Library Technology Guides

Document Repository

Smart Libraries Newsletter

Smart Libraries Q&A: Discovery of library resources

Smart Libraries Newsletter [April 2017]


“Within the library field, I often hear comparisons of ILS or discovery search with Amazon and Google capabilities, with the deficiencies of the ILS/discovery being on the losing side. It is natural this occurs since search is so ubiquitous within the online world, but my suspicion for this library search weakness is that the library search market is so small compared to the giant internet companies most associated with search. Is this a correct assumption? Are the number of software developers employed by the companies most often involved in library software a good comparison and a reason for this? Or is there something more fundamental at the root of it all, e.g. MARC21, RDA, not enough money spent? I would say I hear the comparison with library search vs ‘big search' so often among my peers that the deficiencies of the library software are just a given and is not an assumption that requires being challenged, but why is this happening?”

I think that there is room—even the necessity—for multiple approaches to the discovery of library resources. There is not, and likely never will be, a single tool or platform that can prove comprehensive and seamless access to the content and services libraries provide to their communities.

Libraries center on a specific portion of the overall content available on the web. We select, license, purchase, and otherwise vet the specific resources that are deemed to be reliable, objective, and meaningful to the members of the community or institution the library serves. Optimizing a patron's access to that component of the vast content of the general web in the context of well-funded commercial interests presents a major challenge to libraries.

Library discovery needs to take place on multiple levels, each suited to the context, interests, and sophistication of the patron. It is important to make libraryselected materials easily accessible for those who start with Google, Amazon, or other commercial search environments, to offer broad-based library-specific discovery services, as well as more complex search tools design for specific disciplines.

In the realm of global web search, it is essential for libraries to maximize their presence. We can't deny the reality that most researchers start searching for resources through search engines Google, Google Scholar, Bing, or Baidu. Libraries therefore must have strategies to optimize the discovery and access to their resources in that context. The closed nature of library catalogs as well as obscure and dated metadata standards place barriers in the way of more organic inclusion of our content and services via the general web environment. Structures such as MARC21 were developed for the exchange of data between library systems, but do not fit well within the broader web architecture. Given this disconnect, libraries have to implement mechanisms that bridge the gap between library-specific metadata and technologies with that of the web. In the short term, libraries can implement search optimization techniques similar to those used in the commercial sphere. Examples include enhancing the way that resources are exposed on the web to embed structured metadata using, Open Graph, or other structures. Bibliographic containers such as MARC21 and its associated AACR2 or RDA cataloging rules provided essential standardization for libraries to operate efficiently and cooperatively. But it is important to note that these standards are entirely unsupported and not understood outside our own domain. They provide very rich metadata, but in an obscure way from the viewpoint of World Wide Web Consortium (W3C) web standards. Creating crosswalks and mechanisms for expressing data in ways friendly to search engine indexing bots can greatly improve the performance of library resources in broader web search tools. In the longer term, libraries are shifting to more web-friendly models in or management of bibliographic resources. The BIBFRAME proposals to use linked data as the carrier for bibliographic information falls into this category. Moving from library-specific metadata structures to those consistent with the broader web has great potential to strengthen the relevancy of libraries on the web, but it will be an incredibly difficult and expensive transition.

I also see the genre of library-specific discovery services as important and ever improving. These products are optimized to access the content curated by libraries for their communities. They address the print, electronic, and digital resources that the library has selected, hopefully ensuring that the items discovered will be of high quality and easily accessed. While these discovery services have their limitations, they have advanced considerably since their introduction. These tools have largely superseded the traditional online catalog in academic libraries that address mostly the print resources managed within the ILS. Library patrons often did not understand the scope of online catalogs, especially the absence of the articles available through the library's subscriptions to electronic resources.

It is also important for libraries to offer tools and interfaces optimized for specific disciplines or types of content. The online catalog of the ILS, for example, can be seen in this category for those researchers that specifically want an advanced search tool for print resources. Specialists will continue to use the native interfaces to the key resources in their discipline rather than start with the broader discovery service. Another key issue has to do with the objectivity of search and the degree to which a service respects the privacy of the searcher. The top-level Internet services follow a diametrically different business model and set of values than libraries embrace. They are designed to generate revenue through sales and advertising and to collect and exploit personal information to the fullest extent possible. Libraries, in contract aim to provide objective content to their communities without cost and to safeguard their privacy. These ideals of objectivity and privacy may impose some restraint on the degree of personalization in the user experience offered by libraries compared to those like Google and Amazon. Hopefully libraries will continue to develop ways to enhance their services without compromising these essential values.

As the question highlights, those developing search services within the library domain will never have the resources at the level of the top-level Internet giants. But the consolidation of the library technology industry has created a set of companies with the development capacity that has never existed before. These organizations have the capacity to create more sophisticated and powerful technologies than were possible when the environment was more fragmented.

Finally, I don't see library-provided discovery services and the Internet giants as engaged in a win-or-lose competition. Those organizations have little interest in library-specific services. Given the divergent value propositions, I don't think that libraries can or should expect to rely on them entirely in the way that we deliver content and services to our communities. But to the extent that our users make use of those services, it is essential for libraries to be well represented. As libraries shift to linked data as the core model for managing and providing access to collections, we will be more natively part of the web ecosystem. That scenario will come with its own positive and negative implications, but should result in opportunities for libraries to create services more in line with the expectations set by the top level commercial entities.

View Citation
Publication Year:2017
Type of Material:Article
Language English
Published in: Smart Libraries Newsletter
Publication Info:Volume 37 Number 04
Issue:April 2017
Publisher:ALA TechSource
Place of Publication:Chicago, IL
Record Number:22564
Last Update:2022-11-29 00:29:32
Date Created:2017-05-02 18:00:03