A Primer on Metadata Standards

1
A Primer on Metadata Standards

• From Dublin Core to IEEE LOM

Julia Innes Rory McGreal Toni Roberts TeleEducatio
n NB
2
I never met a meta metadata I ever really liked.
(Dorman, 1999)
Like any early inception of any standard, just
understanding the landscape is difficult.
(Luh, 1999)
3
What is METADATA?
data about data
Metadata standards are agreed-on criteria for
describing data to support interoperability
Example January 31, 2001 31 janvier
2001 2001-01-31 01-31-2000 31012000
4
What is METADATA?
Author Banathy,B.H. Year 1973 Title
Developing a Systems View of Education The
Systems-Model Approach Publisher Lear Siegleer,
Inc./Fearon Publishers Description A system
needs to be adaptive If it fails to deliver
expected outcomes 1. it adjusts 2. expectations
adjust 3. it terminates Call Number HX
5
Why METADATA?

• Facilitate information sharing
• Enable search engines on the Internet
• Support intelligent agents Push
• Minimises data loss
• Metadata describes learning objects

No one can sift 100 million docs Every day
6
Learning objects
any entities, digital or non-digital, which can
be used or referenced in technology-supported
learning

• Discoverable
• Modular
• Interoperable

7
Learning objects

1. Segment
2. Lesson
3. Topic
4. Course
5. Programme

granularity
8
Why learning objects?

• COST 1000s of colleges have common course
  topics
• large numbers of courses are going online
• World does not need 1000s of similar learning
  topics
• World needs only about a dozen
• Expensive to develop so sharing is essential
• (From Downes, 2000)

Design courses as a collection of learning
objects NOT HTML
9
Who inputs METADATA?

• Two Camps
• Internal referenced - Users input their own
  metadata
• External referenced Professionals input
  metadata

• number of electronic objects is growing rapidly
• metadata required is too much for third-party
  indexers

10
METADATA characteristics

• a data dictionary of commonly defined elements
• a method for manipulating and communicating
  elements electronically
• rules for identifying and extracting content
• an official standards body
• tools for creating, transmitting, and storing.
• Ahronheim (1998)

11
METADATA conditions

• Mandatory fields (small subset)
• Optional fields
• Extensible
• International interoperability
• Adapted from Griffin and Wason (1997)

12
METADATA challenges
Too much concern with FIELDS NOT enough with
TERMS

• Fields need a common terminology
• Described by a content expert BUT
• TERMS must fit into a universe of knowledge
• AND
• Not be only useful to content experts
• Cross-searching requires compatible
  vocabularies

13
What is RDF? (Resource Description Framework)
RDF is an infrastructure that enables the
encoding, exchange, and reuse of structured
metadata (Bearman et al., 1999)
RDF is syntax independent, and can be expressed
in both XML and HTML. -- World Wide Web
Consortium
14
What is RDF?

• a generalised format for online resources
• expresses all vocabularies with one model and
  syntax
• schema can work in XML

Warning RDF does not solve interoperability
problems with legacy metadata AND a variety of
RDF description schemas are possible
15
Start designing as a knowlege base not HTML see
autonomy.com Online Community Why? Improved
learning Sense of commitment Learning beyond
the content Reduced workload Administration
Content Interaction
16
What is XML? (eXtended Markup Language)
Standard General Markup Language

• Extends HTML without complexities of SGML
• XML is the underlying syntax for the transport
  of information for exchanging structured data

SGML
XML
HTML
17
What is XML?

• any level of complexity
• functions without the server
• vendor independent
• user extensible
• validation human readability

Warning possible Pandora's box of incompatible
metatags
18
Why XML?

• standardized
• uses schemas
• machine-readable
• two entities can use the same data

19
Metadata and RDF/XML
XML syntax
RDF structure
Metadata semantics resources
20
METADATA standards

• Dublin Core
• Warwick Framework
• IMS
• ARIADNE
• IEEE LOM
• AICC
• ADL SCORM
• Merlot?

21
Dublin Core
. . . the HTML of Web metadata (Bearman et al.,
1999) . . . lingua franca for metadata, . . .
at a basic level (Milstead Feldman, 1999) . .
. the most broadly based consensus on resource
description on the Web" (Weibel, 1999)
http//purl.oclc.org/dc/
22
Dublin Core

• coexists with other metadata sets
• all elements are optional
• all elements are syntax-independent
• tagged in HTML, raw XML, or RDF/XML

23
Dublin Core Fields

Title Creator Subject Description Publisher Co
ntributor Date Type Format Identifier Source
Language Relation Coverage Rights
All fields are optional, none are mandatory
24
Warwick Framework

• a higher-level context for the Dublin Core
• modularization of metadata
• facilitates interoperability
• permits selective access manipulation of data

WARNING It can create complexity that is not
needed. (Lagoze, 1996)
http//www.dlib.org/dlib/july96/lagoze/07lagoze.ht
ml
25
IMS
Educause Instructional Management System
MANDATE

1. catalyst for development of instructional
   software
2. creation of an online management infrastructure
   for learning
3. facilitation of collaborative learning activities
4. certification

Partners ARIADNE (Europe), NIST LOM,
Members Apple, Cisco,ETS, IBM, Indust.Canada,
Microsoft, Oracle, Sun, US Defense,etc.
26
IMS Metadata Schema
NOT just a metadata schema

• incorporates extends Dublin Core
• mandatory fields
• simple controlled vocabulary
• sets dictionary values
• reference schemas
• domain-specific taxonomies
• RDF/XML

http//imsproject.com
27
ARIADNE
Alliance of Remote Instructional Authoring and
Distribution Networks for Europe

• fosters the sharing and reuse of electronic
  pedagogical material, by universities and
  corporations.
• a Europe-wide repository for pedagogical
  documents (Knowledge Pool System)
• co-author of IMS Metadata structure

http//ariadne.unil.ch/
28
IEEE LOM P1484.12
Institute for Electrical and Electronics
Engineers Learning Object Management Protocol
MANDATE

• To enable learners or instructors to search,
  evaluate, acquire, and use Learning Objects
• focus on the minimal set of properties needed

http//ltsc.ieee.org/index.html
29
AICC
Aviation Industry CBT (Computer-Based Training)
Committee
Provides guidelines for interoperability for
systems to share data online

http//aicc.org/
AICC Guidelines Recommendations (AGR)
30
ADL SCORM
Sharable Courseware Object Reference Model
Advanced Distributed Learning Network
a set of interrelated technical specifications
built upon the work of the AICC, IMS and IEEE to
create one unified content model

http//www.adlnet.org/Scorm/
31
Merlot
Multimedia Educational Resource for Learning and
Online Teaching California State University
systems Distributed Learning and Teaching
Initiative Multimedia Repository Initiative

• a collection of high quality interactive online
  learning materials people
• an example of a learning object repository
• does not adhere to universal metadata standards

http//www.merlot.org
32
International Metadata Standard for Learning
Objects
IEEE LOM P1484.12
IMS/ARIADNE
Dublin Core
Expressed through RDFgt XML/SGML
33
POOL
Portal for Objects Oriented to Learning
a customizable, intelligent, learning object
repository

34
POOL
35
POOL Metadata Project
New Brunswick Distance Education Network
Inc. University of New Brunswick Electronic Text
Centre

• Specifications for applying IMS to POOL learning
  objects including
• A base schema schemas for describing video,
  audio, and still images at different levels of
  granularity
• Application of schema to the TeleCampus online
  course database

36
The End