2. Metadata Descriptions

• A metadata description uses property-value paired statements to clearly describe the characteristics of a particular thing.
  • Think of labels on food packages (see examples) or bottled drinks that inform consumers about the ingredients and nutritional elements they contain.
• In today’s Web-based environment, structured data describing the characteristics of things are exchanged and used in more flexible ways than a bounded record.

2.1 Embedded in a digital object

• Metadata descriptions embedded in WORD, PDF, iTune files
  Note: For non-HTML text documents like MS Word and PDF files, authoring tools such as Microsoft Word (for word-processing) and Adobe Acrobat (for PDF files) are equipped with templates to support embedding metadata within a file.
  In the following examples, find out what properties a template provides.
  • Open a WORD file from your computer, choose File --> Properties
    
  • Open a PDF file from your computer, or download this one NISO: Understanding Metadata, open from your computer, choose File --> Properties
  • Click on a file from your iTune playlist, choose File --> Get Info
    
    
• Metadata descriptions embedded in webpages
  Note: In many websites, metadata records are embedded in the source code of a webpage. Users usually will not see the metadata when they access and browse a website unless they choose to view the source code.
  Tips:
  . How to view 'source' from a browser: Chrome: View -->Developer-->View Source | Firefox: Tools-->Web Developer -->Page Source
  . How to find out if the page has embedded metadata without viewing the source code: Firefox: Tools-->Page Info
• IBM website| link to site --> view source | cached pdf file |
  
• Go through --> these questions based on the cached source code.
  --Note: the schemas used are Dublin Core and IBM's additional elements
  -- the encoding format is html
  -- we can consider them as multiple metadata 'statements', or still think them as contained in a 'record'.
• A research report | link to site --> view source | cached pdf file |
• California Institute of Technology (Caltech) CaltechAUTHORS
  • | link to a report's metadata record --> view source | compiled cached screenshots
    
    
• Metadata descriptions embedded in digital images
  Note: Some image software allows metadata records about an image to be recorded and attached in the image. When an image is viewed from the software application, it looks as if a record is embedded in the digital image. Values in some elements are automatically captured by the software while others are controlled by metadata creators.
  • An example from Adobe Photoshop CS: | cached: Metadata record; keywords; original image |
    From Adobe Bridge CS: |cached: metadata record |
    -- Note: IPTC Core & Extension = IPTC Photo Metadata, it is the standardized descriptive metadata schema for photos, developed by International Press Telecommunications Council (IPTC)

2.2 Metadata records displayed from databases

Note: Bibliographic databases, digital collections, and digital repositories store metadata records in databases and display the records with a more user-friendly interface. The following examples show how a record could be presented to users and how it is actually encoded or stored in a database.

• Library bibliographic catalogs
  
• Examples from the KentLINK OPAC:
  
  • |cached: Displayed bibliographic record and MARC record |
• Example from WorldCat:
  
  • |link to example ==> go to the end of the record and click 'Linked Data'.
    
  • |cached: display | cached: Linked Data in plain text (Turtle format) and in RDF/XML format
    
    
• Digital collections
• Examples from the *OhioLINK Digital Media Database
  
  • |cached: Image and Full metadata record (2009) |
• Examples from Ohio Memory
  
  • |link to example | cached: Full metadata record and image (2015) |
    
    
• Examples from the Worthington Memory collection, (Worthington, Columbus, Ohio):
  
  • |cached: Search results; Expanded record; Dublin Core record (2006) |
    
  • User-friendly metadata interpretation (2015) |
    
    
• An Example from the GREEN (Green's Functions Digital Library) project, Kent State University
  
  • |link: cached pdf file |
    
    
• Example from the *Alexandria Digital Library (ADL), UC Santa Barbara
  
  • Display records at ADL-- cached PDF files: Standard report | Standard XML
    
  • Records stored at the National Science Digital Library (NSDL) Metadata Repository (see next)
    
    
• Digital repositories
  
  • Examples from the National Science Digital Library (NSDL)
    --Note: This example shows how the encoded record is converted to a different format according to the NSDL metadata structure.
    • The same ADL record displayed at the NSDL-- cached PDF file: metadata record
      
      
  • Examples from American Memory, Historical Collections from the National Digital Library
    
    • | An image thumbnail and full metadata record | (cached 2006)
    • |A set of images, Display record | MARC record | (cached 2015)
      
      • To see other examples in the American Memory collection:
        http://memory.loc.gov/ammem/index.html ==> search or browse
        
        
  • Examples from Europeana, the European digital library
    • | search at http://www.europeana.eu/, filter by sound | A sound recording | (cached 2015)
      
      
• Aggregated Data sets
  • DBpedia
    
    • | Link to "Plato" page http://dbpedia.org/page/Plato | cached (2015)
      Note: follow the linked object such as other persons influenced by Plato, his birth place, etc. to see more.
      
      
  • Google's Knowledge Graph
    
    • | Link to search Mona Lisa | cached 2015
      Note: pay attention to the right side of the screen, follow more linked objects to find more.
      
      
  • Wikidata
    
    • | Link to "Steve Jobs" page https://www.wikidata.org/wiki/Q19837 | cached 2015-08

2.3 Sample records provided by metadata standards or guidelines

Note: Metadata standards usually provide the best practice guides in their specifications. The examples provided focus on the contents rather than the formats. It ensures that metadata elements, structure, and value spaces specified by a standard are correctly understood and controlled when applied to real situations. It is not their intention to provide format guidelines as to how a record should be encoded or displayed.

• Dublin Core
  
  • DC-TEXT Representation of Examples
    *https://www.dublincore.org/specifications/dublin-core/dc-text/2007-12-03/
  • Expressing Dublin Core™ metadata using HTML/XHTML meta and link elements
    *https://www.dublincore.org/specifications/dublin-core/dc-html/
  • RDF/XML Representation of Examples
    *https://www.dublincore.org/specifications/dublin-core/dcq-rdf-xml/
  • Side-by-side example of DC-text representation and RDF graphs
    *https://www.dublincore.org/specifications/dublin-core/dc-rdf/
    
    
• CDWA (Categories for the Description of Works of Art), Getty
  http://www.getty.edu/research/publications/electronic_publications/cdwa/examples.html
  • See over 100 examples under "Catalog Examples".
    
    
• VRA Core 3.0 (Visual Resource Association (VRA) Core Categories)
  http://www.loc.gov/standards/vracore/VRACore3_Examples.pdf
  
  • Pay attention to the data sets in Example 3 that describe a chair documented by a photograph. The photograph was later copied to a slide format and scanned to create a digital image.
    
    
• VRA Core 4.0 (Visual Resource Association (VRA) Core Categories)
  http://core.vraweb.org/vracore_examples.html
  • Metadata records for various types of works are provided. For each work, there is a html display record, a full VRA record, and a record encoded with XML.
    
    
• Cataloguing Culture Objects (CCO)
  *http://vraweb.org/resources/cataloging-cultural-objects/
  • Cataloguing examples can be found in each chapter.
    PDF of the whole book can be downloaded.
    
    
• Encoded Archival Description (EAD) version 2002
  • Encoded examples: http://www.loc.gov/ead/tglib/appendix_c.html
    
    

2.4 Metadata descriptions embedded in a webpage's main <body> section

Note: This is an interactive demo, showing how the webpage's content can be recognized by search engines after embedding semantic metadata.

1. Open this file from a text editor or 'view' source' from a browser. (This is the original source code without embedding metadata in the <body> section.) File to open: Code-Before
   • copy-paste into the Testing Tool website and see what happens. |cached screenshot |
   • copy-paste into the http://rdfa.info/play/ website and see what happens. | cached screenshot |
     
     
2. Now open this page which added metadata tags. File to open: Code-After
   • copy-paste into the Google Testing Tool website again and see what happens.|cached screenshot |
     -- the schemas used is schema.org, a search engines-developed collection of many metadata schemas.
   • copy-paste into the http://rdfa.info/play/ website again and see what happens. |cached graph view | raw data |

All these can use different syntaxes: HTML5' Microdata, or RDFa

2.5 Use a template to create a metadata record and view various output formats

To practice on creating a machine-readable metadata description, check Section 5.

<-- Back to Table of Contents |||| Go to Next Section -->