Standards for Metadata

1. Organizational Framework of Digital Library Resources

2. Metadata Development and Application Framework

Basic Meaning of Metadata Metadata (metadata) is "data about data";

Metadata provides standardized and universal descriptive methods and retrieval tools for various forms of digitized information units and resource collections;

Metadata provides integration tools and links for distributed information systems (such as digital libraries) that are organically composed of multiple digitized resources. Metadata (metadata) is "data about data";

Metadata provides a standardized and universal description method and retrieval tool for various forms of digital information units and resource collections;

Metadata provides an integrated tool and link for the distributed information system organically constituted by a variety of digitized resources (e.g. digital library).

Leaving the metadata of the digital library will be a scattered sand, will not be able to provide effective retrieval and processing.

3. Metadata Application Environment

3.1 Application Purpose of Metadata

(1) Discovery andentification, which is mainly devoted to how to help people retrieve and confirm the needed resources, and the data elements are often limited to simple information such as authors, titles, subjects, and locations. Dublin Core is a typical representative.

(2) Cataloging (Cataloging), used for detailed and comprehensive description of the data unit, data elements include content, carrier, location and access, production and utilization methods, and even related data units, etc., and the number of data elements is often large, MARC, GILS and FGDC/CSDGM are typical representatives of this kind of Metadata. MARC, GILS and FGDC/CSDGM are typical representatives of this type of Metadata.

(3) resource management (Resource Administration), support for resource storage and use management, data elements in addition to a more comprehensive description of the description of the information, but also often includes rights management (Rights/Privacy Management), electronic signatures (Digital Signature), Resource Appraisal (Seal of Approval/Rating), Use Management (Access Management), Payment and Accounting (Payment and Accounting) and other aspects of the information.

(4) resource protection and long-term preservation (Preservation and Archiving), support for long-term preservation of resources, data elements in addition to the description and confirmation of resources, often including detailed format information, production information, protection conditions, conversion (Migration Methods), preservation of the responsibility and so on.

3.2 Application of Metadata in Different Domains According to the data characteristics and application needs of different domains, many Metadata formats have appeared in different domains since the 1990s

For example:

Web resources: Dublin Core, IAFA Template, CDF, Web Collections

Documentation: MARC (with 856 Field), Dublic Core

Humanities: TEI Header

Social science datasets: ICPSR SGML Codebook

Museums and artworks: CIMI, CDWA RLG REACH Element Set, VRA Core

Government Information: GILS

Geospatial Information: FGDC/CSDGM

Digital Images: MOA2 metadata, CDL metadata, Open Archives Format, VRA Core NISO/CLIR/RLG Technical Metadata for Images

Archives and Resource Collections: EAD

Technical Reporting: RFC 1807

Continuous Imagery: MPEG-7

3.3 Degree of Adoption of Metadata Formats

Different Metadata in different fields are at different stages of standardization:

In the area of network resource description, Dublin Core has become a widely accepted and applied de facto standard after years of international efforts;

In the area of governmental information, due to the strong promotion of the U.S. government and the implementation of relevant laws and standards, GILS has become a governmental information description standards, and has been applied to a considerable extent in several countries in the world, and similar to such a geospatial information processing FGDC/CSDGM;

But in some areas, due to the rapid development of technology changes, there are still multiple programs competing, typically Metadata for digital images, many of the proposed standards are in the experimental and improvement stage.

3.4 Metadata format "standardization"

Experience in Metadata development and application has shown that it is difficult to have a single Metadata format that meets the data description needs of all domains, and that even in the same domain, different, but interoperable, formats are needed for different purposes. Even within the same domain, different but interoperable Metadata formats may be needed for different purposes.

At the same time, a unified, centrally planned Metadata format standard is not suitable for the Internet environment, and is not conducive to fully utilizing the market mechanism and various forces.

But in the same area, should strive for "standardization", in different areas, should be properly resolved in different formats of the interoperability problem.

4. Metadata Structure

4.1 General Structure Definition A Metadata format is defined by a multi-level structure:

(1) Content Structure, which describes the constituent elements of the Metadata and their definitional criteria.

(2) Syntax Structure, which defines the structure of the Metadata and how it is described.

(3) Semantic Structure, which defines how Metadata elements are specifically described.

4.2 Content Structure

Content Structure defines the constituent elements of Metadata and may include: Descriptive Elements, Technical Elements, Administrative Elements, and Structural Elements (e.g., links to coding languages, Namespace, data units, etc.).

These data elements are likely to be selected based on certain criteria, and therefore need to be described in the metadata content structure, e.g., ISBD for MARC records, ISAD(G) for EAD, ICPSR Data Preparation Manual for ICPSR.

4.3 Syntactic structure

Syntactic structure defines the structure of a format and the way it is described, such as the organization of elements in sections, rules for selection and use of elements, methods for describing elements (e.g., ISO/IEC 11179 for Dublin Core), methods for describing the structure of an element (e.g., MARC records, SGML, XML), and structural statement descriptor languages (e.g., EBNF Notation), etc.

Sometimes the syntactic structure needs to indicate whether the metadata is bundled with the data object it describes, or exists as separate data but is linked to the data object in some form, and may also describe the way it is linked to definitional standards, DTD structures, Namespace, and so on.

4.4 Semantic structure Semantic structure defines how an element is described, such as the standards, best practices, or customized instructions used to describe the element.

Some metadata formats define the semantic structure themselves, while others specify the semantic structure by the specific adopter, e.g., Dublin Core suggests ISO 8601 for date elements, Dublin Core Types for resource types, MIME for data formats, and URLs or DOIs or ISBNs for identification numbers;

Another example is OhioLink's use of VRA Core, which requires the use of A&AT, TGM, and TGN for subject elements and ULAN for person elements.

5. Metadata Encoding Languages and Production Methods

5.1 Metadata Encoding Languages

Metadata Encoding Languages) refers to the specific syntax and semantic rules for defining and describing metadata elements and structures, often referred to as Definition Description Languages (DDLs).

In the early stages of metadata development, people often use custom record languages (such as MARC) or database record structures (such as ROADS, etc.), but with the increase in the number of metadata formats and the requirements of interoperability, people began to adopt a number of standardized DDLs to describe the metadata, such as SGML and XML, of which XML has the most potential.

5.2 Metadata Production Methods

(1) Specialized preparation of modules (e.g., for MARC, GILS, FGDC, etc.)

(2) Automatic preparation during data processing (e.g., for Dublin Core, etc.)

(3) Automatic preparation during physical processing of data (e.g., for some metadata parameters during digital image scanning)

(4) *** enjoyment of metadata (e.g., OCLC/CORC, IMESH

6. Metadata Interoperability

6.1 Metadata Interoperability Problems

Because of the fact that there are often multiple metadata formats in different domains (or even in the same domain), there is a problem of metadata interoperability when retrieval, resource description, and resource utilization is carried out among the resource systems described in different metadata formats.

Since there are often multiple metadata formats in different domains (or even in the same domain), when retrieval, resource description, and resource utilization are performed between resource systems described in different metadata formats, there is the problem of metadata interoperability:

Reading and interpreting multiple different metadata formats, conversion, and transparent retrieval of digital information resource systems described by multiple metadata formats.

6.2 Metadata Format Mapping

The conversion of different metadata meta-formats using specific conversion procedures is called Metadata Mapping/Crosswalking.

A large number of conversion programs exist for transforming between several popular metadata formats, such as

Dublin Core and USMARC; Dublin Core and EAD

Dublin Core and GILS; GILS and MARC TEI

Header and MARC FGDC vs. MARC

Multiple metadata formats under the same format framework can also be converted using a mediator format, e.g., the UNIverse project utilizes the GRS format for conversion of various MARC formats and other record formats. The format mapping conversion is accurate and the conversion efficiency is high. However, the efficiency of this approach in the face of multiple metadata formats coexist in the open environment is clearly limited.

6.3 Standard Description Framework

Another way to address metadata interoperability is to establish a standard resource description framework that describes all metadata formats, so that as long as a system can parse this standard description framework, it can decode the corresponding Metadata format. In fact, XML and RDF play a similar role from different perspectives.

XML solves the problem of interpreting different formats by allowing all systems that can interpret XML statements to recognize the Metadata format defined by the XML_DTD through its standard DTD definition.

RDF defines a basic model consisting of three types of objects: Resources, Properties, and Statements, where the relationship between Resources and Properties is analogous to the E-R model, and Statements describes the relationship specifically.

RDF establishes a framework for defining and using metadata through this abstract data model, where metadata elements can be viewed as properties of the resources they describe.

Furthermore, RDF defines a standard Schema that specifies mechanisms for declaring resource types, for declaring associated attributes and their semantics, and for defining relationships between attributes and other resources. In addition, RDF specifies mechanisms for invoking existing defined specifications using XML Namespace methods.

6.4 Digital Object Approach

The creation of digital objects containing metadata and mechanisms for their transformation may address metadata interoperability from another perspective.

The Cornell/FEDORA project proposes composite digital objects consisting of a kernel (Structural Kernel) and a function propagation layer (Disseminator Layer).

The kernel holds the content of a document in the form of a bit stream, the metadata describing the document, and the data related to access control of this document and metadata.

The function dissemination layer, the PrimitiveDisseminator, supports services related to the deconstruction of kernel data types and the reading of kernel data, and there may be Content-Type Disseminators, which may have embedded metadata format conversion mechanisms.

For example, a digital object with MARC-formatted metadata in its kernel has a Content-Type Disseminator loaded at the function propagation layer that requests the Dublin Core format and its conversion services. When a user of a digital object requests to read the metadata in Dublin Core, the corresponding content type propagator will request the digital object with Dublin Core and its conversion service program over the network, and then convert the MARC-formatted metadata in the requested digital object to the Dublin Core form before outputting it to the user.

7. A few suggestions

Track the development of metadata, actively participate in the development of metadata standards, accelerate the application of metadata, and pay attention to international convergence.

Speed up the research on mechanisms for effective use of metadata for retrieval (including transparent retrieval by heterogeneous systems), relevance learning, personalized processing, etc.

Accelerating research on ways and methods of organic integration of metadata with digital objects and digital resource systems.

Promote research on the use of metadata for knowledge-based data organization and knowledge discovery.