2012 Annual Meeting Minutes

The CEAL Committee on Public Services and Committee on Technical Processing Joint Program

Discovery, Delivery, and Dissemination of Information in a Multilingual and Multicultural Environment

Wednesday, March 14
1:20-4:30 PM
(Sheraton Centre Toronto Hotel, Dominion Ballroom South)

• 1:20-1:25 p.m.   Introduction of the joint program by Jidong Yang and Shi Deng

Jidong welcomed audience and introduced the contents of the program, as well as the CPS members. Shi introduced the CTP members and introduced guest speakers for the program.
[Click for presentation slides]



• 1:25-2:15 p.m. SESSION ONE: Introduction to the Semantic Web and Library Linked Data
  Guest speakers: Marlene van Ballegooie, Metadata Librarian and Alastair Boyd, Head of Cataloguing at the University of Toronto Libraries

[Click for presentation slides]

Marlene started with introduction on where libraries are heading in the future. To respond to the user needs and user preference to search engines as an information platform over the library, library bibliographic data needs to move from the closed database model to the open web-based model. She mentioned two LC reports, Working Group on the future of bibliographic control (Jan. 2009) and A Bibliographic Framework for the Digital Age (Oct. 2011) that shows LC's leadership in moving libraries into the library-technological environment where the bibliographic control will be based on linked data principles and RDF basic data model.

She then introduced four principles of linked data and the Resource Description Framework (RDF), the semantic web standard. She called the RDF the building block of linked data and illustrated that RDF uses the concept of triples to describe a relationship between two things. The combination of these triples creates RDF graphs that are the formation of a linked open data cloud. She followed up by showing several library linked data examples.
In her final section, she discussed about linked data and RDA. She used an example to illustrate what a bibliographic description record would be look like to catalogers from MARC to linked data based on RDA data elements. To conclude her presentation, she believed that Libraries can make an important contribution to the Semantic Web. Because we are data experts with a wealth of experience managing, preserving, describing and delivering data to users. The fundamental challenge for the development of linked data is lack of awareness within libraries. Are we ready to take the challenge?

Following Marlene's presentation, Alastair discussed the aspects of library technical services and linked data as the following:

• There are two aspects of Linked Data of particular interest from my perspective as Head of a Cataloguing Department at a large academic library
• I'm caught between the fascinating speculations & predictions of "Library Futurists" and the mundane requirements of short- and medium-term planning decisions; predicting the future in detail is almost impossible
• Marlene's point that we librarians are the experts is my starting point: bibliographic Control is all about establishing links in databases
• In the past the focus has been local; in the future we need to share actual records, not just all make copies of the same records
• "Library Services Platforms" are the "next generation" replacement for ILS's, remotely hosted, designed to share not only records but Services and Apps between libraries, covering the traditional ILS modules (Acquisitions, Cataloguing, etc.);
• Today at UofT we have a complicated patchwork (Sirsi ILS, Endeca Discovery layer, Serials solutions e-resource manager plus Summon) plus uploads to WorldCat, plus a local file of authority records)
• All these have to programmed to talk to each other, and to stay synchronized; this requires a lot of work
• Library Service Platforms hold out the promise of seamless integrated systems that include e-resources as part of the main stream
• One other point about linking us to the wider world: we can't catalogue the web, but we can use our catalogues / discovery layers to provide a front end to the web: the concept of "extend your search" links from records out to Google
• e.g. my proposed "Petrucci Project" links for digitized music scores: as an alternative to cataloguing all 200,000 online scores in the set, links in our catalogue could send precise tightly focused Google queries directly to the site. This is crude compared to the possibilities just outlined by Marlene, but can be done now, even without using RDF framework

Q&A:

Q. Question on technology about linked Data: what are the copyright implications?

A. It is difficult to be sure. Any time we open up our data to the outside world we lose a lot of control over how it might be used. When it comes to links that we make ourselves from our databases out to other places, we must ensure that the target is legal, or that the target itself handles the copyright question. For example, if we implemented A. Boyd's idea of linking to the IMSLP "Petrucci Project" repository of free public domain music scores, that site itself carries the copyright disclaimer concerning the different definitions of public domain that apply in different places.

Q. What about an owner of data doesn't want to be linked free, to follow up on that topic?

A. Having a copyright disclaimer attached to the data will help users understand how they can and cannot use the data.

Q. A few national libraries have already started exposing their data as linked data. What is UofT planning? How can individual libraries prepare?

A. At the University of Toronto Libraries we want to learn more about it and experiment with it as a research project. For other libraries, probably the best thing that they can do to prepare is to do some research and learn about the principles and practice of linked data. A general awareness of the main concepts of linked data will put them in a good position for future developments.

Q. Is there a danger in the increasing reliance of faculty and students on Google, if for instance, it should go bankrupt?

A. If Google went bankrupt it would probably be because a more successful competitor had arisen. The idea is not to surrender our role to Google, but to recognize that such companies have enormous resources, and to take advantage of them by combining our expertise with their technology. We can't catalogue the web, but by linking to it from our catalogues we can in effect also search the web with greater precision.

• 2:25-3:20 p.m.  SESSION TWO: DISCOVERY, DELIVERY, AND DISSEMINATION OF INFORMATION IN A MULTILINGUAL AND MULTICULTURAL ENVIRONMENT: THE PUBLIC SERVICES' PERSPECTIVE

This session included three presentations. The presenter, Xi Chen, was absent due to her illness.

• "Building Relevance within the Classroom"
  By Shirin Eshghi, Japanese Language Librarian, University of British Columbia
  [Click for presentation slides]


• [Cancelled] "Integrating Digital Collections into the Liberal Arts Curriculum: A case study of Shansi Digital Collection at Oberlin College Library"
  By Xi Chen, East Asian and Web Development Librarian, Oberlin College Library
  [Click for presentation slides]


• "Turn Library Instruction on its Head: Capturing Information Literacy in a Blended Learning Model"
  By Asako Yoshida, Reference Librarian, Elizabeth Dafoe Library, University of Manitoba
  [Click for presentation slides]


Q&A: none


• 3:20-4:15 p.m.  SESSION THREE: DISCOVERY, DELIVERY, AND DISSEMINATION OF INFORMATION IN A MULTILINGUAL AND MULTICULTURAL ENVIRONMENT: THE TECHNICAL SERVICES' PERSPECTIVE

This session included three presentations.

• "The ISNI and Its Role in Information Discovery in a Multilingual Digital Environment"
  By Xiaoli Li, Head, Cataloging and Metadata Services Department at UC Davis Library
  [Click for presentation slides]

The presentation covers several topics associated with a new ISO Standard (ISO 27729) named International Standard Name Identifier (ISNI). Xiaoli began her talk with a brief introduction to ISNI. She emphasized that ISNI, a 16-digit code, is a persistent identifier which can be used to uniquely identify Public Identities of parties such as authors, composers, performers, publishers, organizations, or even fictional characters. Then she provided an overview of the ISNI Organization which consists of national libraries, OCLC, and societies and associations across multiple fields of creative activity. The ISNI database administrated and maintained by the Organization was discussed next. Xiaoli pointed out that the database has millions of names already and approximately 600,000 new names will be added every year.

Second half of the presentation was focused on the roles ISNI plays in improving users' resource discovery experience. Using Yimo Zhang, famous Chinese movie director, as an example, Xiaoli illustrated how various resources and information about or by an individual could be linked together using the ISNI identifier assigned to him/her. She concluded that this type of linkage can significantly enhance users' experience in searching resources that are in different formats and/or in multiple languages.

Xiaoli ended her presentation by discussing implications ISNI may have for East Asian libraries and CEAL. She speculated that: 1) East Asian libraries /or CEAL may play an essential role in promoting the adoption of the ISNI system in China, Japan, and Korea; 2) East Asian libraries and/or CEAL may be asked to help find a solution to differentiate the non-unique CJK personal name headings so those names can be represented in the ISNI database; 3) East Asian libraries may be involved in adopting ISNI at their own institutions and in educating their library users about the ISNI system.

• "Strengthening the Chinese Electronic Resources Supply Chain with Standards and Best Practices"
  By Bie-Hwa Ma, Chinese Language Electronic Resources Cataloging Librarian for CDL Systemwide Collections Services, UC San Diego Libraries
  [Click for presentation slides]

Bie-hwa started her presentation with the cataloging issues she faced for the last 4 years when processing Chinese electronic resources. She found some E-resources have incorrect metadata, incomplete or misleading content presentation, url link problems, etc. The source of the above mentioned problems could be traced back to any parties involved in the supply chain. The feasible solution is for all parties to comply with standards and practices established by OpenURL, KBART and PIE-J Working Group.

She then introduced OpenURL and its benefits. In theory, implementing OpenURL is able to provide appropriate copies, reduce linking failure, lower the maintenance cost, increase inter-operability for data exchange. Currently OpenURL has been widely implemented. However according to couple studies, about 20% of OpenURL still has broken links. Most of these failed linkages resulted from the incorrect, incomplete or inconsistently formatted metadata in the knowledge bases and citations.

KBART (Knowledge Base And Related Tools) Working Group was established in 2008. Its goal is to ensure the proper supply of metadata (or title lists) to both link resolvers and their knowledge bases. The phase I recommended practices of KBART are: 1. method and frequency of exchange 2. mandatory data fields, field labels, order and data format 3. specifics for certain type of metadata.

For the University of California's link resolution system, Bie-hwa found out that only about half of the journal packages are included in its knowledge base. In addition many providers have data exchange issues. She felt that all the E-resources publishers/providers should follow KBART's recommendations closely in order to meet the cataloging needs and to provide accurate and timely information as well.

Presentation and Identification of E-Journal (PIE-J) Working Group is another prominent group who has developed recommendations on the best practices for the e-journal. Their best practices include retention of original title and citation information; accurate issn number, title and publication histories; creation of presentation for easy access; and lastly retroactive digitization. Bie-hwa then gave a few E-journal examples for those that followed PIE-J's best practices closely and those that didn't.

Bie-hwa pointed out that there are still some issues that have not been addressed, such as the discrepancy on pinyin word division used by various parties involved in providing E-resources and metadata. The missing content and poor resolution required attention too. In conclusion Bie-hwa called for everyone's support to keep standards and best practices in mind when processing Chinese E-resources. She further suggested forming a working group to promote best practices and future developments for the Chinese E-resources.

Q&A
Q: Comments from Martin Heijdra: I have been told that even if it is now possible for us to add items locally, the granularity of our SFX database is insufficient to really treat our data well: They need a first and last issue, while e.g. the journals indexed in the Quan guo bao kan suo yin databases are often an issue here and an issue there, and as far as I have been told that is not really identifiable in knowledge databases--- that should change!

A: Bie-hwa agreed with Martin's comments and said KBART has special format for solving this kind of problem.

• "Building an improved resource discovery experience for users: a cataloger's experience and perspective"
  By Jia Xu, East Asian Cataloging Librarian at the University of Iowa Libraries
  [Click for presentation slides]

Jia Xu put forward an argument, that is, to build an improved resource discovery experience for users, an important cataloging procedure is to achieve consistency of bibliographic headings. To support her argument and introduce to the projects and tasks that she was going to present, she first used an example to demonstrate the importance of making consistent bibliographic headings for successful user tasks, from public and technical services point of view. To further support her argument and share her experience and insights, she talked about the three projects or tasks that she and her colleague conducted, all aiming to achieve consistency of bibliographic headings within the bibliographic records for Chinese and foreign videos and films. She then summarized what she and her colleague achieved and learned. She concluded that it is very important for catalogers to look at bibliographic data from user's perspective and to focus attention on fulfilling user tasks.

• 4:15-4:30 p.m. SESSION FOUR: RDA UPDATE FROM LC
  Tom Yee, Acting Chief, the Asian and Middle Eastern Division at the Library of Congress

[Click for presentation slides]

Tom Yee provided an update on how LC is preparing to implement RDA.

LC stopped cataloging with RDA on December 31, 2010 after the U.S. national libraries test was completed. LC restarted RDA cataloging at the end of November 2011, in order to address issues raised in the recommendations by the U.S. RDA Test Coordinating Committee. Approximately 30 catalogers and technicians have returned to RDA cataloging, and produced about 3,700 records in November and December. LC offered RDA refresher training for the catalogers, which is basically a reduced version of the original training. The training also included a number of changes to policies and training approaches based on the experiences and feedback from the original testers. Based on the questions raised by catalogers and record reviewers, special topic sessions were conducted to examine challenging issues more closely.

The new RDA website was launched in summer 2011. One of the topics that are covered in the website is LCPS (Library of Congress Policy Statements). There have been two major updates to LCPS, first in November 2011 and secondly in January 2012. The next update is scheduled in April 2012, and more revisions and additions are expected in the future. LCPS is freely available through RDA Toolkit.

PCC has been actively developing decisions and training plans that are necessary to implement RDA.

Four types of RDA developments are ongoing, which are, minor error fixes, more substantial content changes, re-wording of selected chapters by copy editor, and fast track proposal process, which helps pass decisions that are not controversial in a shorter turnaround time. Re-wording of instructions has been conducted in five chapters to be complete by June. This development intends to make the RDA instructions clearer and easier to understand by average catalogers without changing the contents. One example is the use of the word "script." Since the word can have different meanings, a clear definition will be provided when it first appears in the instruction.

LC has determined to target March 31, 2013 as RDA implementation Day 1. Training for more than 500 cataloging staff will begin this summer, with the first group in June and all supervisors in July. NAL and NLM are also aiming at early 2013 for RDA implementation.

Q&A
Q: How is training for the CJK cataloging team being planned?

A: It is being planned. Another related issue that has been raised is training for vendors. LC is still discussing the best way to train them. More information will be shared later.