Catalog and Metadata Services: Strategic Vision Forum


Conclusions from Vision and Directions Task Force, and Some Notes on How we Got There Extensive review of white papers, reports, and professional literature (cf. annotated bibliography and quotations handout.) Conclusion 1: Engage skills, talents, and interests of all staff members. Organizational transparency-->better understanding of workflows-->improved ability to use judgement, innovate, and find efficiencies Invest in staff development Provide rich training opportunities (e.g., "the best training is by doing"--Indiana Task Group). No one is going to be left behind Give ourselves "permission to fail" "Many of life's failures are people who did not realize how close they were to success when they gave up." " I have not failed. I've just found 10,000 ways that won't work." --Thomas Edison Find ways to improve efficiency (e.g., Process improvement ideas) so that we have time to take on new projects Encourage innovation from staff at every level Conclusion 2: Optimize metadata for resource discovery and interoperability Faculty regard "easier electronic access to scholarly information" as University's most urgent IT need (Provost's Cyberinfrastructure Survey Report October 3, 2006) High-quality metadata necessary for success of Yale's digital library initiatives. E.g., "Long-term preservation of digital objects depends heavily upon the use of reliable, well-documented content formats and upon the consistent application of metadata. Likewise, the success of cross-collection searching (whether federated or harvested) also depends heavily on predictable metadata that adheres to national standards and well-documented practices" (DPIP Production and Content Integration Working Group, Aug. 2006 Report ) "Make data work harder" (Dempsey). E.g., classified browsing at NCSU and WNCLN Conclusion 3: MARC has great strengths and significant limitations as metadata standard MARC is extensible and highly structured. Makes online cooperative cataloging possible. But: Designed expressly for data exchange within and among library catalogs Playing well with other catalogs no longer enough. 21st century metadata must be able to play with diverse resources and services on the Web. Pure MARC is invisible to the Web ("dark matter" and "silo" images come to mind ) XML (eXtensible Markup Language) similar to MARC but uses word tags instead of numbers. Similar to HTML, but, like MARC, determines meaning and structure of data elements, not just formatting of Web pages. XML conversion/encoding supports "mashups" of cataloging data with other resources on the Web, e.g., Mashing up Amazon, LC catalog, social bookmarks, blogs, etc.: librarything.com Mashing up Amazon, WorldCat, Google, wikis, etc. in worldcat.org. Note permalinks (findable through Google, deeplinks back to Orbis; cf.: Team of Rivals) MARC-XML at Los Alamos National Laboratory,exploiting extensibiilty of MARC, i.e., large number of available well-defined data elements make MARCXML standard unusually robust, transparent, and extensible. "although there are well over one thousand tag/subfield combinations, only thirty-seven are used in 90% of MARC records". MARC is not dead. It has taken on new life in the Web environment. Those with deep knowledge of MARC have much to contribute. Conclusion 4: Orbis is one of many entry points into Yale's information universe. E.g., Orbis on its own terms and as information gateway: search, e.g., "XML for Dummies" Orbis seen as one database among many in Metalib Catalog data viewed through YaleInfo module (and Library News as RSS feed; see also Portal Opportunities Group) Requires metadata that can move across systems and platforms (portability and interoperablity) Conclusion 5: Convergence of tools needs metadata coordination. Yale Library supports many databases with unique platforms, e.g.: Beinecke Digital Library, MSS&A Finding Aids DB, Visual Resources Collection, Divinity Digital Library Scholars, teachers, students, and staff, want to to search Yale's collections comprehensively, regardless of originating database or department (see Conclusion 2) Solutions: Adopting or creating uniform standards; Providing centralized support for heterogeneous metadata Local standard: Y.E.S. (Yale Element Set) already in place. Crosswalks to reconcile different standards, e.g., Getty Metadata Crosswalks Institutional Digital Repository to preserve diverse metadata, bring together local standards, international standards, crosswalks, etc. Collections Collaborative, e.g., WW-I Experience (involving metadata element set to support diffuse special collection)

Conclusions from Vision and Directions Task Force, and Some Notes on How we Got There

Extensive review of white papers, reports, and professional literature (cf. annotated bibliography and quotations handout.)

Conclusion 1: Engage skills, talents, and interests of all staff members.

Organizational transparency-->better understanding of workflows-->improved ability to use judgement, innovate, and find efficiencies
Invest in staff development
- Provide rich training opportunities (e.g., "the best training is by doing"--Indiana Task Group). No one is going to be left behind
- Give ourselves "permission to fail"
"Many of life's failures are people who did not realize how close they were to success when they gave up." " I have not failed. I've just found 10,000 ways that won't work." --Thomas Edison
- Find ways to improve efficiency (e.g., Process improvement ideas) so that we have time to take on new projects
- Encourage innovation from staff at every level

Conclusion 2: Optimize metadata for resource discovery and interoperability

Faculty regard "easier electronic access to scholarly information" as University's most urgent IT need (Provost's Cyberinfrastructure Survey Report October 3, 2006)
High-quality metadata necessary for success of Yale's digital library initiatives. E.g.,

"Long-term preservation of digital objects depends heavily upon the use of reliable, well-documented content formats and upon the consistent application of metadata. Likewise, the success of cross-collection searching (whether federated or harvested) also depends heavily on predictable metadata that adheres to national standards and well-documented practices" (DPIP Production and Content Integration Working Group, Aug. 2006 Report )

"Make data work harder" (Dempsey). E.g., classified browsing at NCSU and WNCLN

Conclusion 3: MARC has great strengths and significant limitations as metadata standard

MARC is extensible and highly structured. Makes online cooperative cataloging possible.
But: Designed expressly for data exchange within and among library catalogs
Playing well with other catalogs no longer enough. 21st century metadata must be able to play with diverse resources and services on the Web.

Pure MARC is invisible to the Web ("dark matter" and "silo" images come to mind )

XML (eXtensible Markup Language) similar to MARC but uses word tags instead of numbers. Similar to HTML, but, like MARC, determines meaning and structure of data elements, not just formatting of Web pages.
XML conversion/encoding supports "mashups" of cataloging data with other resources on the Web, e.g.,
- Mashing up Amazon, LC catalog, social bookmarks, blogs, etc.: librarything.com
- Mashing up Amazon, WorldCat, Google, wikis, etc. in worldcat.org. Note permalinks (findable through Google, deeplinks back to Orbis; cf.: Team of Rivals)
MARC-XML at Los Alamos National Laboratory,exploiting extensibiilty of MARC, i.e., large number of available well-defined data elements make MARCXML standard unusually robust, transparent, and extensible.

"although there are well over one thousand tag/subfield combinations, only thirty-seven are used in 90% of MARC records".

MARC is not dead. It has taken on new life in the Web environment. Those with deep knowledge of MARC have much to contribute.

Conclusion 4: Orbis is one of many entry points into Yale's information universe. E.g.,

Orbis on its own terms and as information gateway: search, e.g., "XML for Dummies"
Orbis seen as one database among many in Metalib
Catalog data viewed through YaleInfo module (and Library News as RSS feed; see also Portal Opportunities Group)
Requires metadata that can move across systems and platforms (portability and interoperablity)

Conclusion 5: Convergence of tools needs metadata coordination.

Yale Library supports many databases with unique platforms, e.g.: Beinecke Digital Library, MSS&A Finding Aids DB, Visual Resources Collection, Divinity Digital Library
Scholars, teachers, students, and staff, want to to search Yale's collections comprehensively, regardless of originating database or department (see Conclusion 2)
Solutions: Adopting or creating uniform standards; Providing centralized support for heterogeneous metadata

Local standard: Y.E.S. (Yale Element Set) already in place.
Crosswalks to reconcile different standards, e.g., Getty Metadata Crosswalks
Institutional Digital Repository to preserve diverse metadata, bring together local standards, international standards, crosswalks, etc.
Collections Collaborative, e.g., WW-I Experience (involving metadata element set to support diffuse special collection)

This file last modified 03/01/07