DynixInstitute Upstream [January 2005]

Web services: a technology trend worth watching

Breeding, Marshall.

Copyright (c) 2005 Dynix

SummaryAs the next decade unfolds, library applications will adopt the Service Oriented Architecture and rely less on library-specific standards and protocols.

As the next decade unfolds, library applications will adopt the Service Oriented Architecture and rely less on library-specific standards and protocols.

Today, libraries find themselves questioning their future in the wake of an onslaught of challenges posed by Internet search engines. Google Scholar recently launched as a search service specializing in academic literature, encroaching deep into library turf. With librarians still ruminating on the implications of Google Scholar, we learned of Google’s intent to digitize millions of books from the collections of the University of Michigan, Stanford University, Harvard University, the New York Public Library, and the Bodleian Library at Oxford—an undertaking of a scale that few predicted would be practical with today’s technologies. The future seems to be unfolding more quickly than many of us ever expected.

Discussion of these developments is widespread throughout the library community. In my opinion, the far-reaching projects of Google are but examples of a larger trend that over the next few years will play out to challenge and transform libraries. This trend involves technologies and services developed outside the library arena forcing libraries to break out of their traditional approach to technology. To compete with formidable external competition, libraries need technologies in tune with the times and the ability to adapt to a rapidly changing environment.

Library automation to date has progressed at a leisurely pace. Our catalogs have evolved through several presentation technologies: index cards, text-based online catalogs, graphical client/server interfaces, and the current generation of Web interfaces—but they have not made significant conceptual advancements. The direction has followed a conservative course, using technology to execute concepts of bibliographic description and search-and-retrieval models developed many decades ago when the universe of concern was entirely print and physical. Library automation is just now beginning to have a vision that surpasses an electronic version of a card catalog in favor of a library information environment that spans a wide array of information sources, with ever-increasing proportions of digital and non-print content.

I have long thought that library standards develop too slowly. Reliance on standards enables interoperability and protects libraries from dependence on proprietary technologies developed by any given company; they must be embraced by both developers and consumers of library automation systems. Yet, relative to the pace of technology advancement in industry and society, library standards develop at a glacial pace. As libraries develop strategies in response to the latest onslaught of challenges from Google, Amazon.com, and other external forces, we will need to find ways to develop compelling technologies rapidly.

Libraries will depend less and less on protocols, standards, and technologies developed for and by libraries and more on those that prevail broadly throughout the computer industry. I’m not saying that libraries should abandon standards, but that we also need other alternatives for interoperability that can be developed and deployed faster with broader scope and impact.

In this context, the key trend that I see impacting libraries over the next decade will involve the adoption of the Service Oriented Architecture (SOA) as the basis for interoperability among library automaton applications and especially between library automation systems and non-library applications. Also known as Web Services, encompassing components such as XML (eXtensible Markup Language), SOAP (Simple Object Access Protocol), WSDL (Web Services Description Language), and UDDI (Universal Discovery Description, and Integration) directory services, these technologies have gained broad acceptance and adoption in software applications across many information-oriented industries. As I follow the software industry, SOA continually emerges as the preferred framework for new applications.

We can see the beginnings of an interest in Web Services in the library world today. One of the early examples of a web service developed in the library automation arena is SRW (Search and Retrieve Web Service) a flavor of Z39.50 build on SOA. SRW emerged through the Z49.50 Next Generation, or ZING committee of NISO. More recently, a number of the major library vendors formed a consortium called VIEWS (Vendor Initiative for Enabling Web Services) to develop Web Services for library automaton systems. These efforts hopefully represent just the initial momentum for the adoption of Web Services as the basis of interoperability for software applications in our industry. Web Services will be a great benefit to libraries as the future unfolds—a future where the communication with external non-library applications may be even more important than the traditional concerns of interoperability within and among library automation system components.

I believe that one of the realities in store for libraries involves the need to provide library functions and services through interfaces provided by others. For example, as course management systems become widely adopted by colleges and universities, courseware will become the default interface for students. It will be vitally important for libraries to provide access to their resources and services through back-end technologies that are tightly integrated with these course management systems. The adoption of the Service Oriented Architecture and the development of Web Services will allow libraries to permeate their work throughout a wide variety of interfaces and systems beyond those directly provided by the library. Over the course of the next decade, libraryoriented Web Services will be an essential component to enable libraries to integrate what they do into the prevailing environment. In following technology trends, it’s extremely hard to predict the ones likely fizzle out versus the ones that will endure. In my view, the Service Oriented Architecture is destined to prevail, and is one that libraries can use to their advantage to deliver services in a future world that may be dominated by information providers outside the library community.

Publication Year:2005
Type of Material:Article
Published in: DynixInstitute Upstream
Publication Info:Volume 1 Number 1
Issue:January 2005
Place of Publication:Provo, UT
Permalink: http://www.librarytechnology.org/ltg-displaytext.pl?RC=11342
Record Number:11342
Last Update:2012-12-29 14:06:47
Date Created:0000-00-00 00:00:00