Copyright (c) 2009 Information Today
Abstract: Enterprise computing ranks as one of the most firmly established trends in IT over the last decade. This approach involves building a single technical infrastructure, designed as an organic whole, that spans an organization. Enterprise networks address each aspect of computing with a single industrial-strength solution, rather than having individual units within the organization each solve the same problem. Libraries gain many benefits as they shift from self-contained silos to full partners within the enterprise networks of their parent organization.
Enterprise computing ranks as one of the most firmly established trends in IT over the last decade. This approach involves building a single technical infrastructure, designed as an organic whole, that spans an organization. Enterprise networks address each aspect of computing with a single industrial-strength solution, rather than having individual units within the organization each solve the same problem. Libraries gain many benefits as they shift from self-contained silos to full partners within the enterprise networks of their parent organization.
The degree to which a library needs to orient itself to the enterprise varies by its organizational circumstances. Libraries within corporations, universities or colleges, and government agencies almost always exist within a technology infrastructure provided by a higherlevel organization. Assumptions for public libraries differ considerably. Many, especially those in small communities or rural areas, have few opportunities to leverage existing computing facilities and need to implement a complete, self-contained system. Public libraries for large municipal, county, or regional systems may find opportunities to be better connected to their communities through closer connections and cooperation with those IT departments. Libraries of all types need to be aware of this trend toward enterprise computing and must take advantage of the computing facilities provided externally when possible.
But for libraries that do reside within large organizations, connecting with the broader enterprise network is a vital strategy, both technologically and organizationally.
Consider email as a typical example of the shift to an enterprise approach in technology. A decade ago, it was common for any given department within a corporation or university to operate its own mail server. These units might have had their own technical support personnel that supported their computing needs and afforded them a significant deal of independence from their parent organizations, which may not have had a great reputation for delivering IT services.
Over time, the business of managing services such as email becomes increasingly complex. Staggering volumes of spam and relentless attacks of viruses, worms, and other malware forced email administrators to implement complex defense systems. Within an organization, the burden of having multiple departments each taxing its resources to solve these problems became untenable. There came a point where it made sense to abandon individual departmental mail servers in favor of a centralized service with hardened security, scaled to meet the needs of the entire organization. This approach also provides institutional benefits by supplying a communications infrastructure that spans the entire organization, which can be extended to provide other related services. In the earlier days of my career, our library at Vanderbilt University ran its own mail server, as did most other university libraries. Today, I'm aware of few libraries that operate their own mail services. We're delighted for that service to be provided as part of the enterprise network of the university.
Similar transitions took place with file storage. It's common today for organizations to rely on large-scale storage systems in the form of storage area networks (SANs), hierarchical storage management, and other kinds of digital storage technologies rather than have multiple departments invest in high-capacity storage implementations. Again, libraries often benefit from sharing institutionally provided storage for digital library projects rather than having to create their own separate storage environments, dealing with the never-ending issues of data backup, replication, scalability, and disaster recovery. Libraries have their own perspective on storage, especially in the arena of digital preservation. But these concerns need to be applied to the broader information assets of the organization and not just those directly managed by the library.
These examples illustrate one of the key principles of enterprise computing - that of taking advantage of services provided at a higher level whenever possible. Enterprise computing concentrates on a single industrialstrength solution to each area of technology rather than leaving it to each department within the organization to solve the same problem with inadequate resources. This approach frees each unit to focus on the issues unique to its role and to amplify the value it adds to the organization. By taking advantage of technology resources provided within the enterprise infrastructure, each of its composite groups has more resources available for the specific applications that will yield a bigger impact.
In many ways, the traditional library automation model, based on an integrated library system (ILS), tends more toward the departmental model of computing than the enterprise computing paradigm. The ILS is a self-contained system that largely deals with the operation of the library outside the context of the broader organization.
One of the characteristics of typical library automation systems involves the tendency toward an isolated, selfcontained approach. The design of the ILS attempts to provide a complete automation environment for a library, but it does not necessarily extend its reach into other components of the broader environment. Most libraries maintain their own database of users and one or more repositories of information resources and provide their own specialized interfaces to deliver access to its services.
The current model of the integrated library system remains rooted in the realm of departmental computing and has not made the transition to enterprise thinking. How well does your current automation environment integrate with the enterprise? Does the library maintain its own authoritative patron file, or does it rely on the authorization and authentication environment of the larger network? Does the library maintain its own self-contained business processes for the procurement of materials? Or does it take advantage of the accounting and procurement applications of its home institution? Does the library assume that users will flock to its own website and search tools, or does the library provide access to its content and services through the portal of its home institution?
It's also common to see libraries in intermediate stages of enterprise integration. A typical university library will populate the patron database of its ILS with records derived from multiple sources throughout the campus, imported through a batch process and regularly synchronized. Some libraries perform some aspects of materials acquisitions within the local ILS and then hand off the transaction to the institution's financial systems for final payment. Enterprise integration is not an all-or-nothing proposition - it's a continuum that should be working its way toward increased levels of shared processing.
One of the key advantages to an enterprise computing approach for libraries involves the opportunity to become more interconnected with their institutions' strategic priorities. On the flip side, libraries that fail to integrate into the environment of their broader organization risk diminished relevancy and may miss out on new opportunities for support.
It's to the advantage of the library to position itself as a vital part of the business strategy of the overall organization, especially in areas related to the management of information assets. Libraries tend to look inwardly, implementing automation components optimized for their own internal priorities, and they may not give enough consideration to their role as a component of the larger enterprise framework.
The shift to enterprise computing is driven by the need to eliminate redundancies of computing and personnel resources and to deploy a technical infrastructure well aligned with the organization's strategies. A key responsibility of the chief information officer or the chief technology officer of the organization involves molding the technology assets to meet the needs of the broader organization. The CIO should see the library as one of the key components of information infrastructure. If the library is working within its own self-contained silo and has not yet engaged with the realm of enterprise computing, it risks not being in the strategic conversations regarding issues in which it has a large stake. Rather than trying to keep a safe distance from the influence of the organization's IT sphere of influence, libraries need to exploit any possible opportunity to participate in the strategic enterprise infrastructure.
Since I have asserted the value of libraries moving beyond isolated models of computing and becoming more oriented to the broader institutional enterprise infrastructure, I need to also mention some practical measures to accomplish this goal. At a minimum, it involves a heightened level of awareness about the issue, but technology also plays a role.
At one level, participation in the enterprise isn't a technology issue but an organizational strategy. It's just sound management to keep the library's strategic goals and objective well aligned with that of the larger institution. From my personal observations, this approach is already well-established in most libraries. I see much more of a mixed message on the technology front. It's not at all unusual to see library computing departments that resist involvement with their organization's broader IT infrastructure services and that prize the independence of the library's own computing resources. Many libraries have very positive working relationships with higher-level IT organizations and are generally satisfied with an arrangement that involves greater dependency on enterprise. The ability for a library to rely on these services depends on a high degree of trust, reinforced with formal agreements that ensure adequate levels of support and reliability.
Beyond these organizational issues, library automation products need to evolve and allow libraries to more eas- ily participate in the technical infra- structure of the enterprise. I see the service-oriented architecture (SOA) as one of the key factors that allow com- plex organizations to build an enter- prise infrastructure in a sustainable way. SOA relies on small, reusable components that can be applied across many different areas of functionality. Few organizations have evolved to the point of having a fully implemented SOA infrastructure, but this approach continues to gain broader acceptance as the preferred technological framework. In the relatively near future, I hope to see a new generation of library automation products designed as serviceoriented business applications.
The transition to SOA, if it happens, will take many years to play out. In the interim, libraries can take small steps to improve their positions within their organizations through technology. Through the use of web services and other open protocols, libraries can break out of their automation systems built as silos and extend their reach to other automation components throughout their organizations. Many tools and techniques are readily available today that can deliver library services and content to external applications. RSS, Atom feeds, and other XML application programming interfaces (API) can be creatively employed to dynamically push library content outside the applications directly managed by the library. Many libraries are finding ways to tap into institutionally provided data sources to avoid unnecessary duplication of effort. Single-sign-on systems that allow services provided by the library to use the same login credentials as other institutional resources, for example, are becoming more common. Every small measure of new interoperability can help a library be more plugged in to its parent institution, both technically and strategically.
When it comes to institutional relevancy, independence isn't a virtue. It's interdependence that makes a library a more valuable organizational asset.
|Type of Material:||Article|
Computers in Libraries|
|Volume 29 Number 06|
|Systems Librarian Column|
|Last Update:||2012-12-29 14:06:47|
|Date Created:||2009-09-11 08:02:10|