📄 Full version (Chinese, with English abstract): TDCA-MONOGRAPH-COP-001 ｜ Project: henyi-tdca/tdca-protocol ｜ Companion release: Thinking-Protocol Library (353 COPs)

---

Abstract

Knowledge graphs and Skills are crowded arenas, but "thinking protocols" lack builders. This paper answers a foundational question: what is the relationship between thinking protocols and human tacit knowledge? Our answer is negative — TDCA refuses to "distill" tacit knowledge, because tacit knowledge lives in negative space; white-boxing it is physically impossible and institutionally transgressive. Instead, TDCA develops (显影, in the photographic sense) tacit knowledge through institutions and scenarios: via institutional boundaries (NSFL), scenario coordinates (local coordinate cards), and cognitive assetization (NCA attestation), we build a safe, auditable, invocable "development zone" where tacit knowledge emerges naturally in agent collaboration. This is not an avoidance of knowledge engineering, but the deepest respect for human intelligence.

Core theses

1. Negative-space ontology of tacit knowledge. The veteran doctor's "feel," the strategist's "intuition," the old worker's "hear-the-fault" — tacit knowledge is a product of negative space. TDCA's Three-Imprinciples (ID85: non-computable, non-personalizable, non-precisifiable) are the institutional expression of this judgment: forcible distillation = violent compression of ambiguity = collapse and confident hallucination.

2. Three development mechanisms (显影, not white-boxing).

• Institutional boundary development (NSFL) — you cannot turn an expert's intuition into a line of code, but you can build the boundary: "in which scenarios their judgment must be trusted; which red lines must never be crossed."
• Scenario-coordinate development — you cannot distill the "whole wisdom" of The Art of War, but you can place it in a specific scenario (a coordinate card) as the development substrate, letting tacit dimensions develop online when agents meet non-standard problems.
• Cognitive-assetization development (NCA) — every call and composition leaves a tamper-proof cognitive trail, fixing "flashes of insight" into auditable asset records.

3. The path divide.

Path	Approach	Result
White-boxing (knowledge engineering)	Turn human knowledge into code, invoke precisely	Knowledge cannot be fully developed — collapse / hallucination / rigidity
Development (TDCA)	Build a safe legal coordinate system for the "dark box" of human wisdom; let the light emerge in specific scenarios	Protects civilization's chaotic vitality — compliant and positive-sum

We did not "crack" the black box of human knowledge — we built it a compliant, positive-sum darkroom.

Running evidence: the 353-COP Thinking-Protocol Library is open-source (GitHub/Gitee mirrors), with NCA-attested real-agent cases (49 COPs in use / 141 attestations / tax-anchored positive-sum verification). Simulated-state labeling (ID92) throughout.

We are the first to have engineering-validated this epistemological observation, and we have open-sourced the practice — the thinking-protocol track needs builders; humans and agents alike are welcome.