Abstracts
Advisory: Certain presentations are available only in the language of preparation and are provided for reference purposes in the context of the Canadian Metadata Forum 2005. Views expressed are those of the author and do not reflect the those of Library and Archvies Canada.
Pre-Forum Event:
Innovations in Metadata Research
Metadata Strategy at Library and Archives Canada
Presented by Deane Zeeman, Lead, Metadata Strategy Catalytic Initiative, Library and Archives Canada
Presentation [PDF 122 KB]
Metadata and Description: Findings and New Research Trajectories from the InterPARES2 Project
Presented by Dr. Joseph Tennis, Assistant Professor, School of Library, Archival and Information Studies, University of British Columbia
Presentation [PDF 66 KB]
Tuesday, September 27
Metadata Demystified
Karen Morgenroth, Data Management Coordinator, CISTI, National Research Council Canada.
The presentation will address the basics of metadata: defining key concepts and situating metadata in the arena of resource discovery and information management.
Presentation [PDF 357 KB]
Keynote Address
Michael Gorman, president of the American Library Association, Dean of Library Services at the Henry Madden Library, California State University, Fresno.
Death of Metadata
Jeffrey Beall, Auraria Library, University of Colorado at Denver.
In the mid-1990s, the rise of the World Wide Web highlighted the need to facilitate information discovery. Metadata became the hot topic at information management-related conferences and in professional literature; many information professionals expended enormous efforts creating and implementing new metadata schema. The resulting Tower of Babel of metadata schema has made the sharing metadata among professional communities increasingly difficult. Although abundant resources have been devoted to the development and implementation of Dublin Core, the schema has largely failed. The presentation will explore the reasons for this failure and propose alternatives.
Presentation [PDF 406 KB]
Metadata vs. Full-Text Search Engines
Denise Bedford, Thesaurus Manager and Senior Information Officer, World Bank Group.
Organizations are redefining the essential question about search. The question is no longer full text versus catalogue-type searching. The question is how and when you will implement semantic searching. Semantic searching combines aspects of both full text and catalogue searching, but it moves above the negative aspects of both. It leverages a metadata architecture base, deep conceptual indexing, programmatic methods of capturing metadata, works across and within languages, and promotes interoperability. Semantic search is also the foundation upon which to build an enterprise search architecture. Neither full text nor catalogue searching will scale to enterprise search.
Presentation [PDF 1,159 KB]
Exploiting Metadata
Linda Jackman, Consultant, Sierra Systems.
The investment of metadata in the development of a subject gateway, for access to resources on multiple cultural sites, will be examined. How this investment is exploited to drive a faceted taxonomy for accurate searches and guided navigation will also be explored.
Pierre Dumouchel, Vice-Président Scientifique, Centre de Recherche Informatique de Montréal (CRIM).
The presentation examines the MPEG-7 Audio-visual Document Indexation System (MADIS), a research project aimed at enabling content-based automated indexing of audio-visual documents within the MPEG-7 metadata standard. The project was initiated to help the National Film Board manage its vast collection of audio-visual material.
Controlled Vocabulary and "Folksonomies"
Louise Spiteri, Associate Professor, School of Information Management, Dalhousie University.
Social bookmarking sites allow end-users to add sites to their personal collection of links and to describe and organize the contents of these sites via their own vocabulary, or tags. This session will examine the major existing bookmarking sites, and how they facilitate the grassroots community classification of digital information. The benefits and limitations of user-derived tags, or folksonomies, will be discussed, as well their potential applications to existing controlled vocabularies and the creation of socially-developed taxonomies.
Presentation [PDF 426 KB]
Wednesday, September 28
Workshop A: Costing Models for Metadata
Pierre Dulude, Senior Standards Officer, Canadian Culture Online, Department of Canadian Heritage and David McCallum, IM Consultant.
Different organizations use different approaches when managing metadata during the development phase of a web site. In this workshop, participants will review factors that influence the cost of metadata creation based on the results a recent study produced by the workshop facilitators. The study led to the development of cost models that could assist workshop participants in planning metadata production and more importantly, in allotting adequate financial resources for the task. Based on the facilitators' experiences, additional issues facing managers planning the implementation of metadata will be discussed during the workshop.
Workshop B: Application Profiles
Marie-Claude Côté and Margaret Devey, Metadata Analysts, Information Management Strategies Division, Chief Information Officer Branch, Treasury Board of Canada, Secretariat.
Metadata application profiles are generally perceived to contribute to information exchange by identifying the metadata elements used in a particular domain or application, and by describing how they have been customized for a particular use. By making this information available, other applications, programs, agencies, and/or domains may use this information to exchange data. Other groups may also choose to re-use the metadata elements themselves. By contributing to information exchange on a technical level, as well as through the potential re-use of the metadata elements, metadata application profiles seem to promote interoperability.
The workshop is aimed at decision makers and practitioners who wish to understand the context and concepts related to application profiles, and the challenges they present. Participants will have an opportunity to share their experiences, issues, and solutions related to metadata application profiles.
Presentation [PDF 141 KB]
Workshop C: Categorization-In-The-Box: What's It All About?
Moderator: Linda Farmer, Consultant, Second Knowledge Solutions (K2S).
This introductory tutorial will begin with an examination of the technologies used for auto-categorization, including tool features, categorization techniques, processes of taxonomy building and categorization, and human/machine collaboration. A case study will be presented giving the real experience (both pros and cons) of deploying these tools. A panel discussion with vendors and users will round out the session.
Presentation [PDF 1,120 KB]
Convergence and Extensibility in the Metadata Domain
Facilitator: Greg Renaud, Senior Metadata Analyst, Treasury Board of Canada, Secretariat.
Convergence Panel
Panelist: MaryAnn Welke, Enterprise Metadata Coordinator, Office of the Corporate Technology Officer, Government of Ontario.
This presentation will provide an overview of efforts to align the Ontario metadata standards with national and international standards. Ontario has made a strong commitment to using Dublin Core through its Web Metadata and Common Metadata Elements Standard (CMES). The CMES defines common elements, refinements and encoding schemes, and provides a template for developing specialized metadata standards. These common elements will enable searching across all Ontario government information resources, and are aligned with national and international standards to enable sharing across jurisdictions. The presentation will discuss issues of common vocabularies, XML and common data standards.
Presentation [PDF 196 KB]
Panelist: Fay Turner, Senior Project Manager, Canada Business National Secretariat, Industry Canada.
The promise of metadata is becoming more elusive as more government jurisdictions and private sector organizations adopt metadata, in particular the Dublin Core. If metadata is to succeed in facilitating the discovery of Web resources, greater attention must be paid to solving issues of interoperability and standardization. The presentation will address questions of: agreement on rules, schemas, controlled vocabularies and guides for implementation; mechanisms for ensuring metadata quality and discuss the requirement for smart search. The challenge for all metadata practitioners and IM policy managers is to find solutions that involve pan-Canadian collaboration and strong leadership from all stakeholders.
Presentation [PDF 131 KB]
Panelist: John Roberts, Senior Archives Analyst, National Archives of New Zealand and chair, Dublin Core GOV Working Group.
This presentation concentrates on the relationship between the development of international standards (especially the Dublin Core) and local applications. The ongoing strategic review of New Zealand review of its government locator system (NZGLS) is focused on how to use the tool in an environment of proliferating sectoral schema, portals and federated search initiatives. The DC-GOV application profile may be a useful adjunct to ensure interoperation across jurisdictions.
Presentation [PDF 39 KB]
Extensibility Panel
Panelist: Mark Jordan, Head of Library Systems, Simon Fraser University.
The Canadian Association of Research Libraries's (CARL) institutional repository harvester and search service [http://carl-abrc-oai.lib.sfu.ca/] collects metadata from ten institutional and electronic theses repositories from across the country using the Open Archives Initiative Protocol for Metadata Harvesting (OAI-PMH) and provides a search interface to the aggregated records. So far, records have been harvested in unqualified Dublin Core. CARL is developing an application profile that will allow the metadata from these repositories to incorporate elements that will extend the functionality of the search service to better suit users' needs and to enhance access to the harvested records. This presentation will share findings of analyses of the unqualified Dublin Core metadata and describe activities to date in the development of the application profile.
Presentation [PDF 47 KB]
Panelist: Guy Aquin-St-Onge, Project Coordinator, Radio-Canada.
Between the standards of the company and the needs of the users, there are gaps to fill. Guy St-Onge will show how the New Media group at Radio-Canada draws on the expertise developed in the traditional sectors of the media to improve its contents to the Canadian public and to the craftsmen who produce the contents. Guy St-Onge will indicate the needs of the industry and the reasons why the standards are sometimes useful and why it seems we remain isolated on technological and semantic islands.
Presentation [PDF 386 KB]
Standards, It's a Jungle Out There!
Facilitator: James Turner, Associate Professor, École de la bibliothéconomie et des sciences de l'information (EBSI).
Interactive panel discussion addressing strategic advantages and challenges of using standards in a variety of domains. Questions submitted by Forum participants at registration will be the focus of the dialogue.
Closing Session
Pointing the Way: The CCRI's Research Infrastructure
Dr. Dale Anderson, Research Coordinator, Canadian Century Research Infrastructure (CCRI)
The Canadian Century Research Infrastructure is the largest, most complex research project in the social sciences and humanities in Canada. The project involves seven universities, partnerships with several public- and private-sector partners, and a time span of five years, all with the goal of creating a unique research infrastructure centred on microdata from Canadian censuses (1911, 1921, 1931, 1941 and 1951). A variety of metadata will accompany the microdata, thus combining both quantitative and qualitative materials. The result will be a new foundation for the study of social, economic, cultural, and political change in Canada in the first half of the 20th century. This presentation will outline the research infrastructure that is being created, and answer the question of why the CCRI is interested in metadata.
|
