Citation - Svenonius, E. (2000). The intellectual foundation of information organization: MIT press.

Keyword -

• The first part: intellectual foundation of information organization.
  • Chapter 1 [Introduces and defines: intellectual foundation, information ,and document] It establishes a conceptual framework that identifies the central purpose of systems for organizing information: bringing like things together and differentiating among them (同其所同,異其所異). It considers the function of principles in the context of systems design and concludes with an illustration of some of the problems encountered in the design of organizing systems.
• The second half of the book moves from generalities to particulars. It presents an overview of three bibliographic languages used to organize information – work languages, document languages, and subject languages – and looks at these languages in terms of their vocabulary, semantics, and syntax.
  • Chapter 8 [Subject language - alphabetic LC vs classification DDC]
  • Chapter 9 [Subject language - semantics]
  • Chapter 10 [Subject language - syntax]

bibliographic – information retrieve system's objective:

• the traditional objective:
  此節回顧歷史發展
  • Antonio Panizzi(1850), 作者名書目 (字順式目錄)規則
  • Charles Ammi Cutter(1876): 尋求(finding), 集中(collocating), 選擇(choice)
  • Seymour Lubetzky(1960): 區分作品(work)與書本(book)。
  • Paris Principle (1961). 目的中，未包含選擇，也未確定說明 依主題聚集相關書目的目的。只是暫時性的書目系統主張。
  • IFLA (1997): FRBR
    • find 查詢: 找尋讀者所敘述的查詢條件, 無論是在檔案或資料庫中的單一款目或是一組款目, 都能以款目的屬性或關係被檢索到。
    • identify 辨識: 辨識實體, 即辨識相同的兩個或兩個以上的實體。
    • select 選擇: 選擇適合讀者需求的實體, 即選擇內容或外觀適合讀者需求的實體, 或拒絕不符合讀者需求的實體。
    • acquire or obtain 獲取: 被描述的實體, 可追過採購、借閱或電子連結的方式即可獲取。
• Sufficiency of objectives:
  此節 Svenonius 討論了 navigate 的目的
  • 瀏覽: 找到關於某作品的其他作品，如：與此作品具有概括、聚集、相連關係，或在屬性上等同、階層、相關的其他作品。
    To navigate a bibliographic database (that is, to find works related to a given work by generalization, association, or aggregation; to find attributes related by equivalence, association, and hierarchy).
• Objectives of a Full-Featured Bibliographic System:
  在此 Svenonius 提出涵蓋前兩項的五目的架構。
• Operationalization of Objectives:
  
• The Collocating Objectives: 集中。從 IR 的 precision, recall, relevance 探索 collocating 的 measurement
• Open-Ended Objectives: 處理無盡的描述與格式是不經濟的，討論核心(core)詮釋資料。
• Necessity of Objective:
• Implementation of Objectives in Present Systems:
• Implementation of Objectives in Future Systems:
  • Cost
  • User need: 一些研究顯示使用者並不那麼需要傳統完整的書目資料
  • Different system has different purpose.
  • Full-feature bibliographic system still has its user.

• Introduction: ontology, information entities 資訊實體
  • Objects of Description:
    • Ontology is a theory regarding the entities. The entities in bibliographic theory are Works, Editions, Authors, and Subjects.
    • Modeling bibliographic description: E-R model
  • conceptual and operational description
• Entity Types
  • Documents: “information-bearing messages in recorded form”
    • The set of all documents sharing essentially the same information (Work)
    • The set of all documents sharing the same information (Edition)
    • The set of all documents descended from a common origin (Superwork)
    • The set of all documents by a given Author
    • The set of all documents by a given Subject
  • Works: (never been satisfactorily defined) an intellectual or artistic creation. Lubetzky: “the material book and the intellectual work”.
  • Superworks: form Domanovzky's criterion of descent form a common origin. A superwork may contain any number of works as subsets, the members of which while note sharing essentially the same information content are nevertheless similar by virtue of emanating from the same ur-work.
  • Editions, Subedtion, Version:

bibliographic languages constructing principles: 書目語言五原則

1. user convenience, 便利用戶
2. representation, 表徵
3. sufficiency and necessity, 充分性與必要性
4. standardization, 標準化
5. integration. 整合

可參看 張慧銖（2003）。圖書館目錄發展研究。第四章, 第一節, 書目記述的原則。其論述基本上根據此章的譯成。

the difference between bibliographic…

• objective: codify what a user can expect…
• principle: directives for the design of bibliographic language (codes or rules) used to create such a system
• rule

充分性與必要性: Occam's Razor: = Principle of parsimony 簡約原則 (Ranganathan: canon of impartiality)

• Lubetzky 1953: “Is this rule necessary?”
  • 簡化編目規則: “…as his reform to simplify cataloging, set about eliminating redundancy in the rules for providing identifiers for bibliographic entities.”
• two studies
  • Library of Congress (US, 1946)
  • OCLC (1993): 根據13個資料欄位設計出來的機制，足以提供重複書目資料比對的工作 / 13 weighted data elements are sufficient for identifying duplicate records in 36 million records bibliographic database.
  • Seal (1983)

[Work language - Anglo-American Cataloging tradition]: 用來描述作品的語言: 語意結構與語法規範

• names attributes, entities, and relationships – to provide the means to find information
• derive and assigned metadata – to provide the normalization required to organize it
• syntax – to disambiguate vocabulary and order bibliographic displays
• others: membership, inclusion, equivalence, aggregation, sequence, and commentary

• 屬性 / Attributes
• 根據一般使用習慣選名 / Choice of Names: Commons Usage
• 明確名稱 / Disambiguation of Names: add Elements, Context, and Syntax
• 建立語意關聯 / Mapping Names: Semantic
• 實體 / Entities
  • Work ID: 作品識別子: 英美傳統以 Author-title 為主。其理由與挑戰。
  • Edition identifiers, superwork identifiers
• 關係 / Relationships (Tillett)
  • Membership
  • Inclusion
  • Equivalence
  • Aggregation
  • Sequence
• 未來展望 Observation: 傳統的書目編目需要調整到數位資訊時代的需求。

p.88-9

The definition of main entry is surrounded by confusion and controversy. Defined in terms of what it refers to, it can denote:

• A description of a document, in particular its full description (a meaning that is now obsolete);
• The attribute of a document by which it is most often accessed, its so-called main access point, such as a person, corporate body, or title.
• A work, which is designated either by a normalized name title or by a normalized title.

(p.95)

(p.95-6)

• 使用複合屬性作為主款目的原因
  • 沒有單一屬性足以當作作品的唯一性識別
    no single attribute suffices to identify a work uniquely
  • 作者寫出很多作品，且題名通常沒有識別性
    authors write numerous works, and titles are often not distinctive
• 以作者名作為前導項，因為
  • 傳統重視作者的作法
    honor tradition, authorship
  • 回應讀者習慣
    respond to users expectations

• 挑戰
  • 沒有作者、多作者、一人多名
  • 非西方讀者不必然由作者名開始查找館藏書本
  • 當代許多團體作者與非書資料的作者性難以判定，使用者也習慣用題名進行檢索

[Document languages - AACR2] Document languages are the languages used to describe space-time embodiments of information. 描繪能體現在特定時空的資訊

• 傳統的描繪對象: 書、文件
  • 描繪文件的三種面向
    • 物質或載體 / physical or carrier attributes
    • 出版 / publication attributes
    • 取用 / access attributes
• 傳統文件語言遇到新媒體時，所產生的問題:
  • The problem of classifying these media 分類問題
  • The problem of deciding what physical characteristics of nonbook media should or can be described in bibliographic records. 應該描繪哪些物質特徵的問題
  • The problem (brought to the fore by documents in digital form) of how to organize entities that lack essential descriptive attributes, because they are unstable or in flux 如何描繪易變且流動的數位文件的問題
  • The political and technological problem of creating stable, standard document identifiers 如何建立穩定、標準的文件識別：政治性與技術性的問題

• Description Principle: 混雜了不同的對象與原則
• Work(作品): 作者是否是必要屬性?

file link - Google Schloar, XXC