Metadata Encoding Transmission Standard Overview

Metadata Encoding Transmission Standard Overview Elizabeth Rogers Metadata Encoding Transmission Standard History The origins of the Metadata Encoding Transmission Standard, or METS, can be traced back to a 1996 attempt by The University of California Berkley to address the institutions inability to rebind pages of a book once they had been digitized. The program created by UC Berkley, called Ebind, was not successful. However, it paved the way for another attempt at solving this problem, the Making of America II Project. This project, started in 1998, was taken on by several U.S. universities, including UC Berkley (McDonough, 2005). While MOA2 was a step in the right direction towards the organization of digital objects, it was discovered that MOA2 had limited ability to fulfill this role. In 2001, a group of libraries working on digital library development programs decided that a replacement for MOA2 was needed, which led to the development of METS (McDonough, 2005). Metadata Encoding Transmission Standard Importance and Significance When an institution creates metadata for a resource, particularly a book, the metadata can be used to aide users in finding the book, and helps the library keep an accurate account of its collection and holdings. However, if the library fails to create accurate structural metadata, that does not mean the resource is lost or that the pages of the book will be forever separated. The same cannot be said of books once they are digitized. When a book is digitized, each page becomes a separate resource, and before the creation of METS, there was no encoding standard that provided a platform to create the structural metadata necessary to digitally bind these resources to ensure that they would be findable and able to used and evaluated as a cohesive unit (METS: An Overview Tutorial, 2016). Brad Westbrook, a librarian at the University of California in San Diego, describes METS as an XML standard that is a type of digital wrapper. It functions to relate the components of a digital resource (Rose, 2005). METS was specifically created for the digital library community to allow for the digitization and encoding of complex digital objects, like books or presentations. These resources can contain a variety of parts as well as different types of files. For example, one presentation can contain text files, images, video, and sound files. Using the structural metadata elements included in METS, institutions can ensure that all components of a resource are linked, even if they are stored in different places. METS also enables institutions to use structural metadata to control the presentation of resources and ensure that the objects are presented in the way they were intended to be (Rose, 2005). As previously stated, METS was born out of MOA2. METS did not replace MOA2, but rather built upon the work that had already been done. One of the major shortcomings of MOA2 was its lack of flexibility at the local level with administrative and descriptive metadata elements. METS allows for flexibility at the local level with administrative and descriptive metadata, as it does not require either of these to be included in an objects METS document. If descriptive or administrative metadata are used, METS does not require the use of controlled vocabularies for many elements, and allows for the use of whatever metadata element set the record creator chooses, furthering its increased flexibility over MOA2 (McDonough, 2005). Additionally, MOA2 was limited by its ability to encode only texts and still image media. MOA2 was unable to encode audio or video resources. Even in 2001, this would pose a significant challenge to any library with a well-developed collection that was serious about di gitizing all of its resources (McDonough, 2005). METS gives institutions and repositories the ability to encode audio and video resources, in addition to print objects. Finally, METS was created to allow for improved sharing of digital objects between repositories, which MOA2 could not facilitate (McDonough, 2005). Metadata Encoding Transmission Standard Schema Description METS documents are created using XML, so that the document is machine readable. A METS document can include up to seven sections, METS header, descriptive metadata, administrative metadata, file inventory, structural map, structural links, and a behaviors section. The only required, and most important, section for a METS document is the structural map. The structural map defines a hierarchical structure for a digital object, this section is where the relationship between the digital objects files is described. The information found in this section is what allows users to more easily look through a digital object, much in the way a user would look through a physical book. The structural map can also links the objects digital files back to their descriptive and administrative metadata (McDonough, 2005). The structural map is a unique aspect of this scheme because it can be represented by an actual diagram that illustrates the relationships between the parts of an object and the objects metadata. Another unique section of the METS document is the structural links section. This section is generally used in the archiving of websites. It allows the document creator to record hyperlinks between items in the structural map. A METS structural map can show the page hierarchy of a website, showing the relationship between a parent page and subsequent child pages underneath it. The structural links section allows for the recording of links between the child pages that would not be displayed in a traditional hierarchical organizational structure (METS: An Overview Tutorial, 2016). The behaviors section of a METS document is used to record behavioral metadata. This section records any metadata related to software or applications that may be needed to view, or use, a digital object. The behaviors section enables institutions to exercise control over how users experience a digital object. However, this section can also create significant challenges for repositories. Software and applicat ions change consistently and often, rapidly. If a behavior changes, a repository manager would need to modify the record for every object associated with this behavior (McDonough, 2005). The seven required sections of a METS document are also some of the top level elements used in METS. Other elements unique to METS include, structural requirements, technical requirements, maintenance agency, behavior files, and description rules. This is a departure from other schemas that tend to include top level elements that lend themselves solely towards descriptive metadata. In METS, these traditional descriptive elements are found within the descriptive metadata element. Here, elements from Dublin Core, MARC, MODS, EAD and VRA can be wrapped inside METS sub elements to describe the digital work (METS: An Overview Tutorial, 2016). This distinction is important because it reinforces that while METS does allow for the inclusion of descriptive metadata, its focus is on the administrative and structural metadata that is necessary to maintain the objects original structure and presentation. Metadata Encoding Transmission Standard Resources METS Schema Documentation. (2016, August 9). Retrieved March 12, 2017, from http://www.loc.gov/standards/mets/mets-schemadocs.html Schema Documentation. (2011, July 1). Retrieved March 12, 2017, from https://www.loc.gov/standards/mets/profile_docs/mets.profile.v1-2.html METS: An Overview Tutorial. (2016, February 9). Retrieved March 12, 2017, from https://www.loc.gov/standards/mets/METSOverview.v2.html#structlink Rose, Trish. (Summer 2005). METS: A Data Standard for Access and Preservation Now and Into the Future. Digital Letters, 8, 1-4. McDonough, Jerome. (2006, February 1). METS: standardized encoding for digital library objects. International Journal on Digital Libraries, 148-158. Metadata Encoding Transmission Standard Example Alabama blues Lomax, Alan 1915-2002 Recordist Hurston, Zora Neale Recordist Barnicle, Mary Elizabeth 1891-1978 Recordist Metadata Encoding Transmission Standard Conclusion METS was created in response to a void in the Library and Information Science community, with respect to archiving digital objects. Repositories that curate digital objects are tasked with organizing and disseminating a number of resources that can far exceed the collections of institutions with only physical objects in their collections. Once objects become digitized, they also present the unique challenge of taking on characteristics they did not possess as physical objects. In addition to their physical characteristics, these objects now have digital characteristics as well. Ensuring the that the integrity of the objects remain intact is important, if the objects are going to best serve patrons and users. The encoding scheme provided by METS enables institutions to organize and display vast collections of digital objects, while maintaining the objects integrity. METS achieved the flexibility that institutions felt MOA2 was lacking. However, one of the challenges presented by increased flexibility can be decreased interoperability. The lack of controlled vocabularies and required schema elements in METS makes it more difficult for individual repositories to share digital objects that have been encoded using METS. The future of METS will be focused on overcoming this challenge, and working towards interoperability between repositories. Jerome McDonough suggests that the creation of METS profiles by institutions is a step that can be taken on the road towards interoperability. In a METS profile document, institutions can detail restrictions on, and guidelines for, creating METS documents. Institutions can include directions about the schema and controlled vocabularies that should be used in the creation of METS documents. Additionally, McDonough suggests that a METS profile could contain guidelines for the forms that should be used for digital objects. This way, repositories could easily communicate with each other regarding the forms of objects that can accept and give. (McDonough, 2006). The LOCs METS website already has information on developing a METS profile for a digital object and describes the requirements for a complete profile. The requirements laid out by the LOC for a METS profile include information about an objects title and creation date, contact information, related profiles, profile context, external schema, rules of description, controlled vocabularies, structural requirements, technical requirements, tools and applications, and examples. (METS Profile Components, 2011). As more institutions start to adopt this profile format, they will be able to move towards increased sharing of metadata and records. Categories for the Description of Works of Art History Cataloging non-print items has always proved challenging for the Library and Information Science community. The uniqueness of the objects held by galleries and museums makes standardization and interoperability difficult. As a possible answer to this conundrum, Categories for the Description of Works of Art, or CDWA was developed. CDWA can trace its roots back to the early 1990s when it was created by the Art Information Task Force, also known as the AITF. This task force was comprised of art historians, museum curators and registrars, visual resource professionals, art librarians, information managers, and technical specialists. CDWA is the basis for CDWA Lite, an XML schema used to describe works of art that was developed out of CDWA (Categories for the Description of Works of Art: Introduction, 2015). Categories for the Description of Work of Art Importance and Significance The art documentation and museum communities realized that developing a data structure standard for the explicit purpose of describing art, architecture, and material culture was a necessity (Baca, 2007). As the community was developing and changing, so was its need to describe its collections and holdings. Traditional data structure standards and schema, such as MARC, were primarily intended to describe textual works, as evidenced by the data element sets that include elements such as creator and publisher. These elements simply dont apply to visual works of art. In contrast the CDWA includes 532 categories and subcategories directly related to visual works of art (Baca, 2007). It was important for this community to develop a schema that had a wide variety of categories because repositories, like art museums that hold a wide variety of objects. Art objects can include textiles, paintings, pottery, sculpture, and works of architecture just to name a few. With the development of CDWA this community finally had the data structure standard it needed. However, CDWA is not able to be expressed in a machine-readable form. To be able to make this data machine readable, and thus more sharable, another standard needed to be developed by this community. In response to this need, CDWA Lite was created. Now, the art documentation and museum communities had an XML schema that was based off CDWA. Though this schema does not contain as many categories and subcategories and CDWA, CDWA Lite still achieves great depth and flexibility with over 300 elements and sub elements (Baca, 2007). Categories for the Description of Works of Art Schema Description The Getty Institute describes the purpose of CDWA on its website stating, The Categories for the Description of Works of Art (CDWA) are a set of guidelines for best practice in cataloging and describing works of art, architecture, other material culture, groups and collections of works, and related images, arranged in a conceptual framework that may be used for designing databases and accessing information (Categories for the Description of Works of Art: Introduction, 2015). The category/subcategory sets for CDWA are vast and allow for detailed and accurate descriptions of these types of collections and holdings. CDWA has several core categories that are strongly recommended to be included in the description of a work using this schema. While the IATF states that they feel the core categories represent the minimum information necessary to uniquely and unambiguously identify and describe a particular work of art or architecture, they concede that ultimately which core categories to in clude should depend on a particular institutions purpose and users (CDWA List of Categories and Definitions, 2014). The core categories in CDWA are object/work, classification, title or names, creation, measurements, materials and techniques, subject matter, current location, related textual references, person/corporate body authority, place/location authority, generic concept authority, and subject authority (Categories for the Description of Works of Art: Categories, 2014). These categories clearly lend themselves to describing works of art and differentiate this schema from others like Dublin Core or MARC. In contrast, CDWA Lite requires fewer elements, presenting instead a core description of the object (Baca, 2007). Per its website Getty explains that the purpose of CDWA Lite is to describe a format for core records for works of art and material culture, based on the data elements and guidelines contained in the Categories for the Description of Works of Art (CDWA) and Cataloging Cultural Objects (CCO) (CDWA Lite: Specification for an XML Schema for Contributing Records via the OAI Harvesting Protocol, 2006). Records created with CDWA Lite are less detailed that those created using CDWA intentionally, so that they represent the essence of the work. Murtha Baca of the Getty Research institute says The goal of the CDWA Lite schema is to provide core descriptive metadata about cultural works-i.e., an essential metadata record that can be easily shared and contributed to union resources and that provides enough information to enable users to understand what the work is and what instituti on owns it (Baca, 2007). The core categories recommended by the IATF for CDWA are the basis for the twenty-two high level elements in the CDWA Lite schema. In this schema elements 1-19 deal with descriptive metadata and elements 20-22 deal with administrative metadata. (CDWA Lite: Specification for an XML Schema for Contributing Records via the OAI Harvesting Protocol, 2006). Within CDWA Lite, only nine of the twenty-two top level elements are required. These elements are, object/work type wrapper, title wrapper, display creator, indexing creator wrapper, display materials/techniques, indexing dates wrapper, location/repository wrapper, and record wrapper (CDWA Lite: Specification for an XML Schema for Contributing Records via the OAI Harvesting Protocol, 2006). One of most unique features of CDWA and CDWA Lite is that within the required categories and elements, there are recommended sub elements and sub categories. For example, within the creation category for CDWA the following sub categories are also strongly recommended, creator description, creator identity, creator role, creation date, earliest date, and latest date. (CDWA List of Categories and Definitions, 2014). There are considerably more recommended categories and subcategories required by CDWA than other metadata schemas I have encountered thus far. The inclusion of such a wide variety of elements and sub elements is essential when an institution holds objects in its collection that all possess distinct characteristics. Another interesting aspect of this schema is the harvesting ability of CDWA Lite. CDWA Lite records foster interoperability between museums and other institutions that use the Open Archives Initiative, or OAI by allowing repositories to harvest metadata from one another (CDWA Lite: Specification for an XML Schema for Contributing Records via the OAI Harvesting Protocol, 2006). According to Murtha Baca of the Getty Research Institute this is an advantage because the metadata comes from (or should come from) the institution that owns the corresponding objects or items, and is therefore accurate and authoritative (Baca, 2007). As was mentioned earlier, an XML record that is created using CDWA Lite represents only the most core information about the work. This was done to make the harvesting of metadata using this schema achievable and manageable (Baca, 2007). Categories for the Description of Works of Art Resources Categories for the Description of Works of Art: Introduction. (2015, October 6). Retrieved March 20, 2017, from http://www.getty.edu/research/publications/electronic_publications/cdwa/introduction.html Categories for the Description of Works of Art: Categories. (2014, April 9). Retrieved March 19, 2017, from http://www.getty.edu/research/publications/electronic_publications/cdwa/categories.html Categories for the Description of Works of Art (CDWA). (2012, August 9). Retrieved March 23, 2017, from http://www2.archivists.org/groups/standards-committee/categories-for-the-description-of-works-of-art-cdwa CDWA Lite: Specification for an XML Schema for Contributing Records via the OAI Harvesting Protocol. (2006, July 17). Retrieved March 22, 2017, from http://www.getty.edu/research/publications/electronic_publications/cdwa/cdwalite.pdf Baca, Murtha. (Spring 2007). CCO and CDW Lite: Complementary Data Content and Data Format Standards for Art and Material Culture Information. VRA Bulltein, Vol. 34, Number 1, 1-8. Categories for the Description of Works of Art Example 12345 http://www.getty.edu/art/collections/objects/o1091.html 12346 http://www.getty.edu/art/collections/objects/o1092.html 98077 full view, oblique view from the right front corner general view oblique view 1996 98076 detail of the desktop with inlaid coat of arms overhead view detail view coat of arms Metadata Encoding Transmission Standard Conclusion Museums and art institutions recognized the need for a cataloging system that enabled them to accurately represent the works held in their collections which led to the development of CDWA. As more institutions and repositories began to develop digital collections, a new need also arose. Institutions that previously had distinct collections found themselves having the same digitized copies of works in their online repositories. Interoperability, which wasnt as a great of a need before now needed to be realized. It was this realization that led CDWA Lite. Both CDWA and CDWA Lite have element and sub element sets with a great amount of depth that are specific to describing works of art. Within CDWA this allows for the development of rich and accurate records. Within both CDWA and CDWA Lite the element and sub element sets allow institutions to use one system to catalog a variety of works that have dissimilar properties. This need sets this community apart from the library and archival communities which mainly house textual objects. These schemas rely heavily on descriptive metadata, and most of the elements in the CDWA and CDWA Lite element sets fall under the descriptive metadata umbrella. Through CDWA Lite institutions are also able to harvest metadata from each other, enabling museums and art repositories to create accurate and up to date records for digitized works. CDWA and CDWA Lite are still relatively young schemas, CDWA just reached is barely twenty years old. Considering the future of the schemas, one change that may come for CDWA Lite is the inclusion of more of the elements and sub elements from CDWA (CDWA Lite: Specification for an XML Schema for Contributing Records via the OAI Harvesting Protocol, 2006). Adding more elements to CDWA Lite would enable museums to make the metadata for more of the objects in their collections harvestable and would increase the quality of the harvested metadata. It will be exciting to see how this schema develops and continues to improve standardized cataloging and interoperability in the museums and art repository communities. Works Cited: Baca, Murtha. (Spring 2007). CCO and CDW Lite: Complementary Data Content and Data Format Standards for Art and Material Culture Information. VRA Bulltein, Vol. 34, Number 1, 1-8. Categories for the Description of Works of Art: Introduction. (2015, October 6). Retrieved March 20, 2017, from http://www.getty.edu/research/publications/electronic_publications/cdwa/introduction.html Categories for the Description of Works of Art: Categories. (2014, April 9). Retrieved March 19, 2017, from http://www.getty.edu/research/publications/electronic_publications/cdwa/categories.html Categories for the Description of Works of Art (CDWA). (2012, August 9). Retrieved March 23, 2017, from http://www2.archivists.org/groups/standards-committee/categories-for-the-description-of-works-of-art-cdwa CDWA List of Categories and Definitions. (2014). Retrievied March 23, 2017 from https://getty.edu/research/publications/electronic_publications/cdwa/definitions.pdf CDWA Lite: Specification for an XML Schema for Contributing Records via the OAI Harvesting Protocol. (2006, July 17). Retrieved March 22, 2017, from http://www.getty.edu/research/publications/electronic_publications/cdwa/cdwalite.pdf McDonough, Jerome. (2006, February 1). METS: standardized encoding for digital library objects. International Journal on Digital Libraries, 148-158. METS Example Documents. (2016, February 9). Retrieved March 14, 2017, from https://www.loc.gov/standards/mets/mets-examples.html METS: An Overview Tutorial. (2016, February 9). Retrieved March 12, 2017, from https://www.loc.gov/standards/mets/METSOverview.v2.html#structlink METS Schema Documentation. (2016, August 9). Retrieved March 12, 2017, from http://www.loc.gov/standards/mets/mets-schemadocs.html Rose, Trish. (Summer 2005). METS: A Data Standard for Access and Preservation Now and Into the Future. Digital Letters, 8, 1-4. Schema Documentation. (2011, July 1). Retrieved March 12, 2017, from https://www.loc.gov/standards/mets/profile_docs/mets.profile.v1-2.html