Staff Papers

Libraries Australia: Simplifying the Search Experience

A paper by Tony Boston, Bemal Rajapatirana and Roxanne Missingham, National Library of Australia at the Information Online 2005 Conference.

Keywords

National union catalogues; Kinetica; Resource sharing; Resource discovery; Search Tools and Techniques; FRBR

Abstract

The National Library of Australia is redeveloping its Kinetica service. A new search service called Libraries Australia was released in late 2004 which aims to simplify the search experience and improve performance for Kinetica users. The new service incorporates improvements such as parallel searching of multiple databases; Google-like searching with phrase, synonym and spell-checking features; results sorting options; an alerting service and improved options to get bibliographic items in Library collections; and other services such as online bookshops. A specification for application of the Functional Requirements for Bibliographic Records (FRBR) to the Australian National Bibliographic Database (ANBD) is also described. FRBR provides a conceptual model for the entities of interest to users of bibliographic records and defines a hierarchy of work, expression, manifestation and item entities. In a future release of Libraries Australia, it is planned to use FRBR concepts to simplify the search experience for users by clustering of results as has been done recently by a number of North American bibliographic agencies.

Introduction

Since 1981 the National Library of Australia has provided an online service to support resource sharing and collaboration between Australian libraries.

This service, which since 1999 has been known as Kinetica, enables Australian libraries to contribute to the national online catalogue and to benefit through shared collection building, cataloguing and interlending. It has also enabled libraries to link to overseas union catalogues and services such as those of the Research Libraries Group (RLG), the Online Computer Library Centre (OCLC) and the National Libraries of New Zealand and Singapore .

The service is an important part of the Library’s strategic aim to:

“provide rapid and easy access to the wealth of information that reside in libraries and other cultural institutions – and to break down the barriers that work against this”

National Library of Australia , Directions 2003-2005 [1]

At the core of the Kinetica service is the Australian National Bibliographic Database (ANBD) which currently records the locations of about 38 million items held in over 1,100 Australian libraries.

In 2004 and 2005 the Kinetica service is being redeveloped through a modularised two stage project [2]. The first stage is now complete and involved migration of the ANBD to a new software platform, implementation of a new directory service for customer and access management, and development of a new Kinetica search service which is known as Libraries Australia. The new search service is a major advance, enabling users to ‘find’ and ‘get’ resources in Australian libraries and online through an easy to use and high performance interface.

Figure 1 shows the components of the Kinetica service with the shaded areas of the diagram indicating the developments that took place during stage 1 of the Kinetica Redevelopment Project.

Figure 1 - Kinetica Redevelopment Project Stage 1

Figure 1: Kinetica Redevelopment Project – Stage 1

The reasons for the redevelopment of Kinetica are:

  • to introduce improvements for customers in areas such as search functionality, record contribution and online performance (response time);
  • to give the Library an improved capacity to innovate, and thus introduce further future improvements;
  • to integrate the Kinetica Chinese Japanese and Korean (CJK) service into the core system, and to provide the ability to handle all non-roman scripts supported by the Unicode standard;
  • to develop new search interfaces to meet the needs of individuals; and
  • to provide an affordable solution, supporting service development within the available budget with an appropriate risk management approach.

The redevelopment of Kinetica will deliver a solution which more closely realises the wishes of Australian libraries and users, and provides an innovative technical solution to enable improved resource discovery and resource sharing across the country.

Libraries Australia – the new Kinetica Search Service

Libraries Australia, the Kinetica search service, was released on 8 December 2004. The new service enables users to search and browse the ANBD as well as many overseas catalogues. Users have extended ‘get’ options; can save queries and records from any database; print, email and download records in a variety of formats and schedule saved queries to run periodically using a new alerting service.

This paper presents some of the technological innovations that have been introduced in Libraries Australia which include:

  • New software platform with improved search performance;
  • Unicode UTF-8 MARCXML compliant database;
  • Phrase, synonym and spell checking features;
  • Parallel searching of multiple remote databases;
  • Sorting of ANBD records;
  • New advanced search interface;
  • IP authentication;
  • Improved ‘get’ options;
  • Simplified print, email and download options;
  • Saved queries and records; and
  • Alerting service

New software platform

The ANBD has been migrated to the TeraText software platform [3] which has been used by the National Library since 2002 for other resource discovery services like PictureAustralia and Music Australia. TeraText is a high performance Z39.50 compliant XML database developed by InQuirion, a company spun out of RMIT. Software development work for TeraText is carried out in Melbourne and the software is used at sites in Australia, North America and Europe.

The new database has much improved performance characteristics and on its current hardware platform can support up to 350 users doing simple searches with below 2 second search and present times and up to about 1000 concurrent users with below 5 second search and present times. This represents a database throughput of about 100 searches per second. The increase in response times by number of concurrent users is approximately linear as shown in figure 2.

Figure 2 - TeraText Database performance

Figure 2: TeraText database performance

The Libraries Australia front-end web application has been developed using Java SDK 2.0 [4] and XHTML 1.0 [5] within the Library’s applications development framework and the TeraText Java API.

Unicode UTF-8 MARCXML compliant database

Data in the ANBD is held in MARC21 MARCXML format [6]. The data is encoded in Unicode UTF-8 [7]. Unicode compliance makes the data more accessible and meets the latest standard for encoding the world’s language scripts, including non-roman scripts. The TeraText database supports the variety of non-roman characters that occur in the ANBD and after completion of stage 2 of the project in late 2005, will include data from the Kinetica CJK database.

Phrase, synonym and spell checking features

Today’s users have come to expect instant and easy access to information resources through use of the Internet. Search engines such as Google [8] provide highly relevant results quickly through an easy to use single search box.

Libraries Australia has adopted a “Google style” simple search interface through which users can do phrase and synonym searching using exactly the same conventions as Google. For example, terms for phrase searching are simply enclosed in quotes, eg “Australian Story”, and searching by synonyms is denoted by including a tilda before the search term. A synonym search such as ~car will also search for the terms automobile, auto, motorcar and machine. Synonym searching has been implemented using the WordNet software [9].

There are no stop words in Libraries Australia; all words are searched for and the case of entered search terms is not significant. Terms entered into the basic search interface are automatically ANDed together, and the boolean operators OR and NOT are supported if entered in uppercase. This behaviour is also modelled on Google.

A dictionary [10] of over 270,000 English words, names from the Kinetica Authorities database and Australian places supports spell checking so that the system makes suggestions where terms are entered that are not in the dictionary. For example a search for “ularu” in the ANBD returns no hits and the user is directed to try “uluru”.

The simple search interface includes checkboxes that can be used to subset the ANBD by resources of various formats, such as newspapers, theses, pictures or music; or types such as Australian, online, government or conference. The basic search interface for Libraries Australia is shown in figure 3.

Figure 3 - Libraries Australia basic search interface

Figure 3: Libraries Australia basic search interface

From the basic search interface, by default the system searches the ANBD and PictureAustralia. The ANBD searches are run against the Title, Author and Subject Heading indexes and separate sets of results are provided for each index. This separation of results enables the user to quickly identify, for example, works that are about Patrick White as opposed to works by Patrick White.

These new features combine to make basic searching both easy to use and powerful, a feature familiar to users of Internet search engines.

Parallel searching of multiple remote databases

Libraries Australia is capable of searching multiple remote Z39.50 targets in parallel via a broadcast search. The new service currently allows searching of twelve remote targets (the ANBD, Kinetica Chinese Japanese and Korean Database, PictureAustralia, British Library Catalogue, Library of Congress Catalogue, OCLC WorldCat, Research Libraries Group Union Catalogue, SCIPIO, Singapore National Union Catalogue, Te Puna and RLG Union Catalogue) and also includes two new targets, the Australian full text journals (APAFT) and Canada Institute of Scientific and Technical Information (CISTI).

Search results are presented on a single web page with the first ten records from each target displayed. Links are provided in the summary area to results from each target separately and no attempt has been made to integrate the results. Users are also given the option to search Google on the results page using the same terms that they entered. Figure 4 shows the summary results page from a search for “uluru” against the ANBD and PictureAustralia.

Figure 4 - Libraries Australia summary search results

Figure 4: Libraries Australia summary search results

Sorting of ANBD records

Results sets of up to 10,000 records from the ANBD and Kinetica Authorities databases can be sorted by title, author or date of publication (ANBD only). This feature will be gradually extended to other databases available through Libraries Australia that support sorting.

New advanced search interface

The advanced search interface is most similar in design to the old KineticaWeb search interface. However a number of enhancements have been made that simplify the search experience for Libraries Australia users:

  • The number of items on the pulldown menu available for searching has been reduced without compromising functionality and the terminology has been simplified. Twenty-six items on the KineticaWeb Search interface have been reduced to twelve items in Libraries Australia by:
    • amalgamating searches into one pulldown menu item, for example a Number search searches all the identifier indexes and a Classification search searches National Library of Medicine (NLM), Library of Congress (LC) and Dewey classification number indexes; and
    • providing a pulldown menu of keywords such as “all of these”, “any of these”, “as a phrase”, “starts with” and “exact match” that controls the indexes that are searched within the TeraText database.
  • Users can now enter expanded values for languages and place of publication instead of the MARC codes. For example for languages french instead of fre can be used or for place of publication new south wales instead of xna can be used which allows for simpler and more intuitive searching.
  • When published has been added as a search option so that users can find the latest material in the database or material published before or after a particular year or within a range of years.
  • Where held has been added to allow users to search for items just held within libraries in one or a number of state/territories, or within their own library.

IP authentication

Users of Libraries Australia can be authenticated via an IP address instead of a username and password. This enables libraries that use the service to provide access through computers within a range of IP addresses and eliminates the need to enter username and password information through the Libraries Australia interface, providing a better level of security and enabling much easier end user access.

Improved ‘get’ options

With the advent of the Internet and the world wide web, user expectations have been significantly raised and it is no longer best practice to provide services that simply enable users to find catalogue records that describe an information resource. Users also expect to be able to ‘get’ items through streamlined and integrated workflows.

Libraries Australia attempts to meet this need by providing a number of options for users to ‘get’ items:

  • Online – providing direct access to the resource if it is in digital form via a URL;
  • Through their library – with links to their local catalogue and checks for availability where this is technically possible;
  • Through other libraries – again with links to their local catalogue and checks for availability where this is technically possible and, where the subscribing library agrees, the ability to place an ILL request online; and
  • Through other suppliers including bookshops, document supply services such as Infotrieve, CISTI and SCIPIO or the National Library’s Copies Direct service (for material held by the National Library).

For bookshops that support a Web Services interface such as Amazon [11], the new service is able to automatically check availability via an ISBN or author/title search against the bookshop’s database. This enables users of Libraries Australia to link directly to the page on the booksellers site that displays the item and associated ordering and price information.

For bookshops that do not support a Web Services interface (about 25 at present), links to their home page are displayed within Libraries Australia. The National Library encourages booksellers not yet listed to provide their details so that they can be added to Libraries Australia.

Figure 5 - Item availability information from the National Library's catalogue

Figure 5: Item availability information from the National Library’s catalogue

Figure 6 - Item availability from other libraries

Figure 6: Item availability information from other libraries

Figure 7 - Links to online bookshops for Uluru: an Aboriginal history of Ayers Rock

Figure 7: Links to online bookshops for Uluru : an Aboriginal history of Ayers Rock.

Simplified print, email and download options

Libraries Australia supports printing, emailing and downloading multiple records at a time. Users can simply check a box against records of interest and then select the print, email or download buttons. A variety of output formats for bibliographic data are supported including brief and full text, MARC and RIS format for import into bibliography management tools such as EndNote or ProCite as shown in figure 8.

Figure 8 - Emailing selected records

Figure 8: Emailing selected records

Saved queries and records

In Libraries Australia users can save queries and run them again the next time they login. They can also save records permanently that they have found that are of interest from any of the available targets.

Alerting service

Another new feature of Libraries Australia is an alerting service in which users can schedule saved queries to be run periodically, for example weekly, monthly or every three months. In figure 9, an alert is being created that will find all articles from the APAFT database that contain “endangered species” as a subject term.

Figure 9 - Save alert screen

Figure 9: Save alert screen

Future directions

In stage 2 of the Kinetica Redevelopment Project, Libraries Australia will be enhanced to:

  • support generation of record products for download to Kinetica customer systems;
  • allow some personalisation of the interface linked to individual user registration;
  • improve sorting performance and expand its use for non-ANBD databases;
  • incorporate data from the Kinetica Chinese, Japanese and Korean database; and
  • improve ‘get’ options including better links to document supply services and online bookshops.

A strategic direction for the Library is to enable a wider range of users, including end users, to search the ANBD more effectively. The Library would like to open access to Libraries Australia to as many Australians as possible and exposure of ANBD data to search engines like Google may also be trialled during stage 2 of the Kinetica Redevelopment Project. The OCLC Open WorldCat program [12] provides a model of what is possible in this area. The Senate Environment, Communications, Information Technology and the Arts Reference Committee‘s 2003 report on Libraries in the online environment stated:

The Committee recommends that the National Library of Australia receive additional funding to provide improved access to Kinetica for all Australian libraries and end users. [13]

The Library also is planning to investigate technologies such as the Open Archives Initiative Protocol for Metadata Harvesting (OAI-PHM) and Rich Site Summary (RSS) as mechanisms for sharing ANBD metadata.

FRBR specification

Overview

The development of the ‘Functional Requirements for Bibliographic Records’ (FRBR) Model [14] enables records, and the resources they represent, to be considered in terms of contextual relationships.

FRBR provides a conceptual model for the entities of interest to users of bibliographic records.

The model proposes a hierarchy of 4 entities or levels

  • A work (which is realised through)
  • An expression (which is established in)
  • A manifestation (which is exemplified by)
  • An item.

Library and union catalogues have, for many decades, stored bibliographic records according to format and content standards which, though generally rendering data at the manifestation level, will typically include information from all of these entities. These standards include MARC21, Anglo-American Cataloguing Rules, and Library of Congress Subject Headings.

The emergence of FRBR offers the potential to take the intellectual activities undertaken by cataloguers into a new environment where information is no longer held or viewed in a ‘flat’ structure. It can expose the simple and complex relationships recorded in bibliographic records which can then be translated into searching and display interfaces based on a more flexible understanding of resources. In this way FRBR can enable users to find resources which are related through catalogue entries which, while containing the information traditionally described in bibliographic records, are ‘clustered’ together in new ways.

As part of the Kinetica Redevelopment Project, work commenced in 2004 on developing the application of the FRBR model to the new Kinetica Search Service of the National Bibliographic Database (NBD). The expectation is that users of the Kinetica Search Service would be able to choose to cluster their search results by work name, and selected expression attributes to provide better precision and a simpler search experience.

A project was undertaken in 2003 [15] to investigate the potential to support and exploit FRBR concepts for the ANBD. The scope of the inquiry was:

  • analysis of records on the ANBD to determine those records which would be clustered through FRBR;
  • in depth analysis of a popular Australian work; and
  • consideration of issues relevant to the application of FRBR to the ANBD.

The study concluded that it would be possible to proceed with the design and implementation of an FRBR interface against the ANBD taking into consideration the following key factors :

  • The lack of consistent or sufficient data within bibliographic records on the ANBD to accurately differentiate expression level entities.
  • Utilisation of expression (E) and manifestation (M) attributes that are most useful to users and more importantly, are already available in ANBD records. Re-cataloguing is not an option and corrective or ‘value-add’ programming cannot address all data legacies.
  • Data quality issues in bibliographic data including problems such errors in MARC coding (e.g. incorrect material type codes, missing language of publication codes, relator codes etc), incorrect construction or choice of work headings, variant forms of names and name titles.
  • Scoping the application of FRBR selectively on the ANBD to a subset of bibliographic data for example to popular literary or theological works where careful aggregation and linking supports the users discovery of variant expressions and manifestations.
  • Data schemas to enable effective storage and transmission of hierarchical and parallel relationships
  • Scalability, performance and design of the FRBR interface to ensure rapid clustering, accommodate data volumes (current and future), and user friendly data display and navigation.

What follows is description of the design and specification undertaken to-date.

Scope and search screen

Initial design work was inclusive of all material types on the ANBD. However, given the complexity of delivering even a simple FRBR interface, the initial implementation was scoped and limited to monographs solely on the ANBD. Attempting to cluster across other target databases accessible through the Kinetica Search Service, including the Kinetica Authorities database, was eliminated early in the scoping process.

It is envisaged that any FRBR implementation would be piloted prior to adoption in the main Kinetica search service.

Access to FRBR clustering would be supported from the basic search interface as searches need to be based on author, title or subject only. An option under consideration is a checkbox (turned off by default) that would enable users to turn FRBR result clustering on or off against the ANBD target. An issue for resolution is how to support the limit options on the basic search in tandem with FRBR clustering.

Appropriate screen labelling, terminology and layout are primary considerations in assisting users to anticipate, understand, and fully exploit the delivery of search results that are FRBR clustered. The exact configuration of search and result pages needs careful design and usability testing and what is presented here is an initial starting point for development of the FRBR interface.

Several interfaces that use concepts based on the FRBR model have been examined during the development of specifications including OCLC’s FictionFinder [16], RLG’s RedLightGreen [17], Library of Congress’s FRBR display tool [18], and the AustLit gateway [19] Of these, RedLightGreen was considered to be the most user friendly and has influenced the screen specifications.

Result screens

Result screens need to present works clustered by workNames with links to appropriate expression-manifestation level keys, collectively referred to as ‘versions’. The decision to present un-differentiated manifestation and expression level information is driven partly by the data (or lack thereof) available in ANBD records and also the need to streamline the presentation of results and minimise the interface layers that the user must negotiate.

On selecting the “group results” option (an interim name for invoking FRBR result clustering) the work summary results page is returned and should present a list of workNames sorted alphabetically. A workName is a combination of Author and Title headings (MARC 1XX and 240 or 245) or a uniform title (MARC 130) heading if it exists. workNames should be derived on insertion of bibliographic records into the database utilising a workName algorithm based on the OCLC FRBR Work-Set Algorithm [20]. Authors and titles from the bibliographic database are first normalised using modified NACO rules [21] and then checked against the Authority database to determine their preferred form before a workName is constructed.

Work summary screen

The work summary screen will present workNames ranked by availability and relevance. Availability is related to the number of holdings for a given workName. For example, for a search on “Jane Campion”, a list of works similar to the following might be returned:

Search for: Jane Campion

  • The Piano Jane Campion

Fiction

22 versions between 1990 and 2003 in English and 6 other languages. Includes books (braille, large type) and talking books.

Available in 45 libraries including yours

Primary subject: Piano Motion Picture

Reviews available. Abstract: Based on Jane Campion's award winning film, this novel tells the story of Ada , who with her piano and her nine-year-old, illigitimate daughter Flora, migrates to the remote bush of nineteenth century New Zealand for an arranged … more.

 

  • Piano: A Novel Jane Campion

Fiction

6 versions between in 1994 and 1996 in English. Includes books (large type).

Available in 26 libraries.

Primary subject: Piano Motion Picture

Reviews available. Suitable for adult audience.

  • Piano: The Novel Jane Campion

Fiction

2 versions in 1994 in English.

Available in 26 libraries.

Primary subject: Piano Motion Picture

Reviews available

 

  • The Piano Michael Nyman

Motion picture music. Suitable for general audience.

6 version between 1993 and 1998 in English. Includes music (full score) and sound recordings (sound disc).

Primary subject: Piano music

Reviews available. Contents: To the edge of the earth ( 4:06 ); Big my secret ( 2:51 ); A wild and distant shore ( 5:50 ); The heart asks pleasure first ( 1:33 ); Here to there ( 1:02 ); The promise ( 4:14 ); A bed of ferns (0:46); The fling ( 1:29 ); The scent of love … more

Refines Search by

Subjects:
Campion Jane
Campion Jane - Criticism And Interpretation
English Fiction
Piano Motion Picture
Large Type Books

Show More Subjects

Authors:

Campion Jane
Campion, Anna
Pullinger, Kate
Coombs, Felicity
Gemmell, Suzanne

Nyman, Michael

Show More Authors

 

Figure 10: Proposed work summary screen

 

The right-hand side of the page lists works by availability (based on the number of Library holdings for that work) and relevance. Under each workName is its form, primary subjects and number of versions. On the left side of the screen are related subjects and authors.

Work details

In order to support the above display, work-related attributes need to be extracted and aggregated from the Bibliographic database. For a database as large as the ANBD, it could be difficult to generate this display in real-time with sufficient performance. An approach under consideration is to maintain a Work database in concert with the existing Bibliographic and Authority databases. Implementation of this solution would entail developing robust synchronization or update processes between the Work and Bibliographic databases.

It is proposed to use MODS [22] as the schema for the Work database as it is sufficiently flexible and granular in terms of content designation and permits a hierarchical expression of work records. The following table lists Work, Expression and Manifestation attributes that would be used within the MODS record. It is not exhaustive and it should be noted that initially only those MARC sources relevant to books (BK) would be utilized.

Name

Definition

MODS element

MARC Source

Comments

workId

A unique control number assigned to the work

/identifier

 

System-generated

Retain versionId (manifestation) in separate attributes

workName

Name of the work

/titleInfo/title +
/name/displayForm

100,110,111+240,245; 130

 

Author-title or author-uniform title if it exists. Used to link to the Bibliographic database.

*/name currently only consists of author names (1XX).

form

Form of the material – primarily intellectual or literary form

/genre/genre
authority=”marcgt”

008/24-27 (BK)

33/34 (BK)

008/21 (SE)

008/18-19 (MU)

008/30-31 (MU)

008/25 (MP)

008/26 (CO)

008/33-34 (VM,MP)

Eg Thesis, Non fiction, Fiction etc

primarySubject

Subject heading that has most number of linked manifestation records.

/primarySubject

6XX

 

versionIdentifier

Identifier for the version

/relatedItem/identifier

001

Link to manifestation record in Bibliographic database

versionName

Name of the version

/relatedItem//titleInfo
/title+/relatedItem
/name/displayForm

245$a$b$n$p$c (first

 

versionLanguage

Language for an edition

/relatedItem/language/
languageTerm

008/35-37

 

versionNUC

NUC holdings for an edition

/relatedItem/location/
physicalLocation

850 $a

NUC symbols for Libraries holding a version

ISBN

ISBN of the version

/identifier type=ISBN

020 $a

Retrieved from Bib database using AN identifier

edition

Edition of the version

/relatedItem/
originInfo/edition?

260$a

 

targetAudience

Target end user group

/ targetAudience authority="marctarget">

008/22 (BK, CF, MU, VM)

E.g. adolescent, adult preschool

Table 1: Mapping work, expression and manifestation data to MODS

Business rules have been defined to harvest the details listed above and key to this is the concept of establishing a primary manifestation (bibliographic) record from which to derive the bulk of the work details data, and including a preferral process to determine this primary record. Where data is not available a set of rules have been defined for acquiring data from other bibliographic records within the work set. Where no data is available it will be requirement for the interface to suppress the display of the data label, for example, if no targetAudience is available or the MARC coding is non-specific.

An example MODS record for The Piano is available in Appendix 1.

The work details screen will display work level attributes including a version statement, form information, subjects and genre. Below the work level details will be summary information taken from the manifestation level records for each version of the work, for example a title statement, publisher, ISBN, and number of libraries holding the version. Figure 11 is indicative of the information that would be presented on the work details screen.

 

 

 

An Angel at my table .Frame, Janet 1924-
31 versions between 1984 and 2003, in English, Italian, German, Greek, Chinese and 5 other languages.
Subjects: Frame, Janet, 1924- -- Biography; Authors, New Zealand —Biography
Other Formats: Large print books
 

An Angel at my table : an autobiography / Janet Frame

New York :G. Braziller, c1984

80761042

In 6 libraries

An angel at my table : the complete autobiography / Janet Frame

Bath : Chivers, 1987

 

In 5 libraries

Ein Engel an meiner Tafel : autobiographischer Roman / Janet Frame

München : Piper, 1997

349222413X

In 2 libraries

Fu an tian shi / [Janet Frame] zhu

Taibei : Shi bao wen hua, 1997

9571324175

In 3 libraries

Figure 11: Proposed work details screen

FRBR Future directions

FRBR is a conceptual framework, and in developing the new interface Kinetica staff will continue to explore the issues associated with the implementation of FRBR, seeking where possible to build upon the activities of LC, RLG and OCLC and to adopt best practice. Many issues are yet to be resolved. For example the adoption of data schemas in which to encode and exchange FRBR data, the revision, deployment and uptake of international cataloguing standards to accommodate FRBR and the promotion of network standards to encourage data creation that maximises the exposure FRBR entities within the existing MARC record format.

Not all issues will be resolved immediately but as the base of implementations expands, through continued experimentation and iteration, creative and effective ways will be found to adopt the FRBR model.

Conclusion

The redevelopment of Kinetica is a major project for both the National Library and Australian libraries. The service provides a fundamental building block for national cooperation and resource discovery. The redevelopment will enhance this role while supporting greater efficiencies for Australian libraries and enabling much easier access to Australian library collections.

The redevelopment of Kinetica provides an example of the utilisation of new technologies and standards in an innovative manner. The connecting of libraries and commercial suppliers, such as booksellers and publishers, into a supply chain for the resource discovery service, provides significantly improved ‘get’ functionality for Libraries Australia users.

Finalising the Kinetica Redevelopment Project will not signal an end to the development of the service which underpins the national information network. Migration of the ANBD to a robust and high performance software platform should enable the Library to more quickly respond to user needs and suggestions for enhancements. The new software platform also provides a stable base on which to build value-added services in the coming years.

References

[1] National Library of Australia (2002). Directions for 2003-2005,http://www.nla.gov.au/library/directions.html

[2] National Library of Australia (2003). Kinetica redevelopment project website, http://www.nla.gov.au/kinetica/redevelopment.html

[3] InQuirion Pty Ltd (2004). InQuirion – Makers of TeraText, http://www.teratext.com.au/

[4] Sun Microsystems (2004). Java Servlet Technology, http://java.sun.com/products/servlet/

[5] The World Wide Web Consortium (W3C) (2002). XHTML 1.0: The Extensible HyperText Markup Language (Second Edition),http://www.w3.org/TR/xhtml1/

[6] Library of Congress (2004). MARCXML MARC21 XML Schema, http://www.loc.gov/standards/marcxml/

[7] The Unicode Consortium (2004). Unicode home page, http://www.unicode.org/

[8] Google (2004). Google, http://www.google.com/

[9] Cognitive Science Laboratory, Princeton University (2004). WordNet®: a lexical database for the English language, http://www.cogsci.princeton.edu/~wn/

[10] The Solution Café (2004). Jspell Web Spell Checker, http://www.jspell.com/

[11] Amazon (2004). Amazon Web Services, http://www.amazon.com/webservices

[12] OCLC (2004). Open WorldCat program, http://www.oclc.org/worldcat/open/default.htm

[13] Australia . Parliament. Senate. Environment, Communications, Information Technology and the Arts References Committee (2003). Libraries in the online environment. Canberra : The Committee, 2003

[14] IFLA Study Group on the Functional Requirements of Bibliographic Records (1998). Functional Requirements of Bibliographic Records: final report. München: K. G. Saur. http://www.ifla.org/VII/s13/frbr/frbr.pdf

[15] Rajapatirana, Bemal (2004). Testing FRBR Against the National Bibliographic Database , http://www.nla.gov.au/lis/stndrds/grps/acoc/rajapat2004.doc

[16] OCLC Research (2004). OCLC FictionFinder, http://fictionfinder.oclc.org/

[17] RLG (2004). RedLightGreen,http://www.redlightgreen.com/

[18] Library of Congress (2004). FRBR Display Tool Version 1.0,http://www.loc.gov/marc/marc-functional-analysis/tool.html

[19] (2004). AustLit – Australian Literature Gateway, http://www.austlit.edu.au/

[20] Hickey, Thomas B. and Toves, Jenny (2003). FRBR Work-Set Algorithm, http://www.oclc.org/research/software/frbr/frbr_workset_algorithm.pdf

[21] NACO Program for Cooperative Cataloging (2001). Authority File Comparison Rules (NACO Normalization) , http://www.loc.gov/catdir/pcc/naco/normrule.html

[22] Library of Congress (2003). Metadata Object Description Schema (MODS) v3.0, http://www.loc.gov/standards/mods/

Appendix 1: Example MODS record for The Piano.

 

<?xml version="1.0" encoding="UTF-8"?>
<modsCollection xmlns:xlink="http://www.w3.org/1999/xlink" xmlns:xsi="http://www.w3.org/2001/XMLSchema-instance" xmlns="http://www.loc.gov/mods/v3" xsi:schemaLocation="http://www.loc.gov/mods/v3 http://www.loc.gov/standards/mods/v3/mods-3-0.xsd">
<mods version="3.0">
<identifier>wn12345678910</identifier>
<titleInfo>
<title>The Piano</title>
</titleInfo>
<name type="personal">
<namePart>Campion, Jane</namePart>
<displayForm>Jane Campion</displayForm>
</name>
<!-- Workname is a combination of title and displayForm of name(s), eg "The Piano by Jane Campion" -->
<typeOfResource>text</typeOfResource>
<genre>fiction</genre>
<originInfo>
<dateValid encoding="iso8601" point="start">1990</dateValid>
<dateValid encoding="iso8601" point="end">2003</dateValid>
</originInfo>
<subject authority="lcsh">
<topic>Man-woman relationships</topic>
<geographic> New Zealand </geographic>
</subject>
<subject authority="lcsh">
<topic>Mothers and daughters</topic>
<geographic> New Zealand </geographic>
</subject>
<subject authority="lcsh">
<topic>Motion picture plays</topic>
</subject>
<subject authority="lcsh">
<topic>Mute persons</topic>
<geographic> New Zealand </geographic>
</subject>
<subject authority="lcsh">
<topic>Women Pianists</topic>
<geographic> New Zealand </geographic>
</subject>
<abstract>”The piano” is …</abstract>
<relatedItem type="constituent">
<identifier>000010494320</identifier>
<!-- Link to the Bibliographic database via the Australian National Bibliographic Number (AN). -->
<titleInfo>
<title>The piano</title>
</titleInfo>
<name type="personal">
<namePart>Campion, Jane</namePart>
<displayForm>Jane Campion</displayForm>
</name>
<name type="personal">
<namePart>Pullinger, Kate</namePart>
<displayForm>Kate Pullinger</displayForm>
</name>
<originInfo>
<dateValid encoding="iso8601" point="start">1994</dateValid>
<dateValid encoding="iso8601" point="end">1995</dateValid>
</originInfo>
<typeOfResource>text</typeOfResource>
<genre>fiction</genre>
<language>
<languageTerm authority="iso639-2b">eng</languageTerm>
</language>
<language>
<languageTerm authority="iso639-2b">fre</languageTerm>
</language>
<!-- language(s) of this version/edition -->
<location>
<physicalLocation>ADFA</physicalLocation>
</location>
<location>
<physicalLocation>ANL</physicalLocation>
</location>
<!-- locations are the unique NUC symbols for the holdings of all the manifestations of this version/edition -->
</relatedItem>
<relatedItem type="constituent">
<identifier>000010954414</identifier>
<!-- Link to the Bibliographic database via the Australian National Bibliographic Number (AN) -->
<titleInfo>
<title>The piano lesson: Screenplay for a Film</title>
</titleInfo>
<name type="personal">
<namePart>Campion, Jane</namePart>
<displayForm>Jane Campion</displayForm>
</name>
<originInfo>
<dateValid encoding="iso8601" point="start">1991</dateValid>
</originInfo>
<typeOfResource>text</typeOfResource>
<genre>fiction</genre>
<language>
<languageTerm authority="iso639-2b">eng</languageTerm>
</language>
<!-- language(s) of this version/edition -->
<location>
<physicalLocation>NMQU</physicalLocation>
</location>
<location>
<physicalLocation>NNCU:A</physicalLocation>
</location>
<location>
<physicalLocation>NTSM</physicalLocation>
</location>
<!-- locations are the unique NUC symbols for the holdings of all the manifestations of this version/edition -->
</relatedItem>
<!-- 7 more versions/editions here -->
</mods>
</modsCollection>

Back to the top