By Janifer Gatenby
This entry was contributed by my colleague Janifer Gatenby, who works at the OCLC Leiden office. In her OCLC role, Janifer--a frequent speaker and standards specialist--identifies trends and opportunities for Web data services and system interoperation.
If I like a novel of one author, I look to see what else he or she has written. The same is true if I find somebody who writes well on topics that interest me, so I would like to find all he or she has written in any form - whether book, article or blog. The importance of grouping the works of authors together is clearly important to me; titles give me an idea about subject whereas authors give me an idea about quality and pertinence. Yet traditional library catalogues, though enriched with authority control of names, place the main emphasis on titles.
When the AustLit was first released in 2001 it drew lots of attention because it presented the literature primarily from the point of view of the author, with title of work as a secondary access and was also an early adopter of the FRBR model. The author details in AustLit go far beyond the traditional details in library name authority records, including things in the author's life that had a significant influence on the author's creativity, such as literary schools, peers, places lived and education. Prizes awarded are also included. AustLit was also before its time in that the data had been hand crafted, not by librarians, but by literary scholarly experts who signed the entries, thus being a forerunner to databases founded on user contribution such as Wikipedia.
AustLit was one of the inspirations for WorldCat identities, which creates a web page for each author within the WorldCat database, not by hand crafting some 600,000 records, but by data mining 110+ million bibliographic records to reveal new information about authors within WorldCat. Like AustLit, this new rich information includes works by and about the author. It also includes a time-line for the publication of the works and it provides external links to further author information, namely to the Virtual International Authority File (VIAF) and Wikipedia. WorldCat identities, incorporated within worldcat.org, is accessible from the details display of a manifestation. In the figure below, the entry for Ernest Hemingway is accessible from a record for The Old Man and the Sea.
National libraries continue to place emphasis on rich name authority records. The VIAF initiative links the records of the Library of Congress, Deutsche Nationalbibliothek and the Bibliothèque Nationale de France. Linking these three files results in a rich list of alternative name forms, all linked. More institutions are expected to contribute to VIAF in 2009.
Currently library resources are not treated equally by name authority records. Lorcan Dempsey pointed to a clear need to expand the scope of authority control to include authors of articles, papers and presentations in journals, institutional repositories and other sources. This is not an easy task for a large database such as WorldCat with 110+ million records with some name authority control and 57 million recently added article records with no name authority control. For scalability it will require some clever data mining to produce richer information in authority records including affiliations and co-authors that can be used to match and distinguish names and hence correctly group their works. There is also some valuable information that could potentially be inherited from the trade such Thomson ISI's highly cited database and Scopus Author Identifier.
Elsewhere, information created by the publishing sector is being brought into library databases systems as base data for selection, ordering and cataloguing. Much more could be adopted from the publishing sector in the way of author information. The various sectors in the publishing industry realise the potential benefits of sharing author information and this had led to the creation of a new standard identifier, the International Standard Name Identifier (ISNI). Participating in the creation of this standard are representatives from industry institutions which control authors' rights and royalties, led by CISAC, the International Confederation of Composers and Authors Societies. Significantly, the working group also includes library representatives from the Bibliothèque Nationale de France, the British Library and OCLC. ISNI promises to provide a central database of disambiguated authors and contributors with a facility to connect to enriched data in external databases. The ISNI database has potential for libraries to access authoritative data created by organisations with direct connections with their authors and contributors. Libraries have been invited to the table and will be encouraged to contribute information on authors and contributors of works in the "long tail"; works of no commercial or marginal commercial value.
Dorothea Salo, in an article soon to be published in Cataloging and Classification Quarterly, laments that retrieval from institutional repositories is severely limited by lack of authority control. "Once name variants creep into an institutional repository they are phenomenally difficult to extirpate". One notable exception is in the Netherlands, where all university researchers receive a unique authority number in the form of a Digital Author Identifier (DAI) that is registered in the authority file of the national union catalogue. The aim is to use these identifiers in all repositories, research databases and library catalogues. Publishers too, are sending ONIX records to the Dutch national union catalogue as part of national deposit and also aim to use the Dutch national name authority file in the creation of the ONIX creator data. Systematic use of the DAI identifier ensures an enviable tidiness in the way that the works of authors appear in catalogues and repositories of the Netherlands. Germany too provides a good example where contributions to the national authority file are made in real time from systems in libraries and networks using the SRU record update protocol.
With the Dutch DAI as a potential model, libraries have an opportunity to respond to the challenge to produce author views of a variety of resources. It will be interesting to see who will take up this opportunity. For success on a large scale, multiple initiatives are necessary, including data mining, employment of commercial data, at least some manual input, and enrichment via contribution from the public at large. The challenge is to create environments with enough web-scale and authority to attract public input via appeals and intriguing campaigns, even competitions. For example, "Create a publications list in a citation format of your choice for your curriculum vitae". Grouping the works of authors together at web-scale provides new opportunities to re-invent and redeploy traditional name authority control methods to create fresh new services for both end-users and authors.
]]>
