CC:DA/TF/Metadata/4

June, 1999

Task Force on Metadata

Summary Report

Charge #1: Analyzing the resource description needs of libraries.

The group decided at Midwinter that it would:

• review existing principles of the purpose of the catalog
• define library or catalog users and their expectations
• look at the context in which we are describing our resources

Principles or purposes of the catalog, users, and the context in which catalogs are used would be part of any analysis. The first step was to sketch out a general approach or understanding of these three elements — the purpose of catalogs, the users of catalogs, and the context in which catalogs are used.

1. From catalog principles to user tasks (consists of a summary from a preliminary review of existing statements of principles for catalogs)

   Based on a preliminary review of the literature, the group determined that one useful way to analyze resource needs is by breaking the needs into five basic tasks. The five tasks are derived from the Functional Requirements for Bibliographic Record’s (FRBR) four basic user tasks: find, identify, select, and obtain. The fifth task, taken from Rahmatollah Fattahi’s paper, “AACR2 and catalogue production technology” (Toronto conference paper, 1997), is what he calls “housekeeping”, meaning management or administration. Thus, “manage” or “administer is added to the tasks of find, identify, select, and obtain. The management task is simply a heading under which may be grouped those tasks which are necessary to an institution, its staff, or its business partners/vendors/etc. and which indirectly relate to fulfilling the four basic user tasks. The user needs are paramount; the manager needs are subordinate.

   Clearly, within any operational context, so loosely defined a task as manage would need to be further analyzed into precisely defined sub tasks. However, the same can be said of the FRBR’s four basic tasks without any diminishment in the usefulness of these concepts for analyzing the resource description needs of libraries or for analyzing the appropriateness of a metadata data element set, a library information management system, or a library taken as a whole, service-oriented enterprise. These high-level abstractions — find, identify, select, obtain, and manage — provide a solid basis for further analysis. Together they provide the point we may stand on as a profession to have a point-of-view.

   Our resource description needs are grounded in the needs of our users to find, identify, select, and obtain some information thing (book, article, map, score, data set, etc.) We judge our tools — catalogs, indexes, search engines, etc. — primarily by how well they do these tasks. And not only must they perform the user tasks well, but they must make the management task or house keeping as simple, easy, flexible, and cheap as possible.

2. The myth of the library user, or, One size doesn’t fit all (a reminder of the complexity of users both individually and as a group).

   Based on arm-chair speculation, recollections from experience, and a preliminary perusal of the literature, the group determined that it is not useful for this discussion to categorize users into types. Such categories as expert, novice, scholar, student, researcher, citizen, child, young adult, color blind, library staff, company employee, non-English speaking, only English speaking, etc. may indeed be highly valuable in some particular instance, but the main point to glean from this multiplicity is just that — users, even individual users, are (from the point of view of our definition of users into categories of users) multiple, complex, and protean.

   Thus, library resource descriptions, although grounded in meeting the four basic tasks of “the user,” cannot be based on any firm categorization of users into static, defined types. (For a similar view, see Carl Lagoze’s D-Lib article, “From Static to Dynamic Surrogates: Resource Discovery in the Digital Age” (D-Lib, June 1997). Lagoze’s point is that the multiplicity of roles that just one user can take on during the resource discovery process is one of the main contributors to the complexity of the resource discovery process.) We may be convinced from this that institutions or tools that base their resource description needs on simple, static notions of user roles will be inadequate to the demands of real users. A catalog or metadata scheme designed around the idea that the users only care about numeric data and not about images (or vice versa) will fail the user who cares about both.

3. Libraries and catalogs in a new context, or, How do networks affect libraries and their catalogs?

   The fact that we need to provide a “coherent environment for users” and that the library catalog is no longer at the center any more serves as a starting point for the group’s discussions and analysis of the context of the library catalog.

   The group determined that when thinking of the resource description needs of libraries we cannot think only of the catalog. Or, rather, that when we do think of the catalog, we cannot usefully think of it as a s tand alone or isolated tool. When we do think of the catalog, we must think of it as one of the important tools among a host of tools. Thus, it is a new breed of catalog that we must imagine.

   The network is making local or, more precisely, isolated catalogs extinct. We create our institutional catalogs from shared resources and we share our catalogs via networked resources. Our institutional catalogs are open to the world in ways we did not imagine they would be. A user 10,000 miles away can use our catalog. That user may be searching a hundred other catalogs at the same time. That user may want to integrate and format the search results in unpredictable ways. An isolated or local only catalog can’t contribute much to “a coherent environment for users.”

   All the thinking about catalogs that we did before the ascendency of the Internet (and the end of the local-only or isolated catalog) must now be revised. Look at the statement in the Paris principles: “the catalogue should be an efficient instrument for ascertaining . . .” It isn’t enough to be an efficient instrument for ascertaining anymore. In a networked environment, the tool isn’t done working until it has delivered the goods or come as close as possible. For some electronic materials the catalog or ancillary tools will be able to deliver the items; for materials that are not online, the catalog should at least present the user with the options for obtaining the items — call an item from local stacks or storage, recall an item in circulation, fill out an ILL request, connect to a document delivery service, etc.

   Our catalogs have become one tool among many, but those many are not separate or isolated from one another. The catalog is one tool in a network of tools. The basic or necessary principle of tools in a network of tools is interoperability (The definition of which group 3 is charged with). Or if interoperability is too high a demand, each tool in a network of tools must be compatible with each other from the user’s perspective. For example, a user employing an institution’s catalog to find data sets relating to census and voting in Hartford, Connecticut may well need to analyze the data once it has been located, and then format the analysis into a presentation document dominated by images not numbers. While the catalog per se is only one tool in this scenario, it should be compatible with a wide range of other tools that may be used as functional extensions of the discovery and retrieval process. What is desirable is a network of tools that are portable, flexible, agile, mappable, extensible, adaptable, a coherent network of tools. The library catalog can be part of such a coherent environment, but only if it is designed, maintained, and used as one tool in a network of tools.

The next step for this group may be to analyze these resource description needs in terms of future needs.

Charge #2: Building a conceptual map of the resource description terrain/landscape and developing models for accessing/using metadata both inside and outside the library community.

Working Group Number Two was charged with drawing up a conceptual map of the resource description landscape. It submitted such a map. The map draws on the list discussion in the fall of 1998.

The map consists of five columns. For each resource category, the group has outlined what modes of description and access existed before computers (if relevant), what exists currently, and what it sees for its future. For some resource categories, there are additional comments. Some categories are amalgams of like materials, for example, pamphlets, vertical files and offprints or technical reports, dissertations, etc. (See Appendix 1.)

The group will continue to work on developing models for accessing/using metadata based on its conceptual map.

Charge#3: Devising a definition of metadata and investigating the interoperability of newly emerging metadata schemes with the cataloging rules and MARC format.

The subcommittee decided to employ a two-phase approach as its operating procedure. In phase one, three of the subgroup members would collect and submit definitions of “metadata,” “interoperability” and of newly emerging “metadata schemes” for open discussion and comment on the CC:DA’s electronic mail system, metamarda-L. In phase two, these definitions would be filtered and evaluated against AACR2 and the MARC format to devise a working definition(s).

The three team members supplied a number of definitions culled from various sources that were disseminated onto the electronic mail server for consideration and open discussion by members of the subcommittee and other members of the entire task force. Beginning with the term “metadata” followed by “interoperability” and “metadata schema”, each term received potential definitions for consideration and discussion over a period of several weeks followed by a period of reevaluation and refinement followed by reconsideration and discussion before proceding to the next term. Overall, this sequential process of volunteered contributions for each term resulted in 27 potential definitions for metadata, nine for interoperability, and ten for metadata schema.

This method of operation proved to be a contextual and an evolving process, i.e. definitions submitted first were reconsidered and reworked in light of subsequent evaluations, discussion, and new submissions. The course of the discussion revealed that the varying definitions for each concept appeared to depend on intent and context informing and surrounding the definition. These deliberations also revealed that concepts found within the definitions of the terms being considered were not as transparent and understandable as originally assumed. After soliciting further input and comment on these developments the consensus seemed to be that the terms under consideration by the task force remain somewhat nebulous concepts that would benefit from further refinement. One possible alternative to an denotative definition satisfactory to all parties and considerations may be to offer a definition followed by examples that provide further elaboration and clarification. This approach along with the definitions submitted and considered for all three terms are included in Appendix 2. However, for purposes of this report, the majority of participants felt that we should offer a best attempt at working definitions for all three concepts.

The formal working definitions for the three terms deliberated and submitted by the task force subcommittee for charge #3 follow below (submitted May 17, 1999):

• METADATA are structured, encoded data that describe characteristics of information-bearing entities to aid in the identification, discovery, assessment, and management of the described entities
• INTEROPERABILITY is the ability of two or more systems or components to exchange information and use the exchanged information without special effort on either system.
• A METADATA SCHEME provides a formal structure designed to identify the knowledge structure of a given discipline and to link that structure to the information of the discipline through the creation of an information system that will assist the identification, discovery and use of information within that discipline.

Using these working definitions, the group will continue to focus on interoperability of emerging metadata schemes with cataloging rules and MARC.

Charge 4: Recommending ways in which libraries may best incorporate the use of metadata schemes into current library methods.

At ALA Midwinter 1999 this group was asked how best to proceed in this direction. The group came up with a definition of what a “prototype” library catalog would look like in the future (given below):

• have the patron use ONE search interface to access all information, whether it is a number of different metadata types and standards, databases, and OPAC(s)
• provide a seamless transition to the user to all information available, moving from the ILS system of a front-end search mechanism that accesses numerous resources, to a search interface that can access all information available in any standard, format, location, or subject. (Example: interface can search local OPAC, World Wide Web, metadata standards (EAD, TEI, GILS, Dublin Core), special collections, museum holdings, etc., and present results in a useable format to the patron through one search mechanism)

Our definition of prototype is: a virtually seamless access to information and relevant retrieval of information from the user’s point of view.

The following “prototype” library catalogs were mentioned by members of the Task Force. This is merely a list that members have said MAY be prototypes; they have not been examined or explored by members of this group for comparison with the definition of “prototype.” As a next step, the group will analyze these prototypes further.

Propotypes:

• BIBLINK
  http://hosted.ukoln.ac.uk/biblink/
• Cooperative Online Resource Catalog (CORC)
  http://corc.oclc.org
• Arts and Humanities Data Service (articles about at)
  http://ahds.ac.uk/public/metadata/discovery.html
• Getty Research Institute auction catalog records
  http://opac.pub.getty.edu
• a.k.a. project by Getty Research Institute (public site to be pulled shortly)
  http://www.ahip.getty.edu/aka/
  [Powerpoint presentation on a.k.a.]
• Arthur project by Getty Research Institute
  http://www.ahip.getty.edu/arthur/
• Other projects of interest
  http://facesla.org

The groups will meet at the Annual 1999 meeting in New Orleans and determine how to further proceed. The agenda for the meeting is as follows:

8:30-10:15

• Call to order
• Presentations on 4 charges by leader(s)/other group members of midwinter breakout groups
• Proposed preconference on metadata for Annual 2000: ideas as to what The Ideal Preconference on Metadata would have as subject matter, speakers, etc.
• Plan for coming year (TF goes out of existence annual of 2000)

10:45-12:30

• RDF and XML / Eric Miller; Diane Hillman
  • Presentations
  • Questions and answers

Those interested in attending the RDF and XML session should RSVP to Mary Larsgaard since seating may be limited. Those who responded will get first choice on seating.

The Task Force will continue its deliberations with the goal of analyzing charge #5 and submitting rule revision recommendations as needed. Work should be completed by Annual 2000.

The Task Force is also involved in presenting a preconference on metadata at the ALA Annual meetings in Chicago in the year 2000. Daniel Kinney is putting together a committee of the CC:DA Task Force on the metadata preconference.

Metadata:

1. Data about data.

2. Tool to accomplish various processes — Clifford Lynch from Keynote speech, Managing Metadata for the Digital Library: Crosswalks or Chaos?

3. A cloud of collateral information around a data object — Clifford Lynch (ibid.)

4. Descriptive data about a resource that relieves the user of having to have full access to the resource in order to know of its existence — Don Waters from “I Know it’s out There but Where?” Metadata conference

5. Data that documents or tracks the change or uses of data — John Perkins from his presentation on the Coalition for the Interchange of Museum Information, Metadata conference

6. Structured data about data — Carl Lagoze from “The Warwick Framework : Managing Chaos through Containment”

7. All of it is just data — no Meta-metadata — all data requires mechanisms for discovery, management and access control — Clifford Lynch again, Metadata conference

8. Metadata describe the content, quality, condition, and other characteristics of data. Metadata help a person to locate and understand data — FGDC content standard

9. Meta is a prefix that in most information technology usages means “an underlying definition or description.” Thus, metadata is a definition or description of data and metalanguage is a definition or description of language. Meta (pronounced MEH-tah in the U.S. and MEE-tah in the U.K.) derives from Greek, meaning “among, with, after, change.” Whereas in some English words the prefix indicates “change” (for example, metamorphosis), in others, including those related to data and information, the prefix carries the meaning of “more comprehensive or fundamental.” — whatis.com

10. Data about data. Metadata describes how and when and by whom a particular set of data was collected, and how the data is formatted. Metadata is essential for understanding information stored in data warehouses — Webopedia.

11. Metadata is “data about data” (for example, a library catalog is metadata, since it describes publications) or specifically in the context of this specification “data describing Web resources”. The distinction between “data” and “metadata” is not an absolute one; it is a distinction created primarily by a particular application, and many times the same resource will be interpreted in both ways simultaneously. — Resource Description Framework Model and Syntax Specification.

12. Metadata is an abstraction from data. It is high-level data that describes lower-level data — Briefing paper: What is metadata computerwire.com.

13. Structured description about an object or collection of objects; with descriptive, administrative, and structural applications — Roy Tenant, during the Internet Librarian 98 Conference in Monterey (Supplied by George J. Janczyn)

14. The term ‘metadata’ commonly refers to any data that aids in the identification, description and location of networked electronic resources — p.1 of Hudgins, Jean, Grace Agnew, and Elizabeth Brown. 1999. Getting mileage out of metadata: Applications for the library. Chicago: American Library Association.) (Supplied by Erik Jul)

15. In data processing, meta-data is definitional data that provides information about or documentation of other data managed within an application or environment. For example, meta data would document data about data elements or attributes, (name, size, data type, etc) and data about records or data structures (length, fields, columns, etc) and data about data (where it is located, how it is associated, ownership, etc.). — Free On-line Dictionary of Computing.

16. Information about data, or more specifically, the descriptive information rovided in meta tags in an HTML or XML document header about that document. — Glossary of Internet Terms.

17. The descriptive information meta-data supplies allows the user to locate, evaluate, access, and manage online available learning resources. Describing learning resources (whether materials, activities, people or enterprises) digitally is similar in the physical world to attaching a label to an object, such as a can peas or a package of light bulbs. The label provides information about the contents of its container without having to actually open the container. . . . Examples of meta-data for learning materials are the title, the author, the targeted learning level, and the educational objectives of the material. . . . Meta-data is distinct from, but intimately related to, its contents. . . . Overall, meta-data serves as a complement to its contents and reflects the content’s attributes to interested users. — Instructional Management Systems Project: What is Meta-data? “The Meta Data Coalition (formerly Metadata Coalition) regroups vendors and users allied with a common purpose of driving forward the definition, implementation and ongoing evolution of a meta data interchange format and its support mechanisms. The need for such standards arises as meta data, or the information about the enterprise data emerges as a critical element in effective data management.” — The Meta Data Coalition.(Supplied by Vianne Sha)

18. Metadata is data about data. Cataloging in a library setting is an example of metadata. But in the Internet environment that involves commerce and services as well as objects, it has more functions than description and resource discovery:
    • intellectual property rights including the contractual terms related to the document’s use and distribution
    • electronic commerce to encode prices, terms of payment, etc.
    • content rating to disclose the nature of a particular page’s contents which can be used in filtering content so that, for example, parents can block inappropriate material from children
    • digital signatures: can you trust this document?
    • privacy issues: what information do browsers collect when you visit? what information are users willing to disclose about themselves when visiting a Web site?

19. Metadata is a means of assigning descriptive tags to government information so that it is easily searchable and retrievable by clients from any electronic or online tool, be it an Internet connection through their home or business PC, via a public kiosk or by using a call centre as the middle access environment (the Government Information Centre is a good example of this environment). In non technical terms metadata then is information which describes information.

    Some characteristics of metadata:
    

    • It is readable by both humans and machines.
    • Metadata takes a variety of forms, both specialized (VRA) and general (DC), and may be part of a larger framework (TEI).
    • New metadata sets will develop as the networked information infrastructure matures.
    • Different communities will propose, design, and maintain different types of metadata — Kathleen Forsythe and Diana Brooking, two librarians from the University of Washington Libraries, gave an excellent presentation at the Online Northwest Annual Conference (Supplied by Mark Watson).

20. Metadata. An encoded description of an information package (e.g., an AACR2 record encoded with MARC, a Dublin Core record, a GILS record, etc.); the purpose of metadata is to provide an intermediate level at which choices can be made as to which information packages one wishes to view or search, without having to search massive amounts of irrelevant full text. — Arlene Taylor

21. Metadata is data associated with objects which relieve their potential users of having to have full advance knowledge of their existence or characteristics. A user might be a program or a person, and metadata may support a variety of uses or operations — Rebecca Guenther

22. Metadata is a structured description of the content, quality, condition, usage, and other characteristics of data. They enable users to discover, locate, understand, and evaluate data. They also enable administrators to manage data and control access to them.

23. Metadata is data about data, structured to meet the needs of information olders, managers, and users. It helps users to discover, locate, understand, and evaluate data, and helps data administrators to manage data and control access and use. For example, metadata may describe how, when, why, and by whom a particular data object or set of data was collected or created, what its content is, how it is formatted, and the conditions for its use — Clare Imholtz

24. Metadata is a structured, encoded description of an information package. Metadata provides an intermediate level at which viewing or searching choices can be made in light of data characteristics like content, quality, condition and usage. Metadata enables users to find, identify, select and obtain information packages. Metadata also enables administrators to manage information packages and control access to them. — Mark Watson

25. Metadata pl. n. (used with a sing. or pl. verb) 1. Surrogate information about or related to a resource, 2. A structured, encoded surrogate. — Erik Jul

26. METADATA is a structured, encoded description of an information package which serves an intermediary role between the user and the information package by describing data characteristics and allowing for viewing or searching choices to be made. Metadata enables users to find, identify, select and obtain information packages and also enables administrators to manage information packages and control access to them

27. Metadata pl. n. (used with a sing. or pl. verb) 1. Surrogate information about or related to a resource, 2. A structured, encoded surrogate Definition of surrogate: The OED offers: A. sb. 1. A person appointed by authority to act in place of another; a deputy. and 2. a. fig. and gen. A person or (usually) a thing that acts for or takes the place of another; a substitute. Const. for, of. b. spec. = substitute sb. 6 b. and following up on the above suggestion of “substitute,” we find my personal favorite. 6. In technical use. c. Mech. A short section used when a full-length section is not usable.

Interoperability:

1. Interoperability is the ability of a system or a product to work with other systems or products without special effort on the part of the customer. Products achieve interoperability with other products using either or both of two approaches: By adhering to published interface standards, or By making use of a “broker” of services that can convert one product’s interface into another product’s interface “on the fly.” — adapted from Whatis.com

   A good example of the first approach is the set of standards that have been developed for the World Wide Web. These standards include TCP/IP, HTTP, and HTML. The second kind of interoperability approach is exemplified by the Common Object Request Broker Architecture (CORBA) and its Object Request Broker (ORB).

   Compatibility is a related term. A product is compatible with a standard but interoperable with other products that meet the same standard (or achieve interoperability through a broker).

2. Interoperability is the ability of two or more systems or components to exchange information and to use the information that has been exchanged — [IEEE 90].

3. Interoperability is: a. The ability of systems, units, or forces to provide services to and accept services from other systems, unitsor forces and to use the services so exchanged to enable them to operate effectively together. [JP1] b. The condition achieved among communications-electronics systems or items of communications-electronics equipment when information or services can be exchanged directly and satisfactorily between them and/or their users. The degree of interoperability should be defined when referring to specific cases — Federal Standard 1037C-Glossary of telecommunication terms.

4. Interoperability means the easy integration of products from multiple vendors without the need for custom hardware or software — Lonmark Association

5. Interoperability is the ability to use documents created with one DTD for a particular purpose within another environment. For example, any DTD that uses TEI Extended Pointers for linking creates documents which are interoperable, as regards linking, with TEI-encoded documents. This interoperability applies even if the DTD uses no tags in common with TEI at all, since the extended pointer mechanism is controlled by a particular use of attributes, not by specific element types. Panorama Pro actually supports this particular type of interoperability, by offering built-in support for TEI Extended Pointers — CIMI briefing paper on DTD interoperability.

6. Interoperability is defined as the ability of two or more systems or components to exchange and use information and the ability of systems to provide and receive services from other systems and to use the services so interchanged to enable them to operate effectively together — Police Information Technology Organisation.

7. Interoperability is the ability of computers on a network to fully share application software — Product Data Management Information Center glossary.

8. INTEROPERABILITY is the ability of two or more systems or components to exchange information and use the exchanged information without special effort on either system

9. INTEROPERABILITY is the ability of two or more systems or components to exchange and to use information, the ability to provide and receive services from other systems, and the ability to use the services so interchanged to enable them to operate effectively together.

Metadata Schema:

1. Australian Geodynamics Cooperative Research Centre (AGCRC) Metadata Schema
   Local schema based on existing system:
   
   • descriptive information of text archive metadata (keywords, geographic coverage, etc)
   • file information & processing instructions cf GIS data descriptions
   Also tracks national and international standards:
   • Dublin Core
   • ANZLIC: Geology information contains both geospatial and narrative elements Precision spatial encoding not always appropriate; IR from prose often required

2. Dublin Core Working Group:
   

   Schemas: The diversity of metadata needs on the Web requires an infrastructure that supports the coexistence of complementary, independently maintained metadata packages. The World Wide Web Consortium (W3C) has begun implementing an architecture for metadata for the Web. The Resource Description Framework, or RDF, is designed to support the many different metadata needs of vendors and information providers. The Dublin Core Metadata Initiative expects to support the infrastructure for registries provided by RDF Schemas.

3. Interface Data Repository (IDR)

   The IDR metadata is organized as a standard relational schema. At present we have identified six tables that need to be maintained: Bundles, Stores, Blocks, Block-Equivalency, Movement, and Movement-Associations. This section describes the content and maintenance of these tables in an informal form. For those familiar with RDBMS notation, the Appendix contains a formal relational schema.

4. Dublin Core Workshop Series

   The development of formal ontologies is currently a prominent line of research in digital library communities, aimed at identifying the structure of knowledge in a given discipline, and linking these structures into a larger whole. In contrast, one might think of this workshop series as an attempt to identify an “emergent ontology”, that is, a consensus among experienced practitioners across many disciplines about the basic elements of resource discovery.

5. (A Metadata schema is) information that assists the identification, discovery and transaction processes. (It) provide(s) clients with the opportunity to develop the knowledge that information exists (visibility or identification); assist clients to access or discover the required information; and assist client in the conduct of their business transactions with government using interoperable business systems.

6. NABIR

   (A metadata schema contains) guidelines to be used by Natural and Accelerated Bioremediation Research (NABIR) program investigators in managing their information and data. . . . Includes what these guidelines will and will not do. (E.g.) They will not specify how investigators will handle their data and analysis within their research projects. They will not specify how investigators will exchange data among co-workers (although voluntary agreement to some standards in this area will be strongly encouraged). The guidelines speak primarily to the format and documentation of data and information developed by the investigators that are to be transferred to the NABIR program, either for communicating research results or for basic long-term archiving and distribution of site characterization data to the larger user community.

7. A metadata schema provides an ontology aimed at identifying the structure of knowledge in a given discipline and linking these structures into a larger whole through the creation of a system of information that assists the identification, discovery and transaction processes of the given discipline structures.

8. Metadata Schema — A formal specification of the semantics and structure of a coherent collection of attributes that can be assigned in the description of a esource, as well as constraints that may apply to such descriptions.
   — Stu Weibel notes:
   
   • Formal in this definition means that it is maintained by an authoritative agency. It implies as well that the specification itself has a well-defined structure of fields, field labels, and permissable data types.

   • Semantics refers to the human-understandable meaning of the attributes.

   • Structure refers to the encoding characteristics of the value (is it a discreet value? a range? a ordered compound value . . . )

   • Coherent is used in the sense of an orderly relation of elements that are constituents of a logical whole

   • Attributes are fields or elements, which may in turn have substructure.

   • Constraints may include number and order of possible values, optionality, permissable data types, specification of collating sequence, etc.

9. Aimed at identifying the structure of knowledge in a given discipline and linking these structures into a larger whole, a metadata scheme assists users in the identification, discovery and use of information. A metadata scheme accomplishes this objective by providing a formal specification of the semantics and structure of a coherent collection of attributes that can be assigned in the description of information within a respective discipline.

10. Aimed at identifying the structure of knowledge in a given discipline and linking these structures into a larger whole, a metadata scheme provides a formal specification of the semantics and structure of a coherent collection of attributes that can be assigned in the description of information within a respective discipline. — Mark Watson