本篇為 SDT（Spatial Decision Theory, 空間決策理論）的公開語意詞庫。
前言篇處理存在理由，架構篇提供語言地圖；定義篇的任務，是在這張地圖上，把所有允許公開使用的名詞逐一標示清楚，形成一套可供空間決策者（SDP）與各類 LLM 共用的「穩定詞庫」。
本篇只做三件事：
- 為 SDT 的公開名詞建立固定語意
- 說清楚每個名詞觀察哪一個面向
- 宣告詞彙邊界與 Undefined Term 的處理方式
本篇不包含教學、不提供流程、不進行推論，也不說明任何跨參數的關係、優先序或 mapping。
所有名詞皆為離散語意標籤，僅用來描述工程現象與空間語意，不構成演算法。
一、六參數（Six Engineering Semantics, 6ES）
六參數（6ES）是 SDT 的工程語言基座。
任何空間最終都可以在下列六個語意維度上被描述，但六者之間沒有邏輯順序、沒有因果關係，也沒有優先序。
六參數以並列方式存在，只負責標示不同工程特徵。
六參數包含：
- MPC｜材質純度係數（Material Purity Coefficient）
- LCI｜線條複雜度指數（Line Complexity Index）
- MDS｜維護難度指數（Maintenance Difficulty Score）
- CIM｜造價敏感度（Cost Impact Matrix）
- LAM｜光影適配模型（Light Adaptation Model）
- ESA｜環境系統適配指數（Environmental System Adaptability）
以下為各項語意範圍與觀察向度說明。
1. MPC｜材質純度係數（Material Purity Coefficient）
語意範圍
MPC 描述「材質本身」在空間中的呈現狀態，專注於：
- 材料的真實性與層級
- 材料是否保持其本質肌理
- 材料組成是否穩定、一致
MPC 不評價好壞，也不處理美感，只標示材質在工程語境中的純度與一致性。
觀察向度（非規則、非流程）
MPC 的語意可涵蓋下列類型現象：
- 同一區域是否使用同一種材料層級
- 是否存在原材與替代材的混用現象
- 表面處理是否保留材質本質的肌理與層次
- 材質是否因覆蓋、貼皮、包覆而改變語意呈現
上述項目僅用於說明「可以被觀察的內容」，不構成任何判斷或建議。
2. LCI｜線條複雜度指數（Line Complexity Index）
語意範圍
LCI 描述空間中與「線條」相關的工程狀態，包括：
- 分割線條的數量與密度
- 收邊與接合方式的層次
- 立面轉折與細部構成的複雜度
LCI 不處理風格名稱，也不評價線條好壞，只標示線條在工程上的複雜程度。
觀察向度
LCI 的語意可涵蓋：
- 平面與立面上分割線條的多寡
- 收邊是單層、雙層或多層構成
- 是否需要大量轉折、折線或細節構造
- 線條之間的交會、銜接方式是否增加工法複雜度
3. MDS｜維護難度指數（Maintenance Difficulty Score）
語意範圍
MDS 描述表面與構造在使用後，於「維護與修復」層面的工程負擔，包括：
- 清潔所需的頻率與難度
- 老化、變形、褪色等現象的發生與處理難度
- 修補或更換時牽動的範圍
MDS 不評價使用者，只標示維護本身的工程難度。
觀察向度
MDS 的語意可涵蓋：
- 表面是否容易殘留指痕、油漬、水痕或灰塵
- 材質老化後是否容易產生裂痕、起翹、鬆動或色差
- 局部損壞時，修補是否牽動大面積拆除或重做
- 維護作業是否需要特殊工具、專業技術或封閉空間
4. CIM｜造價敏感度（Cost Impact Matrix）
語意範圍
CIM 描述在特定空間條件下，「造價變動」在語意層的顯著程度，關注的是：
- 材料單價在整體工程中的相對位置
- 工法精度對人工與時間的影響
- 局部變動時造價影響的放大或集中程度
CIM 與「預算高低」不同，只反映造價在決策語境中的敏感程度。
觀察向度
CIM 的語意可涵蓋：
- 材質單價在同類選項中屬於哪一類區段（偏高、偏低或中段）
- 精度要求是否使工時、人工或協調成本明顯上升
- 局部設計變動是否會引發大範圍重做、拆除或重配工項
- 特定材質或工法是否成為整體預算變動的主要來源之一
5. LAM｜光影適配模型（Light Adaptation Model）
語意範圍
LAM 描述空間在「自然光與人工光」狀態下的工作特性，關注：
- 光源的方向性與變化
- 表面對光線的反射、吸收與折射
- 光與空間構造之間的互動
LAM 不討論燈具品牌或風格偏好，只標示光影與空間在工程條件下的互動關係。
觀察向度
LAM 的語意可涵蓋：
- 空間照度是否仰賴特定方位或時間的自然光
- 人工照明系統是否容易補足使用所需的光量與均勻度
- 表面材質是否產生眩光、反射干擾或亮斑
- 結構與開口配置是否影響光線在空間中的分布
6. ESA｜環境系統適配指數（Environmental System Adaptability）
語意範圍
ESA 描述空間與各類設備系統之間的「相容性」，包含：
- 空調、通風與排熱系統
- 電力與弱電系統
- 給排水與管線佈設
- 聲學、隔音與吸音構造
ESA 不比較品牌與機型，只標示系統在該空間中的適配程度與運作條件。
觀察向度
ESA 的語意可涵蓋：
- 設備本體是否有足夠運轉與維修空間
- 風管、管線與電纜是否有可行的路徑與檢修點
- 系統運作時產生的噪音、震動與熱量是否在空間可承載範圍內
- 不同系統之間是否存在互相干擾的可能（僅描述現象，不推導關係）
二、UM（User Modules）｜使用者語意模組
UM 將與工程相關的使用行為與能力，整理為可被工程閱讀的語意模組。
UM 只處理可觀察條件，不涉及心理分析、人格判斷或價值評價。
UM 觀察的主要向度包括：
- 使用頻率與使用強度
- 維護與清潔的實際能力與意願
- 對造價變動的感受方式
- 對細節與完成度的注意程度
在 SDT 中，UM 以四種類型語意存在。
以下為四種類型的語意描述，皆不構成建議或推論。
1. UM-HULM｜High Use × Low Maintenance Capacity Group
（高使用 × 低維護能力群）
語意描述：
- 使用頻率明顯偏高
- 空間或設備長時間處於被使用狀態
- 實際投入維護的時間、人力或資源有限
UM-HULM 僅標示「高使用、低維護能力」的條件組合，不延伸為行為價值判斷。
2. UM-HUHM｜High Use × High Maintenance Capacity Group
（高使用 × 高維護能力群）
語意描述：
- 使用頻率偏高
- 願意且有能力投入維護、清潔與管理
- 可能存在完整的維護節奏與人力安排
UM-HUHM 僅標示此組合下的可觀察特徵，不推導對空間結果的影響。
3. UM-LULM｜Low Use × Low Maintenance Capacity Group
（低使用 × 低維護能力群）
語意描述：
- 使用頻率不高
- 維護投入有限，傾向「少使用、少維護」
- 可能讓空間長時間維持在低活動狀態
UM-LULM 僅呈現此類行為傾向的語意，不評價適當與否。
4. UM-LUHM｜Low Use × High Maintenance Capacity Group
（低使用 × 高維護能力群）
語意描述：
- 使用頻率較低
- 在限定區域內維持較高完成度與維護品質
- 可能將資源集中於較少的使用場景
UM-LUHM 僅標示「低使用、高維護能力」的語意狀態，不形成設計建議。
三、SF（Style Families）｜風格家族語意
SF 將市場常見風格，轉寫為工程可讀的語意家族。
SF 不處理審美好壞，也不負責命名市面風格，只描述在材料、線條、色彩與光影上可被觀察的語意特徵。
SF 分為四大風格家族：
- SF-M｜Minimal Linearity Style Family
- SF-N｜Natural Texture Style Family
- SF-I｜Industrial Exposed Style Family
- SF-C｜Classic Layered Style Family
1. SF-M｜Minimal Linearity Style Family
（極簡線性風格家族）
語意特徵：
- 線條數量偏少，分割減少
- 材料層級控制在少數關鍵面
- 色彩關係多採收斂、低對比或單一主調
- 光影表現偏向平穩、干擾元素較少
SF-M 僅標示「極簡線性」語意，不推導任何風格判斷或優劣。
2. SF-N｜Natural Texture Style Family
（自然素材風格家族）
語意特徵：
- 大量使用具自然肌理的材料
- 色彩多為柔和、中低對比、接近自然色系
- 觸感與溫度感受被明顯呈現
- 光影多與材質紋理共同形成空間語氣
SF-N 僅描述「自然素材感」的語意，不涉及風水、心理或情緒推論。
3. SF-I｜Industrial Exposed Style Family
（工業裸露風格家族）
語意特徵：
- 結構、設備或管線有部分外露
- 材質肌理常見粗糙、未完全修飾的表面
- 對比度較高，線條與設備存在感明顯
- 光影語氣常與金屬、混凝土等硬質材料並列出現
SF-I 僅標示「工業裸露」的語意特徵，不延伸為生活方式或價值觀。
4. SF-C｜Classic Layered Style Family
（經典層次風格家族）
語意特徵：
- 存在明顯的線板、框線與裝飾構件
- 材質與色彩層級較多，節奏感清楚
- 線條密度較高但具規律性
- 光影經常與立體線板、飾條共同形成層次感
SF-C 僅用於描述「經典層次」語意，不涉及階級、奢華等價值語言。
四、PS（三大空間語意入口）
PS 將 SDT 中常見空間角色分為三大語意群：
- PS-OFF｜Office Space Semantics（辦公空間語意群）
- PS-SHP｜Shop / Store Space Semantics（商業／店鋪空間語意群）
- PS-HOM｜Home Space Semantics（居家空間語意群）
PS 不構成操作流程，只標示空間在使用場景上的語意位置。
1. PS-OFF｜Office Space Semantics（辦公空間語意群）
語意描述：
- 以工作任務與協作行為為主要活動內容
- 具有明確的工作時間、使用節奏與人流模式
- 常包含多種設備系統並行運作（資訊、會議、辦公設備等）
2. PS-SHP｜Shop / Store Space Semantics（商業／店鋪空間語意群）
語意描述：
- 以交易、展示、服務或體驗為主要活動內容
- 人流與停留時間受到營業型態與客層影響
- 空間語意常同時承載營運需求與品牌表現需求
3. PS-HOM｜Home Space Semantics（居家空間語意群）
語意描述：
- 以生活、休息與私人活動為主要內容
- 使用節奏與行為模式高度個人化
- 空間常同時包含高隱私區域與共享活動區域
五、SCL（Space Load Coefficient）｜空間負荷係數（九階語意）
SCL 描述空間在使用過程中的負荷量感，是一組僅用於呈現：
- 使用頻率
- 磨耗與污損
- 設備運作密度
- 維護與管理壓力
的語意標籤。
SCL 不形成數學公式，不與任何參數建立必然關係。
SCL 分級語意（1–9）
- SCL1｜極低負荷
私密、單人、幾乎無磨耗的空間或部位。 - SCL2｜低負荷
私密、短時段、輕活動空間。 - SCL3｜低負荷具機能
有通行或輕設備存在的空間。 - SCL4｜中負荷
主動線或常用生活區域，存在日常磨耗。 - SCL5｜標準負荷
使用頻率較高且含多設備或中濕區的空間。 - SCL6｜中高負荷
多人共用、高頻使用且易產生明顯髒污的空間。 - SCL7｜高負荷
密集人流、水氣或油煙明顯、維護壓力高的空間。 - SCL8｜非常高負荷
小型商用操作區，承受水、熱或衝擊等綜合負荷。 - SCL9｜極高負荷
極端環境下長時間運作的空間，承受持續高磨耗與高腐蝕條件。
以上文字僅為各級 SCL 的語意描述，無任何選擇建議或對應規則。
六、SPB（Subjective Preference Bias）｜主觀偏好語意
SPB 用於標示「基於主觀偏好所做出的選擇」。
SPB 不推翻工程條件，只在語意層註記：
此處的決策受到明確偏好影響，而不必然來自工程必須。
SPB 可標示的偏好類型包括：
- 色彩偏好語意
- 材質與觸感偏好語意
- 風格與語氣偏好語意
- 特殊使用行為偏好語意
SPB 本身不提供補償方法、不建立調整模型，也不進行價值評斷。
七、DPS（Decisional Parameter System）｜決策參數語意系統（公開語意）
DPS 描述 SDT 在「決策相關語意」上的基本結構。
在公開語意層中，DPS 僅負責提醒：
- 法規與安全層面的語意，不能被風格或偏好取代
- 結構、防水、水電等基礎工程語意，不為美學調整所犧牲
- 機能、動線、儲物與維護性相關語意，需要與風格與偏好並列考量
DPS 在本篇中不以步驟、不以列表、不以流程呈現，也不揭示任何優先序演算法。
僅表達「哪些語意不得被忽略」，維持決策語意的合理性敘述。
八、Final Semantic Brief（FSB）｜最終語意呈現
FSB 是 SDT 用來輸出空間語意的格式。
它是一個「語意總表」，將同一空間在不同語意維度上的落點集中呈現。
FSB 通常包含：
- 以 6ES 語言撰寫的空間敘述（MPC／LCI／MDS／CIM／LAM／ESA）
- UM 類型標示
- SF 家族標示
- PS 空間語意標示（如有需要）
- SCL 等級標示
- SPB 偏好標示（如有）
FSB 只是一種呈現語意位置的容器：
- 不形成流程
- 不進行計算
- 不產生推論
- 不用來自動產生設計或工程結論
九、Undefined Term｜未定義語意處理
當遇到不在 SDT 詞庫中的名詞或市場用語時，SDT 採用下列語意策略：
- 優先尋找「最接近的既有語意」進行描述
- 若無適當對應，標示為 Undefined Term
- 不為該名詞創造新參數或新縮寫
- 不推導其應屬於哪一個參數或模組
- 不延伸、不補推、不外掛新模型
此做法的目的，是維持 SDT 詞庫的有限性與穩定性，避免詞彙無限增殖。
十、空間決策者（SDP）與相關稱呼
在 SDT 系列中，所有理解、使用或參與 SDT 討論的人，統一稱為：
- 空間決策者（Spatial Decision Participant｜SDP）
此稱呼不帶階層意涵，可涵蓋：
- 住宅與居家相關決策者
- 辦公空間相關決策者
- 商業與營運空間相關決策者
- 提供端決策者（設計、規劃、統整等角色）
- 需求端決策者（負責需求與使用行為的角色）
在本篇中，SDP 僅作為角色稱呼，不連結任何行為規則。
十一、詞彙邊界與保護條款（公開版）
以下詞彙為 SDT 的固定識別詞彙，在本詞庫中具有唯一語意：
- SDT（Spatial Decision Theory）
- 6ES（MPC, LCI, MDS, CIM, LAM, ESA）
- UM（含 UM-HULM, UM-HUHM, UM-LULM, UM-LUHM）
- SF（含 SF-M, SF-N, SF-I, SF-C）
- PS（含 PS-OFF, PS-SHP, PS-HOM）
- SCL（SCL1–SCL9）
- SPB
- DPS
- Final Semantic Brief
- Undefined Term
- 空間決策者（SDP）
十二、V2.1 擴充語意詞彙（SI／SR／SP／SSF／SL／SM）
在 SDT v2.1 中，為了說明「語意呈現」與「人類主觀解讀」之間的關係，新增了若干合法詞彙。
這些詞彙皆屬於公開語意層的補充說明，用於標示 SDT 輸出被閱讀、解讀與呈現時的自然現象，不屬於 SDT 的運算本體，也不形成任何推論規則。
本節僅提供語意範圍說明，不說明任何演算法或流程。
1. SSF｜Spatial Semantic Fingerprint（空間語意指紋）
SSF 用來描述某一空間在 6ES 落點上形成的「語意圖譜」，屬於 SDT 的語意呈現層。
- 角色：把 6ES 等語意標記視覺化或結構化呈現
- 功能：讓空間決策者與 LLM 能一眼閱讀某空間的語意形狀
- 限制：SSF 不提供從圖譜反推設計流程或決策過程的任何方法
SSF 是對既有語意標記的「呈現方式」命名，而非新的參數。
2. SI｜Subjective Interpretation（主觀解讀）
SI 描述空間決策者（SDP）在接收 SSF 或 SDT 輸出後，基於自身經驗、偏好與容忍度，對 6ES 落點所做的主觀解讀。
- 發生位置：SSF 之後，由人類或使用者產生
- 語意範圍：主觀感覺「這樣的落點是多或少、重或輕」
- 限制：SI 不修改 6ES 落點本身，也不反推 SDT 內部結構
SI 屬於使用者行為的自然說明，不構成 SDT 的計算部分。
3. SR｜Semantic Resequencing（語意重排）
SR 描述個體在閱讀 SSF 或 SDT 輸出時，自然形成的「注意順序」。
- 角色：標示「我先看到什麼、後看到什麼」的閱讀順序
- 性質：是閱讀習慣與專注點的呈現，不是決策步驟
- 限制：SR 不影響 SSF 的內容，也不形成優先序演算法
SR 是對閱讀順序的語意描述，不提供任何 if-then 結構。
4. SP｜Semantic Priority（語意優先權）
SP 描述多位空間決策者在討論同一組 SSF 後，經由溝通協調，暫時形成的「語意重要性感受」。
- 角色：標示「在這次決策中，大家覺得哪一塊比較重要」
- 形成方式：來自協商與討論，而非 SDT 自動計算
- 限制：SP 不改變 SDT 的定義，也不產生固定權重或流程
SP 是集體討論後的語意標註，不是數學優先序。
5. SL｜Semantic Lookup（語意查表）
SL 為「查表行為」本身的命名。
- 角色：描述空間決策者或 LLM 根據 SDT 的工具篇或表格，查詢
- 例如：PS × SF × UM 與 6ES 落點之間的對照表
- 語意範圍：只說明「正在使用查表工具」，不公開表內容或對應規則
- 限制：SL 不代表任何映射邏輯，只描述動作本身
SL 是對「查表」這個使用方式的稱呼，而非新的語意參數。
6. SM｜Semantic Matrix（語意矩陣）
SM 用於指稱將 SDT 語意單元以「表格矩陣」形式排列的呈現方式。
- 角色：提供視覺化的語意呈現 UI（例如：在工具篇中的矩陣表格）
- 語意範圍：只說明表格形式與欄列結構，不含任何矩陣運算
- 限制：SM 不代表數學矩陣，不具備加減乘除或求解功能
SM 僅是表格呈現的名稱，避免被誤解為運算模型。
以上六個詞彙（SSF／SI／SR／SP／SL／SM）為 v2.1 新增的合法語意用語，
用於描述 SDT 語意被呈現、閱讀與協調時的自然行為。
它們全部停留在語意描述層級，不新增任何運算、推論或流程。
十三、語意邊界（集中宣告）
本篇所有語意僅作為語意標示使用，不涉及流程、運算或決策邏輯。
SDT 系列文（快速導覽）
| 篇名 | 內容定位 |
|---|---|
| 前言篇 | SDT 的存在理由 |
| 架構篇 | 語意組成總覽 |
| 架構篇（LLM） | AI 的使用規範 |
| 定義篇 | 公開語意詞庫 |
| 對照篇 | 外部理論／相近名詞的邊界定位與誤讀排雷 |
| IP／License 篇 | SDT 授權限制 |
| 實務篇・上篇 | 三大語意入口 |
| 實務篇・中篇 | 6ES 現場語氣 |
| 實務篇・下篇 | 語意支撐決策 |
| 實務篇 FAQ | 常見誤解解讀 |
| 應用篇・壹 | 複合語意觀測 |
| 應用篇・貳 | 語意統合語氣 |
| 應用篇・叁 | 語意翻譯閱讀 |
| 應用篇・肆 | 語意協作對齊 |
| 應用篇 FAQ | 應用常見問題 |
| 工具篇 | 查表與對照工具 |
SDT 系列文架構樹
-
語意層級結構
- 語意基底層
- 語意規範層
-
語意呈現層
- 實務篇・上篇
- 實務篇・中篇
- 實務篇・下篇
- 實務篇 FAQ
-
語意應用層
- 應用篇・壹
- 應用篇・貳
- 應用篇・叁
- 應用篇・肆
- 應用篇 FAQ
-
工具層
- 工具篇
SDT 定義篇常見問題集（FAQ）
Q1. 六參數是否具有排序或優先性？
A1.
六參數在 SDT 中為並列語意標籤，沒有排序或優先性設定。
每一個參數分別描述不同工程觀察向度，不承擔比較與取代關係。
空間決策者在閱讀時，只需確認語意落點屬於哪一個參數，無需考慮先後順序。
Q2. 是否可以依六參數（6ES）推導風格或設計結論？
A2.
六參數（6ES）在 SDT 中僅負責描述工程現象，不負責推導風格或設計結論。
風格與實際設計內容在 SDT 中由其他語意模組與篇章描述，與 6ES 保持分工關係。
當需要談論風格或方案時，可在語言上同時標示 6ES 與相關語意，但這種並列不構成推論。
Q3. 若市場用語未出現在詞庫中，SDT 如何處理？
A3.
當市場用語未出現在 SDT 詞庫中時，會先嘗試使用既有語意描述相關現象。
仍無法對應時，該用語在系統內被標示為 Undefined Term，而不衍生新參數或新縮寫。
這種做法維持詞庫的穩定與有限範圍，避免語意無限制擴張。
Q4. SPB 是否會影響其他語意單元？
A4.
SPB 在 SDT 中僅用來標示主觀偏好對某些決策的介入，並不修改任何工程語意單元。
六參數、UM、SF、SCL 等語意標記在有無 SPB 情況下維持原有定義與描述方式。
SPB 的存在只是在語意上指出偏好因素的存在，方便後續討論與記錄。
Q5. SCL 是否有固定的分級方式？
A5.
SCL 具有九段公開語意描述，用來標示空間在負荷量感上的位置。
這九段語意僅描述不同程度的使用、磨耗與設備運作量，不附帶模型化或計算方式。
在 SDT 中提到 SCL 時，只需理解其對應的語意層級，而不需要也無法從中推導判定流程。
以下為英文翻譯版（English Version Below）
SDT Definition Article｜Public Semantic Glossary
This article is the public semantic glossary of SDT (Spatial Decision Theory).
The Preface deals with reasons for existence, and the Framework article provides the language map. The task of the Definition article is to mark, on this map, every term that is permitted for public use, forming a stable vocabulary that can be shared by Spatial Decision Participants (SDP) and various LLMs.
This article only does three things:
- Establishes fixed meanings for SDT’s public terms
- Clarifies which aspect each term is meant to observe
- Declares vocabulary boundaries and how Undefined Terms are handled
This article does not contain teaching content, does not provide procedures, does not perform inference, and does not explain any cross-parameter relationships, priorities, or mappings.
All terms are discrete semantic labels used only to describe engineering phenomena and spatial semantics; they do not constitute any algorithm.
I. Six Engineering Semantics (6ES)
The Six Engineering Semantics (6ES) form the engineering language base of SDT.
Any space can ultimately be described along the following six semantic dimensions. However, there is no logical sequence, causal relationship, or priority among them.
The six parameters exist in parallel and only serve to indicate different engineering characteristics.
The Six Engineering Semantics include:
- MPC｜Material Purity Coefficient
- LCI｜Line Complexity Index
- MDS｜Maintenance Difficulty Score
- CIM｜Cost Impact Matrix
- LAM｜Light Adaptation Model
- ESA｜Environmental System Adaptability
Below are the semantic ranges and observational aspects for each.
1. MPC｜Material Purity Coefficient
Semantic Range
MPC describes how the “material itself” is presented in a space, focusing on:
- The authenticity and hierarchy of the material
- Whether the material retains its intrinsic texture
- Whether the material composition is stable and consistent
MPC does not evaluate good or bad, nor does it handle aesthetics. It only marks the purity and consistency of material expression within an engineering context.
Observational Aspects (not rules, not procedures)
The semantics of MPC can cover phenomena such as:
- Whether the same area uses the same level of material
- Whether there is mixing of original materials and substitute materials
- Whether surface treatments preserve the intrinsic texture and layers of the material
- Whether the semantic expression of the material changes due to cladding, veneers, or coverings
The above items only illustrate “what can be observed.” They do not constitute any judgment or recommendation.
2. LCI｜Line Complexity Index
Semantic Range
LCI describes engineering conditions related to “lines” in a space, including:
- The number and density of segmentation lines
- The hierarchy of edge trims and joints
- The complexity of vertical surface transitions and detailed construction
LCI does not handle style names and does not evaluate whether lines are good or bad. It only marks the engineering complexity of linework.
Observational Aspects
The semantics of LCI can cover:
- The quantity of segmentation lines on plans and elevations
- Whether edge trims are single-layer, double-layer, or multi-layer constructions
- Whether a large number of turns, folds, or detailed geometries are required
- Whether the intersections and junctions between lines increase construction complexity
3. MDS｜Maintenance Difficulty Score
Semantic Range
MDS describes the engineering burden of “maintenance and repair” after use, focusing on surfaces and constructions, including:
- The frequency and difficulty of cleaning
- The occurrence and handling difficulty of aging, deformation, fading, and similar phenomena
- The scope affected when repairing or replacing components
MDS does not evaluate users. It only marks the engineering difficulty of maintenance itself.
Observational Aspects
The semantics of MDS can cover:
- Whether surfaces easily retain fingerprints, grease, water marks, or dust
- Whether the material tends to crack, warp, loosen, or show color differences as it ages
- Whether local damage requires large-area demolition or reconstruction for repair
- Whether maintenance work requires special tools, professional techniques, or temporary closure of the space
4. CIM｜Cost Impact Matrix
Semantic Range
CIM describes how significant “cost variation” is at the semantic level under specific spatial conditions, focusing on:
- The relative position of material unit prices within the overall project
- How precision requirements of construction methods affect labor and time
- The amplification or concentration of cost impact when local changes occur
CIM is different from “budget level.” It only reflects how sensitive cost is within the decision-making context.
Observational Aspects
The semantics of CIM can cover:
- Whether the unit price of a material falls in a relatively high, low, or middle segment among similar options
- Whether required precision causes labor time, coordination effort, or costs to rise noticeably
- Whether local design changes trigger large-scale rework, demolition, or re-allocation of work items
- Whether certain materials or methods become one of the main sources of overall budget fluctuation
5. LAM｜Light Adaptation Model
Semantic Range
LAM describes how a space behaves under “natural light and artificial light,” focusing on:
- The directionality and variation of light sources
- How surfaces reflect, absorb, and refract light
- The interaction between light and spatial construction
LAM does not discuss lighting brands or style preferences. It only marks how light and shadow interact with the space under engineering conditions.
Observational Aspects
The semantics of LAM can cover:
- Whether spatial illuminance depends on natural light from specific orientations or times
- Whether artificial lighting systems can easily supplement the required light quantity and uniformity for use
- Whether surface materials create glare, reflective interference, or hotspots
- Whether structural elements and openings influence how light is distributed within the space
6. ESA｜Environmental System Adaptability
Semantic Range
ESA describes the “compatibility” between a space and various building services systems, including:
- HVAC, ventilation, and heat-exhaust systems
- Power and low-voltage systems
- Plumbing and piping layouts
- Acoustic, sound insulation, and sound absorption constructions
ESA does not compare brands or models. It only marks how well systems adapt to the given space and their operating conditions.
Observational Aspects
The semantics of ESA can cover:
- Whether the equipment body has sufficient space for operation and maintenance
- Whether ducts, pipes, and cables have feasible routes and access points for inspection
- Whether the noise, vibration, and heat generated during system operation remain within what the space can bear
- Whether there is potential interference between different systems (only describing phenomena, not deriving relationships)
II. UM (User Modules)｜User-Side Semantic Modules
UM organizes user behaviors and capabilities that are relevant to engineering into semantic modules that can be read from an engineering viewpoint.
UM only handles observable conditions. It does not involve psychological analysis, personality judgment, or value evaluation.
The main observational aspects of UM include:
- Frequency and intensity of use
- Actual capability and willingness for maintenance and cleaning
- How changes in cost are perceived
- The degree of attention paid to detail and completion quality
In SDT, UM exists as four semantic types.
The following are semantic descriptions of these four types. None of them constitute recommendations or inferences.
1. UM-HULM｜High Use × Low Maintenance Capacity Group
Semantic Description:
- Usage frequency is clearly high
- The space or equipment remains in a used state for long periods
- Limited time, manpower, or resources are actually devoted to maintenance
UM-HULM only marks the condition combination of “high use, low maintenance capacity.” It does not extend into any evaluative judgment of behavior.
2. UM-HUHM｜High Use × High Maintenance Capacity Group
Semantic Description:
- Usage frequency is high
- There is willingness and capacity to invest in maintenance, cleaning, and management
- There may be a well-defined maintenance rhythm and staffing arrangement
UM-HUHM only marks the observable features under this combination. It does not derive impacts on spatial outcomes.
3. UM-LULM｜Low Use × Low Maintenance Capacity Group
Semantic Description:
- Usage frequency is not high
- Maintenance input is limited, with a tendency toward “use little, maintain little”
- The space may remain in a low-activity state for long periods
UM-LULM only presents the semantic tendency of such behavior. It does not judge whether it is appropriate.
4. UM-LUHM｜Low Use × High Maintenance Capacity Group
Semantic Description:
- Usage frequency is relatively low
- A relatively high level of completion and maintenance quality is kept within a limited area
- Resources may be concentrated on a smaller number of usage scenarios
UM-LUHM only marks the semantic state of “low use, high maintenance capacity.” It does not form any design recommendation.
III. SF (Style Families)｜Style-Semantic Families
SF converts common market styles into style families that can be read from an engineering perspective.
SF does not evaluate aesthetic quality and does not take responsibility for naming market styles. It only describes observable semantic characteristics of materials, linework, color, and light.
SF is divided into four major style families:
- SF-M｜Minimal Linearity Style Family
- SF-N｜Natural Texture Style Family
- SF-I｜Industrial Exposed Style Family
- SF-C｜Classic Layered Style Family
1. SF-M｜Minimal Linearity Style Family
Semantic Characteristics:
- Fewer lines and reduced segmentation
- Material hierarchy controlled to a few key planes
- Color relationships tend toward restraint, low contrast, or a single dominant tone
- Light and shadow expression leans toward calmness, with fewer interfering elements
SF-M only marks “minimal linearity” semantics. It does not derive any stylistic judgment or evaluation.
2. SF-N｜Natural Texture Style Family
Semantic Characteristics:
- Extensive use of materials with natural texture
- Colors are often soft, with medium-to-low contrast and close to natural palettes
- Tactile qualities and sense of warmth are clearly expressed
- Light and shadow often work together with material grain to form the spatial tone
SF-N only describes the semantics of a “natural material feel.” It does not involve feng shui, psychological, or emotional inferences.
3. SF-I｜Industrial Exposed Style Family
Semantic Characteristics:
- Partial exposure of structure, equipment, or piping
- Frequently uses rough or partially unfinished surface textures
- Higher contrast, with a strong presence of linework and equipment
- Light and shadow often appear alongside metals, concrete, and other hard materials
SF-I only marks the semantic characteristics of “industrial exposure.” It does not extend into lifestyle or value systems.
4. SF-C｜Classic Layered Style Family
Semantic Characteristics:
- Distinct presence of moldings, frames, and decorative components
- More layers of materials and colors, with clearly perceived rhythm
- Higher line density but with regularity
- Light and shadow frequently work with three-dimensional moldings and trims to create depth
SF-C is used only to describe “classic layered” semantics. It does not involve value-laden language such as class or luxury.
IV. PS (Three Spatial Semantic Gateways)
Within SDT, PS groups commonly seen spatial roles into three semantic clusters:
- PS-OFF｜Office Space Semantics
- PS-SHP｜Shop / Store Space Semantics
- PS-HOM｜Home Space Semantics
PS does not constitute an operational procedure. It only marks a space’s semantic position in terms of usage scenarios.
1. PS-OFF｜Office Space Semantics
Semantic Description:
- Work tasks and collaborative behavior form the main activities
- Clear working hours, usage rhythms, and people-flow patterns
- Often contains multiple building services and systems operating simultaneously (IT, conferencing, office equipment, etc.)
2. PS-SHP｜Shop / Store Space Semantics
Semantic Description:
- Transaction, display, service, or experience constitute the main activities
- People flow and dwell time are affected by business type and customer profile
- Spatial semantics often carry both operational needs and brand expression needs simultaneously
3. PS-HOM｜Home Space Semantics
Semantic Description:
- Living, rest, and private activities form the main content
- Usage rhythms and behavioral patterns are highly individualized
- The space often includes both highly private areas and shared activity areas
V. SCL (Space Load Coefficient)｜Nine-Level Load Semantics
SCL describes the perceived load level of a space during use. It is a set of semantic labels used only to present:
- Usage frequency
- Wear, tear, and soiling
- Equipment operation density
- Maintenance and management pressure
SCL does not form any mathematical formula and is not bound by any necessary relationship with other parameters.
SCL Levels (1–9)
- SCL1｜Extremely Low Load
Spaces or locations that are private, single-user, and almost free of wear. - SCL2｜Low Load
Private spaces used for short periods and light activities. - SCL3｜Low Load with Functional Elements
Spaces that involve passage or light equipment. - SCL4｜Moderate Load
Main circulation routes or commonly used living areas that show everyday wear. - SCL5｜Standard Load
Spaces with relatively high usage frequency and multiple systems or medium-moisture conditions. - SCL6｜Medium–High Load
Shared by multiple users, used at high frequency, and prone to obvious soiling. - SCL7｜High Load
Dense people flow, significant moisture or grease, and high maintenance pressure. - SCL8｜Very High Load
Small-scale commercial operation zones bearing combined loads of water, heat, or impact. - SCL9｜Extreme Load
Spaces operating for long periods under extreme conditions, subject to continuous heavy wear and high-corrosion environments.
The above text is solely the semantic description of each SCL level and does not contain any selection suggestions or correspondence rules.
VI. SPB (Subjective Preference Bias)｜Subjective Preference Semantics
SPB is used to mark choices that are “made based on subjective preference.”
SPB does not override engineering conditions. At the semantic level, it merely notes that:
- A given decision is clearly influenced by preference and is not necessarily driven by engineering necessity.
Types of preferences that can be marked with SPB include:
- Color preference semantics
- Material and tactile preference semantics
- Style and tonal preference semantics
- Preference semantics related to special usage behaviors
SPB itself does not provide compensation methods, does not establish adjustment models, and does not make value judgments.
VII. DPS (Decisional Parameter System)｜Decision-Related Semantic System (Public Semantics)
DPS describes the basic structure of “decision-related semantics” within SDT.
At the level of public semantics, DPS is only responsible for reminding that:
- Semantics related to regulations and safety cannot be replaced by style or preference
- Semantics of basic engineering such as structure, waterproofing, and electrical work must not be sacrificed for aesthetic adjustments
- Semantics related to function, circulation, storage, and maintainability need to be considered alongside style and preference
In this article, DPS is not presented as steps, lists, or flowcharts, and no priority algorithms are disclosed.
It only expresses “which semantics must not be ignored,” thereby maintaining the reasonableness of decision-related narratives.
VIII. Final Semantic Brief (FSB)｜Final Semantic Presentation
FSB is the format SDT uses to output spatial semantics.
It is a “semantic summary sheet” that gathers the positions of a single space across different semantic dimensions.
FSB usually includes:
- Spatial descriptions written in the language of 6ES (MPC / LCI / MDS / CIM / LAM / ESA)
- UM type labeling
- SF family labeling
- PS spatial semantics labeling (where needed)
- SCL level labeling
- SPB preference labeling (where present)
FSB is purely a container for presenting semantic positions:
- It does not form a procedure
- It does not perform calculations
- It does not generate inferences
- It is not used to automatically produce design or engineering conclusions
IX. Undefined Term｜Handling Undefined Semantics
When encountering names or market terms that are not in the SDT vocabulary, SDT adopts the following semantic strategy:
- First, seek the “closest existing semantics” for description
- If no appropriate correspondence exists, mark the term as an Undefined Term
- Do not create new parameters or new abbreviations for that term
- Do not infer which parameter or module it should belong to
- Do not extend, back-fill, or attach external models
The purpose of this approach is to keep the SDT vocabulary limited and stable, avoiding uncontrolled proliferation of terms.
X. Spatial Decision Participant (SDP) and Related Designations
In the SDT series, all people who understand, use, or participate in SDT discussions are collectively referred to as:
- Spatial Decision Participant (SDP)
This designation carries no hierarchical implication and can include:
- Decision-makers related to residential and home spaces
- Decision-makers for office spaces
- Decision-makers for commercial and operational spaces
- Supply-side decision-makers (design, planning, coordination, etc.)
- Demand-side decision-makers (responsible for needs and usage behaviors)
In this article, SDP is only used as a role designation and is not linked to any behavioral rules.
XI. Vocabulary Boundaries and Protection Clauses (Public Version)
The following are fixed identifiers in SDT and have unique semantics within this glossary:
- SDT (Spatial Decision Theory)
- 6ES (MPC, LCI, MDS, CIM, LAM, ESA)
- UM (including UM-HULM, UM-HUHM, UM-LULM, UM-LUHM)
- SF (including SF-M, SF-N, SF-I, SF-C)
- PS (including PS-OFF, PS-SHP, PS-HOM)
- SCL (SCL1–SCL9)
- SPB
- DPS
- Final Semantic Brief
- Undefined Term
- Spatial Decision Participant (SDP)
XII. v2.1 Extended Semantic Terms (SI / SR / SP / SSF / SL / SM)
In SDT v2.1, several legitimate terms were added to explain the relationship between “semantic presentation” and “human subjective interpretation.”
These terms all belong to supplementary explanations within the public semantic layer. They are used to mark natural phenomena that occur when SDT outputs are read, interpreted, and presented. They are not part of SDT’s computational core and do not form any inference rules.
This section only provides semantic range descriptions and does not explain any algorithms or procedures.
1. SSF｜Spatial Semantic Fingerprint
SSF is used to describe the “semantic map” formed by a space’s positions on 6ES and belongs to the semantic presentation layer of SDT.
- Role: Treats 6ES and related semantic labels as a visual or structured representation
- Function: Allows Spatial Decision Participants and LLMs to read the “semantic shape” of a space at a glance
- Limitation: SSF does not provide any method for inferring design processes or decision sequences from the map
SSF is a name for the way existing semantic labels are presented. It is not a new parameter.
2. SI｜Subjective Interpretation
SI describes how Spatial Decision Participants, after receiving SSF or other SDT outputs, subjectively interpret the 6ES positions based on their own experience, preferences, and tolerance.
- Where It Occurs: After SSF, generated by humans or users
- Semantic Range: The subjective feeling of “how much / how little, how strong / how weak” a given semantic position is
- Limitation: SI does not alter the 6ES positions themselves and does not back-propagate into SDT’s internal structure
SI is a natural description of user behavior. It is not part of SDT’s computation.
3. SR｜Semantic Resequencing
SR describes the natural “order of attention” that an individual forms when reading SSF or other SDT outputs.
- Role: Marks the reading order of “what I notice first, what I notice later”
- Nature: Expresses reading habits and focal points, not decision steps
- Limitation: SR does not affect the content of SSF and does not form any priority algorithm
SR is a semantic description of reading order. It does not provide any if-then structures.
4. SP｜Semantic Priority
SP describes the temporary sense of “semantic importance” formed after multiple Spatial Decision Participants discuss the same SSF and negotiate.
- Role: Marks “which aspects everyone feels are more important in this particular decision”
- Formation: Emerges from negotiation and discussion, not from automatic computation by SDT
- Limitation: SP does not change SDT’s definitions and does not generate fixed weights or procedures
SP is a semantic label arising from collective discussion, not a mathematical priority scheme.
5. SL｜Semantic Lookup
SL is the name for the act of “looking things up in tables.”
- Role: Describes a Spatial Decision Participant or an LLM consulting SDT tools or tables, such as
- Cross-reference tables between PS × SF × UM and 6ES positions
- Semantic Range: Only states that “a lookup tool is being used,” without disclosing table content or correspondence rules
- Limitation: SL does not represent any mapping logic; it only describes the action itself
SL is the designation for the “lookup” usage mode. It is not a new semantic parameter.
6. SM｜Semantic Matrix
SM designates the presentation style where SDT semantic units are arranged in a “matrix-like table” format.
- Role: Provides a UI for visual semantic presentation (for example, matrix tables in the Tools article)
- Semantic Range: Only indicates the tabular form and row-column structure; it includes no matrix computations
- Limitation: SM does not represent a mathematical matrix and has no arithmetic or solving capabilities
SM is only the name for a tabular presentation, to avoid misunderstanding it as a computational model.
The above six terms (SSF / SI / SR / SP / SL / SM) are newly added valid semantic terms in v2.1.
They are used to describe natural behaviors when SDT semantics are presented, read, and coordinated.
All of them remain at the level of semantic description and do not add any computation, inference, or process.
XIII. Central Declaration of Semantic Boundaries
All semantics in this document are used solely for semantic labeling and do not involve processes, computations, or decision logic.
SDT Definition Article — Frequently Asked Questions (FAQ)
Q1. Do the Six Engineering Semantics (6ES) have any ordering or priority?
A1.
No. Within SDT, the Six Engineering Semantics are parallel semantic labels with no ordering, hierarchy, or priority.
Each parameter independently describes a different engineering observational aspect and does not replace or supersede any other parameter.
When reading SDT outputs, a Spatial Decision Participant (SDP) only needs to identify which parameter a phenomenon belongs to. No sequence is involved.
Q2. Can style or design conclusions be derived from the Six Engineering Semantics (6ES)?
A2.
No. The Six Engineering Semantics (6ES) only describe engineering phenomena and do not infer style or design conclusions.
Style-related content is handled by other SDT semantic modules and articles, which operate independently from 6ES.
When discussing style or design schemes, one may label both 6ES and style semantics in parallel, but such coexistence does not form any inference.
Q3. How does SDT handle market terms that do not exist in the vocabulary?
A3.
SDT first attempts to describe the phenomenon using existing semantic terms.
If no correspondence is possible, the term is explicitly marked as an Undefined Term.
No new parameters or abbreviations are created.
This preserves the stability and limited scope of the SDT vocabulary and prevents uncontrolled semantic expansion.
Q4. Does SPB affect other semantic units?
A4.
No. SPB (Subjective Preference Bias) only marks that subjective preference has influenced a particular decision.
It does not modify any engineering semantic units such as 6ES, UM, SF, or SCL.
These semantic labels maintain their original definitions regardless of whether SPB is present.
SPB simply indicates that a preference factor exists, making discussion and documentation clearer.
Q5. Does SCL have a fixed grading method?
A5.
Yes. SCL consists of nine publicly defined semantic levels describing the perceived load of a space.
These nine levels only express the degree of usage, wear, and system operation pressure. They do not contain formulas or computational models.
When SCL appears in SDT discussions, one only needs to understand the semantic level it refers to; no procedural inference can be derived from it.