Other Metadata Approaches

1
Other Metadata Approaches

• The basics of MODS, ONIX and TEI
• Georgina Lewis and Dan Napolitano

2
XML - eXtensible Mark-up Language

• Widely used for the definition of
  device-independent methods of storing and
  processing texts in electronic format.
• The interchange and communication format used by
  many applications on the WWW.
• A metalanguage
• Specifies how markup is to be distinguished from
  text, what markup is allowed and what the markup
  is.
• 3 distinguishing characteristics
• Emphasizes descriptive rather than procedural
  markup.
• Has a notion of documents as document types
• Independent of any one hardware or software
  system
• Extensible does not consist of a fixed set of
  tags.
• XML must be well-formed according to defined
  syntax
• One Root element enclosing the whole document
• Each element contained in root element or within
  other elements so contained
• Tag marks the start and end of each element.
• Can be validated against a schema of some kind
  (DTD, XML Schema, etc).
• Interested in the meaning of data not
  presentation.

3
ONIX (Online Information Exchange)

• Background
• Used primarily by publishers.
• Created by a consortium of groups AAP, EditEUR,
  BISG, and BIC. 1st public release in 2000. The
  current version is 2.1. In 2002 work began on
  ONIX for Serials.
• Much of ONIX is based on the pre-existing
  standard EPICS
• How it works
• Defines both a list of data elements about a book
  and a standardized container for how to send
  those elements as an ONIX message.
• The goal is increased sales. A product of the
  commercial world. With this in mind, designed to
  carry the rich information of the book jacket
  such as images, reviews, etc.
• Over 230 data elements, each with a standard
  definition. Most of the elements are text based,
  but multimedia files, such as images and audio
  files may also be embedded in an ONIX record.
• Crosswalks between ONIX and MARC21 and UNIMARC
• Tags may use either natural language, or a
  shorter mix of alpha numeric characters
• Users
• Amazon, Barnes and Noble, Library of Congress,
  Springer, Houghton Mifflin, Cambridge University
  Press
• Few small press involved at present, despite
  efforts to keep adoption costs low.

4
There is no I in TEI

• TEI History
• Established in 1987 with guidleines first
  published in 1994 (1300 pages with 600 elements)
• Sponsored by ACH, ACL and ALLC
• TEI Consortium set up in 2000 to maintain and
  develop the standard.
• TEI Basics
• Encoding scheme used primarily in the humanities.
  
• Scheme of choice for production of critical and
  scholarly editions of literary texts, including
  electronic texts among others.
• Interchangeable yet flexible with defined
  elements and attributes.
• No defined metalanguage. Originally supported by
  SGML - as of 11/2007 now fully XML.
• Describes how a text looks in current form, not
  how it should look, though can be used to create
  new texts.
• Example
• Kent
• Now by Apollo, king,
• Thou swear'st thy gods in vain.
• Lear
• O vassal! miscreant!
• Laying his hand on his sword.

5
MODS (Metadata Object Description Schema)

• Background
• Initially developed by the MARC 21 maintenance
  agency at the LC the Library of Congress
  Network Development and MARC Standards Office in
  2002. Current public release is 3.2
• Heavily indebted to MARC
• How it works
• A descriptive format designed to be used with
  METS
• Much richer than unqualified Dublin Core, whose
  ineffectiveness due to lack of interoperability
  and precision it aims to correct.
• Flexible yet consistent and predictable.
• 19 top level and 64 sub-elements.
• Combines some MARC fields, resulting in the need
  for fewer elements
• Designed as a MARC21 derivative, taking into
  account the great number of existing MARC 21
  records (over a billion), and the thousands of
  MARC trained librarians. It is thus very much
  tied to library history
• One of 3 types of records in XML format made
  available by the LC for Open Archives Iniative
  (OAI) for harvesting.
• More compatible with library data than ONIX

6
In conclusion . . . (the future of the future)

• XML is here to stay (for a while)
• MODS, TEI, and ONIX may be seen as 3 points of a
  triangle, representing the library world,
  academia, and the commercial world. As such they
  have different implementations, supporters, and
  success rates.
• MODS is proving difficult to implement, due to
  inertia, lack of widescale adoption, and
  contention over the extent to which MARC records
  may be pared down. Perhaps it is too encumbered
  by the history of MARC.

• ONIX is working, thanks to major adopters like
  Amazon and Barnes and Noble. There seems to be
  more of a push to make it work (and in a
  commercially viable and timely fashion). It is
  unburdened by history, and aided by the mindset
  of spending money to make money rather than
  saving money to save money.
• TEI continues to prove popular and successful
  with its user base. The transition from SGML
  syntax to full XML syntax in order to keep
  current with technology provides an excellent
  example of the flexibility of the encoding
  scheme. This flexibility, as well as the support
  from the TEI consortium, will secure TEIs place
  in the future of academia.