OLA Quarterly

Library Catalogs and Other Discovery Tools

Perspective on Catalogs The Evolution of Library Discovery Systems in the Web Environment The Library Catalog as Experimental Sandbox Reflections from Menucha LibraryFind™: The Development of a Shared Library Platform at State University Libraries The New Summit: Building the Foundation for Enhanced User Services Building Catalogs in the Sand Legacy Metadata and the New Catalog Northwest Digital Archives: Evolving Access to Archives and Special Collections in the Northwest A Usability Survey of Keyword Searching Using a University Library’s Catalog OLAQSpring 2009 Vol 15 • No 1 OLA Quarterly Oregon Library Association Spring 2009 http://www.olaweb.org Vol 15 • No 1 ISSN 1093-7374

Library Catalogs The OLA Quarterly is an official publication of the and Other Discovery Tools Oregon Library Association. Please refer questions 2 20 and input regarding Perspective on Catalogs The New Summit: John Repplinger Building the Foundation for Enhanced the Quarterly to: User Services Al Cornish Diane Sotak 5 The Evolution of Library Discovery OLA Publications Chair Systems in the Web Environment 24 5000 North Willamette Blvd. Mark Dahl Building Catalogs in the Sand Wade Guidry Portland, OR 97203-5798 10 503.943.8047 phone The Library Catalog 28 as Experimental Sandbox 503.943.7491 fax Legacy Metadata and the New Catalog Tom Larsen Richard Sapon-White [email protected]

Graphic Production: 13 31 Reflections from Menucha Northwest Digital Archives: Tobias Weiss Design Stephanie Michel Evolution Access to Archives and 7345 SW Deerhaven Drive Special Collections in the Northwest Corvallis, OR 97333 17 Jodi Allison-Bunnell phone 541-738-1897 LibraryFind™: [email protected] The Development of a Shared 36 Library Platform at A Usability Survey of Keyword www.tobiasweissdesign.com Libraries Searching Using a University Terry Reese Library’s Catalog Elizabeth Ramsey

Upcoming Issue Summer 2009 President’s Conference Issue

OLA Quarterly is indexed in Library Literature. Library Catalogs and Other Discovery Tools

ver the last few years, it has that local catalogs still play a valuable ing the role of usability testing in its become nearly impossible role in meeting user needs, allowing design and development. Archival and Oto read a library journal or libraries to present unique materials in other materials from special collections attend a library conference without creative ways that may not be possible are also the subject of Richard Sapon- reading or hearing about “next genera- in large-scale, shared systems. Stepha- White’s article. Noting that many of tion catalogs.” Many thoughtful and nie Michael sums up these varying per- these materials are not cataloged and vocal critics have pointed out that spectives in her overview of last fall’s therefore not findable using library library catalogs are hard to use, do not ACRL regional conference, which fo- discovery tools, he suggests some ways include information users need, and do cused on next-generation catalogs. The to tackle that problem. not measure up to the non-library sites next two articles describe a variety of Throughout all of these articles, our users frequent, such as Amazon projects intended to enhance access to one message comes through loud and and Google. Vendors and individual library materials. Terry Reese discusses clear: our systems must be built around developers alike have responded, devel- the development of LibraryFind™, an the needs and expectations of our users. oping new systems and new models to open source metasearch tool developed Users should not have to do things help users discover the treasures their at Oregon State University. Al Cor- the library way, using library search libraries contain. Some of this trailblaz- nish tells us about the Orbis Cascade syntax and vocabulary. Users expect our ing work has taken place right here in Alliance’s partnership with OCLC systems to be as easy as the non-library the Pacific Northwest. In this issue, we to develop WorldCat Navigator, the systems they use regularly. The goal feature the work of some modern-day product that now powers the Summit of user-centeredness, however, can be library pioneers who are performing union catalog. achieved in many different ways. Each the hard work required to take our In order to build user-centered library has unique collections and user retrieval systems in new directions. discovery tools, we need to determine needs, and this diversity is reflected in The articles in this issue range what users need and how they interact the variety of systems and models de- from the unconventional (or all- with our systems. Elizabeth Ramsey re- scribed in this issue. The key question encompassing) to the visionary to the ports on her work doing usability test- is, How do we create and implement specific and concrete. John Repplinger ing of keyword searching in the Con- systems that best meet the needs of our offers a plea for more intelligent, cordia University catalog, while Wade users? We hope you find both inspira- flexible catalogs that can adapt to the Guidry describes the changes made tion and food for thought in the pages changing needs of users. Mark Dahl to Beachbooks, the Coastal Resource that follow. shows us how we can help make this Sharing Network catalog, in response vision a reality by moving to network- to usability testing. Allison-Bunnell Guest editors level, global systems that benefit from tells us about the Northwest Digital Laura Zeigen & Janet Crum the participation of large numbers of Archives, a specialized retrieval system Oregon Health and Sciences users. Meanwhile, Tom Larsen argues for archival finding aids, emphasiz- University Library

1 Perspective on Catalogs

by John Repplinger y first experience with an Another good way for people and com- Science Librarian, electronic library catalog was puters to learn is through mistakes. Some Mark O. Hatfield Library, a “dumb” terminal at the local systems anticipate failed searches through M library. I searched for the book The Wizard spell check technology. A good example of Oz, and while I knew the author’s name of a failed search is a query that has too and the title, it remained frustratingly elu- many terms and would yield nothing, in sive in the catalog. With some experimenta- which case the Boolean search automatically tion, I discovered that my title search failed changes from “AND” to “OR” to broaden because I excluded “The” as the first word. the results. While catalogs continue to And my author search should have been, evolve and failed searches are monitored, the “Baum, L. Frank” instead of “L. Frank technology should consistently analyze the Baum.” At the time, I thought these were results to learn about common problems. truly “dumb” computers with rules that As data are compiled on recurrently failed were too stringent; I wondered how many searches, patterns emerge, and alternative people gave up in sheer frustration for not search strategies could be recommended being able to find what they wanted. while the patron types. Some Web browsers, Fortunately, our electronic catalogs like Firefox 3.0, already use this technology. have improved considerably since then, yet As our languages and cultures change there is still room for improvement. For over time, it is important to include new the next generation of catalogs, one thing words and phrases that our societies com- is paramount: they need to be increasingly monly use. Social tagging is one way to flexible to meet the changing needs of their accommodate these changes in language. communities. Some may cringe at the thought of public- Library catalogs should help patrons generated metadata in library records, but become better searchers. They need to within the right environment, social tagging predict errors, anticipate the needs of can be a powerful resource for the library patrons, and offer alternative search and an incredible way to include patrons. strategies that yield additional and higher BiblioCommons (http://bibliocommons. quality information. One way is to make com), is a new library-oriented “social the search process more interactive. It can discovery system” that allows patrons to tag be very difficult to articulate a complex records with keywords and comment on question, which is often where librarians library materials. Similarly, LibraryThing trump technology. A reference interview (http://www.librarything.com) encourages allows for quick feedback to occur be- users to tag books in their own collections tween a librarian and patron, as clarifying and explore tags utilized by other people to questions and answers are shared. Catalogs discover new books. The new Orbis Cascade could unobtrusively ask follow-up ques- Alliance catalog (http://summit.worldcat. tions to help clarify the search, such as org) is another example that takes advantage “did you mean this author or year?” As the of social tagging, in addition to allowing search unfolds, any new search algorithm patrons to write book reviews. employed should be clearly labeled for Reviews by patrons may not seem im- patrons to view. In the process, patrons portant at first glance, but including them is learn what information is useful and how an ingenious way to get patrons involved in 2 search strategies are “phrased” through your library and to gain valuable qualitative their search history. feedback about your collections. Since librar- Vol 15 No 1 • Spring 2009

ies have traditionally relied on usage statistics attitudes towards privacy (Stone 2008). It for collection development, this kind of suggests that while people cherish the idea information from patrons should be coveted. of privacy, they often let their guard down Patrons may see reviews as an invitation, that and provide information about them- libraries value their opinions and want their selves freely online. Libraries will need to feedback. They may also see reviewing as a consider what patrons want, in conjunc- Reviews by patrons way to give back to society, to help others tion with privacy concerns, as these new may not seem locate quality information. Book reviews will technologies develop. important at first probably be more popular with public librar- Libraries will also need to accom- ies whose patrons tend to read for entertain- modate small mobile technologies, such glance, but including ment, rather than academic library patrons as cell phones and personal digital assis- them is an ingenious that tend to read out of necessity. tants (PDAs). According to the 2007 Pew way to get patrons Technology should unobtrusively sug- Internet Research Project survey, 62 percent gest other materials, much like an electronic of U.S. residents are a part of a “wireless, involved in your library reader’s advisory. It should lead users to other mobile population that participates in and to gain valuable books of interest. The catalog could list a few digital [non-voice data] activities away from books from similar subject headings, nearby home or work” (Horrigan 2008). Some qualitative feedback call numbers, or even commonly checked technologies today, such as smart phones, about your collections. out library materials. For example, within provide full-featured Web browser that are the catalog record for “Harry Potter and the much more interactive than the clunkier cell Goblet of Fire,” five specific magic or fantasy phone Web browsers. Software applications titles could be displayed as suggestions, not for these mobile unlike Amazon.com. Catalogs already do devices are being this in a sense with subject headings; they released at an efficiently groups hundreds of similar items incredibly fast together. Unfortunately, most users are over- rate, including whelmed by lists of subject headings. Casual WorldCat.org’s users only want a few specific titles. recently-released Much like an electronic reader’s advisory, WorldCat Mo- patrons need improved current awareness bile application systems to track their favorite authors and (www.worldcat. genres. It would be wonderful if patrons org/mobile). could login to their library account, identify Libraries should their favorite authors/book series, and auto- explore how a matically be placed on a notification list when catalog interface new books are published. This type of service can be built to would be another way to encourage patrons make searching To make it happen, he needs you. to participate in collection development. faster and easier As an information specialist, you do more than connect individuals Privacy issues arise as libraries make to use with these to publications. You help them find inspiration. As the leading information services provider, EBSCO can help you do it. We reader’s advisory and reviews available for on-the-go provide information management systems that free up your time patron use. In both cases, patrons could technologies. so you can focus on your users. leave a trail of personal information about We also need to themselves. The New York Times recently anticipate the ran a story on an upcoming study from the type of digital ebsco.com 3 Carnegie Mellon University about people’s content that OREGON LIBRARY ASSOCIATION

people will want with these new devices, search for not only books, but for movies, such as audio and streamed video. journal articles, current news, and more Within shared consortial catalogs, it through one interface. WorldCat.org for is common to run across a list of identical example is able to search for an increasing titles with slightly varying editions, media array of formats such as articles and Inter- formats, special notes about donors, etc. It net resources, but has issues with limiting can be confusing for patrons to figure out to local collections. which item they want. Let’s simplify the re- These features are a step in the right sults for our patrons by consolidating local direction, yet there is still a long way to go metadata into one general record that can for all catalogs. It is exciting to see many be shared among consortia members, with of these issues being addressed in the lat- the option of displaying local metadata at est platform releases. By becoming more the click of a button. interactive, catalogs can adapt more fluidly Then there is the issue of searching for to the changing needs of patrons. Let’s meet an author. Most catalogs still use the rigid the patrons where they are (on-the-go), format of last name, first name and middle accommodate how they search, utilize the initial. Catalogers use the strict standards language they commonly use, and provide for consistency, but it seems reasonable for a the formats they need. catalog to include the natural language form of an author’s name, such as “Joe A. Smith” Bibliography in addition to “Smith, Joe A.” Some catalogs Horrigan, John. 2008. Mobile Access to Data are capable of searching both forms, but and Information, Pew Internet & American this feature is often buried. Why not make Life Project. http://www.pewinternet.org/ this the standard author search, and make pdfs/PIP_Mobile.Data.Access.pdf (accessed the traditional (stricter) author search the December 2, 2008). advanced option? Patrons want a convenient one-stop Stone, Brad. “Our Paradoxical Attitudes center that allows them to search for a vari- Toward Privacy.” New York Times Tech- ety of information and to access it quickly nology Blog, July 2, 2008. http://bits. either electronically or in print. Traditional blogs.nytimes.com/2008/07/02/our-par- catalogs have been wedded to the book, adoxical-attitudes-towards-privacy/index. but patrons want to search for everything a html?th&emc=th (accessed December 2, library has to offer. One should be able to 2008).

4 The Evolution of Library Discovery Systems in the Web Environment

by Mark Dahl n December 2008, the Orbis Cascade broadly, e-commerce gained ground and Associate Director for Alliance, a consortium of academic people got used to shopping experiences that Digital Initiatives libraries in Oregon and , involved search, discovery, and fulfillment. and Collection Management, I Aubrey R. Watzek Library, launched a new union catalog on OCLC’s In 1998 Google was founded, and by Lewis and WorldCat.org platform. This change the early 2000s it was the most popular resulted in an updated Web interface, bet- search engine on the Internet. Google’s ter keyword searching, and faceted results. clever PageRank algorithm harnessed the However, we also lost some features that collective intelligence of the Web by using worked well in our old system. But the hyperlinks to help determine relevancy. It larger significance of this change might not was a system that benefited enormously be obvious. A shift has taken place, one that from the sheer scale of Google’s computing moves us into a new paradigm for the sys- power. More importantly, it got smarter as tems that support discovery of resources in more people used it. Google proved that a libraries. The Summit catalog is now part of Web scale enterprise could achieve things a great global organism known as WorldCat, that small- and medium-sized players could and that organism is poised to be more dy- not. In a similar way, dot-com crash survi- namic and more ubiquitous than any of our vors like eBay and Amazon established that old local catalogs could have ever been. How in certain markets there was only room for did we get here? I will attempt to answer a few large players on the Web. that question through my personal account While Google was growing its search of library search and discovery as a librarian business, libraries mostly ignored search and technologist since the mid-1990s. and worked on the problem of organiz- I entered library school in 1996. As the ing a growing array of full text resources. Web emerged, I developed a growing curios- Libraries were acquiring access to electronic ity for it and delved into HTML coding, journals by the bucketful, but it was hard Web programming, and Web server admin- to find out if a given library had access to a istration. In those early days, the library particular journal. By 2001, I had moved to community was just digesting the obvious Watzek Library at Lewis and Clark College, advantages that the Web had over previous and one of my first tasks was to develop technologies like Gopher and Telnet: mouse a way to search our electronic and print click hyperlinking and richer graphics. The journals by title. In response I created a underlying discovery systems libraries used database that mixed together data from our continued much as they had in the past with ILS and Serials Solutions and would later prettier Web-based interfaces on top. support an OpenURL resolver. By the late 1990s some transformative In the early to mid-2000s, library changes began to take shape in the online catalogs began to adopt more of the trap- library world and on the Web. In the library pings of mainstream e-commerce sites by world, full text databases and services like incorporating cover art, external links, and JSTOR arrived on the scene, putting large fancier Web design. They remained weak in amounts of actual content, not just indexing, search functionality. In 2005, major figures online. The general online fulltext database in the library technology community like became the bread and butter of our online Andrew Pace and Roy Tennant began ask- offerings at Central Oregon Community ing rather loudly why OPAC College, which we were positioning to sup- search left so much to be 5 port distance education. On the Web more desired when compared OREGON LIBRARY ASSOCIATION

tion of disparate content and systems with Web technologies could create exceptional online services for libraries. We argued that library systems, including discovery sys- tems, would be many dis-integrated units tied together by standards and clever Web programming. Modular digital library tools like OpenURL resolvers, electronic resource management software, and digital asset man- agement software, the trend toward OPACs running atop ILSs, and federated searching systems that relied on new standards like SRU/W (search/retrieve via URL or Web service) all seemed to confirm this thesis. The new Summit catalog has relevancy ranking based in part on library hold- But as we researched the book in late ings as well as next generation catalog features like facets. 2005, it became clear that this model did not explain it all. More and more, users were with commercial Web search (Pace 2005; beginning to encounter library resources on Tennant 2005). the Web outside the “walled garden” context Projects emerged that attempted to of library-managed discovery systems. People significantly improve search functionality in might discover books on Amazon or articles Web OPACs. They included North Carolina on Google Scholar and then acquire the con- State University Library’s catalog based on tent via a library’s physical or virtual gateway. the Endeca search engine and Casey Bisson’s Moreover, Web 2.0 sites like Flickr, del.icio. WPopac (now Scriblio), an OPAC based on us and YouTube allowed users to contribute the modular WordPress blogging software. and organize digital assets in a collective In the early 2000s libraries also began fashion. Like Google, these Web 2.0 sites got to break important new ground with digital better as more people used them and aspired collections mounted on systems such as to a Web-wide audience. ContentDM and DSpace. These were the In April 2006, I heard Lorcan Dempsey first Web-based discovery systems managed of OCLC give a presentation to the Orbis by libraries that harnessed the Web’s global Cascade Alliance Council on “Moving to reach. Library catalogs largely contain refer- the Network Level: Libraries, Readers, and ences to books held by hundreds of libraries Applications.” Dempsey discussed the shift and are typically closed to search engines from vertically integrating services within because of the redundancy of their data. By a single institution to “collaboratively contrast, digital collections contain unique sourcing” services in concert with external materials and are generally open to search players. The Alliance’s own union catalog, engines, allowing people anywhere on the which aggregates supply and demand for globe to find and use their content. books among 30+ academic libraries, served In late 2005 and early 2006 I co-au- as a strong example of regional collabora- thored a book, Digital Libraries: Integrating tion. Dempsey encouraged the group to Content and Systems, with Kyle Banerjee and broaden its thinking to resource sharing 6 Mike Spalti. We started work on the book that would involve “multi-level” collabora- with a loosely-conceived thesis: that integra- tion between individual libraries, regional Vol 15 No 1 • Spring 2009

consortia and global players like OCLC, but with discovery and delivery features JSTOR, and Google. He challenged the tailored to the needs of the Alliance. group to think about “painful” activities Given the growing shift in my thinking, being done at the local or regional level that I saw several advantages in the Alliance move could be more effectively done by higher to WorldCat. The interface is more modern level organizations and systems. than the old Summit and offers conventions In some respects, the idea of outsourc- from the consumer Web such as narrowing ing library systems to larger-scale play- searches by facets and creating user accounts ers went against my instincts. I’d always for favorites. More compelling, however, is enjoyed managing my own servers and the broader concept of having a catalog that is writing my own Web applications. There a part of a larger organic whole. The World- was something inspiring about being able Cat database is a dynamic, ever evolving to load Linux on an old PC and run my thing, updated by a global community of very own Web presence from that little box catalogers. Unlike our local catalogs, where we humming away in the closet. download records and they remain mostly un- Nonetheless, I couldn’t get the “moving changed like a card in a card catalog, World- to the network level” phrase out of my head. Cat operates like a Web 2.0 site: a community In late 2006 and 2007, I discovered that the of people can cooperatively add metadata idea related to the various Web applications to improve digital objects, albeit in a much that I began using at work and in my person- more regulated, library-world way. WorldCat’s al life. Gmail revolutionized my productivity global, ever changing holdings information at work. I benefited from its great search and allows WorldCat.org to have an unparal- organization features, powered by Google’s leled relevance ranking of books, not unlike huge infrastructure far away from my PC. At Google’s PageRank concept. The WorldCat. Watzek Library, we began using Basecamp org platform also supports user-contributed and Google Docs for project management content like ratings and reviews, a service and collaboration. At a time when I support- that will be progressively more useful as more ed a collection of digital images for teaching libraries and users come on board. on MDID digital collections software, I was Moreover, with WorldCat.org, OCLC impressed with how much better Flickr man- takes a lesson from Google and Amazon aged digital assets. Meanwhile, buzz around and understands that Web scale matters. the concept of cloud computing grew, In order for library content to be noticed especially with the publication of Nicholas on the Web, it needs to be presented by a Carr’s The Big Switch in early 2008, which global player, not in a diluted fashion from explains how computing power in far-away thousands of separately managed library data centers is revolutionizing both personal catalogs. Unlike local library catalogs, computing and back-end IT infrastructure. WorldCat.org provides a place to reference In 2008, our library began implement- a book that is useful for anyone on the Web ing two network level discovery services. and maintains relationships with commer- In winter 2007/2008, the Alliance struck a cial search vendors so that its records will deal with OCLC to create a union catalog appear in search engine results. Further- solution based on the WorldCat.org more, it provides a catalog with common platform. WorldCat Navigator is a consortial conventions for searching and viewing re- version of WorldCat Local that provides a cords not unlike Google providing a certain 7 catalog with the wide scope of WorldCat.org consistency in its interface across the Web. OREGON LIBRARY ASSOCIATION

As Watzek Library threw its weight behind the Alliance WorldCat project, we got another innovative network-level initia- tive underway. Our visual resources curator, The challenge of library Margo Ballantyne, and a faculty member in technology and metadata Ceramic Arts saw an opportunity to create professionals will move from an online image collection of contemporary ceramics. The challenge would be collect- managing a library’s own set of ing the images and metadata from artists isolated databases to managing dispersed throughout the world. The Digital Services Coordinator, Jeremy McWilliams, their library’s imprint on shared and I were avid users of Flickr and knew of global discovery platforms. its powerful Web-based tools for managing images. With little money behind the project These will be systems that benefit from the for staff support, we came up with the idea of network effects allowed by Web scale: they having artists contribute images and meta- will get better as people and organizations data and assign copyright through their own use them and contribute to them. The chal- Flickr accounts. We would then assemble the lenge of library technology and metadata images in a Flickr Group and present them professionals will shift from library manage- as a coherent digital collection via a Web site ment of isolated databases to managing their driven in part by the Flickr API. We imple- library’s imprint on shared global discovery mented this idea in the spring of 2008, albeit platforms. Libraries will still strive to pro- with some technical modifications to our vide specialized interfaces and metadata for initial vision (McWilliams 2008). their users, but the work will be done in this This site, http://accessceramics.org, is new global context. a live, growing collection of contemporary If a library develops a special vocabu- ceramics images that reside in individual lary for a subset of its collection, it will add Flickr accounts but are organized together the terms to a global database so that this into a digital collection with a defined set of vocabulary, however esoteric, can have a metadata. In contrast to digital collections broader benefit. With our likely move to that are cataloged centrally, our metadata WorldCat Local here at Watzek, I’ll en- is entered by the contributors. We found courage our cataloger to start adding genre some similarities to this model in the digital headings for videos on WorldCat instead of history projects launched by the Center for doing the work in our local system. Rather History and New Media such as hurrica- than sweating out upgrades to library-man- nearchive.org. We also found affirmation in aged OPAC software, we will enjoy World- our selection of Flickr when the Library of Cat Local’s “software as a service” model Congress launched a collection of images in that assures it is being constantly improved the Flickr Commons in 2008. and upgraded, just like Gmail. When we These recent experiences have convinced feel the need to customize, we’ll use APIs to me that a new model for library discovery create interfaces tailored to our user com- systems may be emerging, one character- munities. We’ll also take the opportunity ized by global discovery systems like Flickr, to mash up data from multiple sources on 8 WorldCat.org, and new ones yet to surface the network. For example, Watzek recently in both the profit and non-profit sectors. created a proof of concept mashup with the Vol 15 No 1 • Spring 2009

WorldCat API and the Google Book Search API that creates a Google Books search with library holdings in the result set. This platform shift should benefit smaller librar- ies like Watzek, who will now have access to a search and discovery infrastructure that is as good as that used by the big players. Hopefully, these shared platforms will spark new innovations in collections and services by both small and large libraries. The movement towards network-level discovery systems for libraries is emerging in an uneven manner typical of new tech- nologies. I welcome the complexity, chaos, and change. As has been the case in the recent past, much of our job will be manag- ing change for our user communities, both accessceramics.org is a collaboratively-sourced digital collection created with artist-con- technically and via communication with our tributed images and metadata. It uses Flickr as the underlying digital asset management constituents. These network level systems system. should make it easier to do basic research and access common material. User expec- Dempsey, Lorcan. Moving to the network tations for more specialized materials and level: Libraries, readers and applications. services should increase. Whereas we have 2006 [cited 12 February 2008]. Available historically concentrated most of our energy from http://www.oclc.org/research/presen- on commonly published material in familiar tations/dempsey/cascade.ppt. forms, in this global discovery environ- ment we may find ourselves working at the McWilliams, Jeremy. 2008. Developing an extremes. We will be curating physically and Academic Image Collection with Flickr. digitally what we have that is unique and of Code4lib Journal (3) (6 June 2008). Avail- interest globally, as well as assisting with new able from: http://journal.code4lib.org/ forms of intellectual output that don’t neatly articles/74 fit the book or periodical categories. Wherever we end up, it should be a Pace, Andrew K. 2005. My Kingdom for an good ride. OPAC. American Libraries 36, (2) (Febru- ary): 48-9. Available from: http://www.ala. References org/ala/alonline/techspeaking/2005colunms/ Carr, Nicholas G. 2008. The big switch: techFeb2005.cfm Rewiring the World, from Edison to Google. New York: W. W. Norton & Co. Tennant, Roy. 2005. “Lipstick on a pig”. Library Journal 130, (7) (April 15): 34. Dahl, Mark, Kyle Banerjee, and Michael Available from: http://www.libraryjournal. Spalti. 2006. Digital libraries: Integrating com/article/CA516027.html Content and Systems. Oxford [UK]: Chan- dos Pub. 9 The Library Catalog as Experimental Sandbox

by Tom Larsen or centuries libraries have made use A number of these “next generation” Database Management of catalogs in one form or another catalogs are being developed, all of which and Catalog Librarian, as a register of the bibliographic show great promise, and none of which Portland State University Library F entities found in a particular library (New is entirely without flaws. An interesting World Encyclopedia contributors 2008). family of new catalog interfaces comes The modern online public access catalog from OCLC and is based on the WorldCat (OPAC) has expanded this concept some- union catalog. OCLC’s WorldCat database what to include not only items owned by contains over 125 million bibliographic the library, but also items freely available records with the holdings of over 10,000 li- via the Internet that have been deemed by braries around the world. OCLC has devel- A number of such “next library staff to be of value to their patrons. oped three new interfaces to this database. generation” catalogs The modern OPAC also improves on previ- WorldCat.org provides an interface are being developed, ous catalogs with enhanced search capa- that displays search results in such a way bilities and by being accessible from any that it guides users to the nearest library all of which show great computer with an Internet connection. that holds the item in question, then to promise, and none The recent advent of concepts such as progressively more distant libraries, thus al- “Web 2.0” and “Library 2.0,” though still lowing the user to find the quickest way of of which is entirely nebulous, illustrates that users no longer obtaining the item from a library. Further- without flaws. think of the library as their only, or even more, certain Web services (e.g., Google their primary, source of information. As Books) allow the user to pass their Web pointed out by Coyle (2007a), “The ques- search on to WorldCat.org to find a copy of tion today is not how do we get users into the item at a nearby library. the library, but how can we take the library WorldCat Local has the added feature to the users. The answer will necessar- of local branding. The holdings of the lo- ily involve a transformation of the library cal library are displayed first, followed by catalog.” Users rarely begin information the holdings of other libraries in the local searches in the library catalog, tending more library’s consortium, followed by other often to start searching the Web using a WorldCat libraries. For many purposes, popular search engine such as Google. The WorldCat Local can serve as the primary question becomes not one of what resources interface to the local library’s collections can be found in the library but rather one (as opposed to the local library’s own Web of what resources are available anywhere OPAC) with the added feature that the and how can one obtain them (Coyle holdings of other libraries are also dis- 2007b). Consequently methods are being played. This feature is useful if the local developed to pass Internet searches on to copy is unavailable or does not exist. a library catalog when appropriate. At the WorldCat Navigator is being devel- same time, new services are being developed oped for Summit, the union catalog of in library catalogs to send a user’s search the Orbis Cascade Alliance. It is similar to beyond the local catalog into the catalogs WorldCat Local except that it is branded of other libraries, into electronic databases, for Summit and displays the holdings into digital repositories, or even into the of Alliance libraries first followed by the Web. In addition, services are being devel- holdings of other WorldCat libraries. oped that allow users to provide value-add- WorldCat Navigator also allows patrons to 10 ed content in the form of tags, reviews, etc., borrow items directly from other libraries thus making the catalog more interactive. in the consortium. Vol 15 No 1 • Spring 2009

None of these new interfaces is without University, we have a history of experiment- problems. For example, the bibliographic ing with our catalog to develop new services records that display in WorldCat interfaces for our patrons. One of our more successful are based on the OCLC master record, so experiments involved exploiting the capabili- any notes (as well as other fields) that may ties of the Electronic Resources Manage- appear, in a library’s local records will not ment module from Innovative Interfaces to appear or be searchable, in the WorldCat integrate one of our digital archival collec- displays. This is particularly troublesome tions into our database, and then to allow for libraries with extensive special collec- our patrons to navigate through the various tions, since these fields are often crucial for hierarchical levels of the collection (Brenner identifying unique copies of rare materials et al. 2006). This system mimics some of (Allison-Bunnell et al. 2008). Furthermore, the hierarchical characteristics of a finding any authority work done in the local cata- aid, yet it consists of sets of linked MARC log is potentially lost in WorldCat unless records which can be searched by author, OCLC has also done that authority work title, subject, etc., with the other biblio- on their master record. There are many graphic records in our catalog. These records records in WorldCat for which such author- also contain links to the digital objects ity work is sorely lacking. OCLC is aware themselves. This system has greatly improved of these problems, however, and hopefully access to this collection. The important point they will find solutions to them. here, however, is that this experiment would The highly networked nature of infor- have been impossible without a local catalog. mation resources, and the decreased need Another experiment we performed in for users to be within close proximity to our local catalog involved the inclusion of resources, have led some to wonder why non-Roman scripts (Chinese, Japanese, libraries even need their own bibliographic Korean, Arabic, and Hebrew) in author- database and user interface (Coyle 2007b). ity records. We discovered this had certain Certainly it seems redundant for a library’s beneficial consequences for searching in our holdings to be represented in WorldCat and catalog. When the vernacular script form of the local database, which sometimes entails a name was entered into a 4XX or 5XX field duplicated effort. While future improve- in the appropriate authority record, a search ments in WorldCat may make it possible using the vernacular script, retrieved records for libraries to abandon their local catalogs with the vernacular form of the name, and and rely solely on a union catalog, this does through the cross references provided ac- not appear to be totally feasible at the pres- cess to records that contained the name in ent time. In addition to problems with re- Romanized form. cords, many libraries use their local catalogs This functionality would benefit our to inventory items such as study room keys, patrons, but this project was not carried laptops, their dark archives, and other items through to completion because the Library that are not appropriate for a world-wide of Congress began including vernacular union catalog. In some cases items (e.g., scripts in their authority records, which those in dark archives) should not even be then appeared in the OCLC authority file. visible to anyone other than library staff. Currently, it does not appear to be possible Another reason why some libraries might in WorldCat to retrieve Roman-script-only want to maintain a local catalog is to have records when searching using the vernacular 11 an experimental sandbox. At Portland State script. Consequently, this type of search still OREGON LIBRARY ASSOCIATION

works best in our local catalog. Again, we FinalReportWCLSpecCollTaskForce.pdf would have never been able to even explore (accessed 16 February 2009). the possibilities if we did not have a local catalog in which to explore them. Brenner, Michaela, Tom Larsen, and Claudia These are exciting times, and it will be Weston. 2006. Digital collection manage- interesting to see what developments will ment through the library catalog. Informa- happen in the world of library catalogs and tion technology and libraries 25 (2): 65–77. … we would have never other information discovery systems. Nev- ertheless, we are not quite ready to abandon Coyle, Karen. 2007a. The library catalog been able to explore our local catalog and throw the proverbial in a 2.0 world. The journal of academic the possibilities if we baby out with the proverbial bath water. librarianship 33 (2): 289–91. did not have a local On the other hand, though, as the baby grows up, it would do well to be flexible. Coyle, Karen. 2007b. The library catalog: catalog in which to some possible futures. The journal of aca- explore them. References demic librarianship 33 (3): 414–6. Allison-Bunnell, Jodi, Nichole Bouché, John Chapman, David de Lorenzo, Jackie New World Encyclopedia contribu- Dooley, Elizabeth Johnson, Mela Kircher tors. 2008. Library catalog. In New et al. 2008. Report of the WorldCat Lo- World Encyclopedia. http://www.new- cal Special Collections and Archives Task worldencyclopedia.org/entry/Library_ Force. http://www.rbms.info/committees/ catalog?oldid=751788 (accessed 12 bibliographic_standards/committee-docs/ January 2009). Dream. Discover. Do! Spokane Falls Community College library technician program offers:

• Online A.A.S. degree option • Library technician certificate option • School Library Media emphasis option • Bachelor’s degree transfer option

Students also can take advantage of college credit for work experience, independent study and internships.

For information contact Paula Duffy Swan, library technician program at 509-522-3809, or [email protected].

www.spokanefalls.com

12 COMMUNITY COLLEGES Community Colleges of Spokane provides equal opportunity OF SPOKANE in education and employment. DL

OLA Quarterly Library Tech Ad — 2009

Updated: January 13, 2009 Reflections from Menucha

by Stephanie Michel ibrarians from across Oregon and It’s all about the user. Reference/Instruction Librarian, Washington gathered on October A fundamental concern for user needs and University of Portland 23 and 24, 2008, in the beautiful preferences was a common thread. The (and Past President L of ACRL-Oregon) autumn sunshine at the Menucha Retreat WorldCat Local panel stated at the begin- and Conference Center for the ACRL ning of their presentation that ultimately, Oregon/Washington fall conference. The it’s all about the user; later they discussed theme, “The Once and Future Catalog,” their adoption of the Google mantra, offered insight into exciting possibilities for “trust the user.” Terry Reese focused on the next-generation library catalogs, as well as need for one interface per user, meaning discussion of issues to consider and obsta- that each user could customize the catalog cles to overcome. Speakers from around the interface to his/her own preferences. He Northwest and across the country discussed also energized the audience by suggesting their experiences with cutting-edge library that the OPAC should be the least used catalogs, including: interface, questioning why users should need to leave their own familiar online • Kristin Antelman, Associate Direc- environments and come to our system to tor for the Digital Library at North find information. Kristin Antelman’s and Carolina State University (NCSU) Li- Tim Daniels’ presentations also addressed braries, presented the NCSU Endeca- seeking user input and gaining user buy-in based library catalog and the Triangle during the planning process for their new Research Libraries Network (TRLN) systems. From these presentations, it be- consortial catalog. came clear that user needs should be central to the planning process for next-generation • Tim Daniels, formerly the PINES Pro- library catalogs. gram Manager with the Georgia Public Library Service (GPLS), discussed the Considering all the options. Evergreen-based PINES system. The speakers stated that they investigated various products before determining which • Terry Reese, Digital Production Unit option would best meet their needs. Each Head at Oregon State University, ad- institution’s choice reflected their main dressed regional opportunities for next- priority for system functionality: discovery generation library catalogs. (search interface) or fulfillment (delivery system). NCSU and GPLS prioritized dis- • Steve Shadle, Serials Access Librarian; covery, choosing Endeca and Evergreen for Anne-Marie Davis, Reference Librarian/ their user-friendly interfaces. UW consid- Collection Development Coordinator; ered Endeca and Encore, which offer better and Kathleen Collins, Reference and discovery, but ultimately selected WorldCat Instruction Librarian from the Uni- Local for its fulfillment system. versity of Washington (UW) Libraries Terry Reese made an interesting ob- discussed implementation and public servation that “search is easy; discovery is services issues with WorldCat Local. hard.” If you give a user a search box, they will find something. But will they find the Several themes emerged from the confer- best or the most relevant ence session and invigorating discussions. search results? The chal- 13 OREGON LIBRARY ASSOCIATION

lenge of next-generation library catalog Collaboration. systems is to optimize discovery, to have Now more than ever, collaboration within a robust system on the back end that can our own libraries and among regional part- connect users to the information they need ners is vital to success, not only of library while offering an intuitive interface that any catalogs but of many library services. The user can easily navigate. GPLS system provided a model of col- laboration; their 275 member libraries Streamlined decision-making. participate in the system at no cost to the Kristin Antelman, Tim Daniels, and the member libraries. Their catalog is centrally- UW panel discussed the importance of a administered, and a statewide library card streamlined and efficient implementation allows users to check out and request team in order to complete the project quickly. materials from any member library across Kristin Antelman commented that if they the state. NCSU also lauded their TRLN had involved all potential stakeholders, they consortial catalog, which allows users to would still be talking about the project years receive items from participating libraries later, rather than having accomplished it. In within one day of the request. The geo- addition, the project needs sufficient sup- graphic proximity of their libraries makes port resources, including staff and funding. this arrangement more feasible. Another key element in the success of the UW noted user demand for materials NCSU and PINES catalogs was the use of from other libraries increased dramati- project managers who oversaw and directed cally after they adopted WorldCat Local. the implementation process. A small, well-or- Summit borrowing increased 59 percent, ganized and adequately-supported implemen- and interlibrary loan borrowing increased tation team helped these projects to achieve 101 percent. As patrons use more materi- their goals quickly. als owned by other libraries, the need for each individual library to own every book Communication is key. is reduced, opening the door to cooperative Speakers acknowledged that communi- collection development. Building collec- cation, both during and after a project’s tions cooperatively with regional partners implementation, is crucial to its success. will allow libraries to reduce duplication Library staff and patrons need to be kept in their collections and free their limited up to date about implementation timelines resources to purchase unique materials that and ongoing improvements to the system. benefit the entire consortium. Additionally, all users should have the op- Collaboration needs to happen to a portunity to provide feedback or to report much greater degree, according to Terry problems with the system. Tim Daniels Reese. He predicted the end of the local in- discussed GPLS’ outreach to member librar- tegrated library system (ILS), and suggested ies, which resulted in a collective sense of that the ILS will move to the network level. ownership and pride in their library catalog. A network-level ILS would be centrally UW librarians learned from their experience administered, either by a consortium or a that communication is vital to positive staff vendor. Participating libraries could utilize perception of the project. Lack of com- the full functionality of the ILS without the munication can leave staff feeling like their responsibility for maintaining the system, 14 input is not valued and might cause nega- thereby reducing costs and duplication tive perceptions of the system that linger in staff time and expertise among librar- after the implementation. ies. For users, a move to the network level Vol 15 No 1 • Spring 2009

would provide a consistent catalog interface tests gather feedback from our core users to across multiple libraries and might lead to ensure that our decisions meet their needs a statewide library card program similar to and expectations. the Georgia Public Library Service. UW implemented WorldCat Local with the idea that if it didn’t work for their Persistence. users, they would turn it off, despite the Persistence was a key factor in the success of significant investment of time and resources. all the projects. Terry Reese stated it best by Another WorldCat Local library did exactly asserting, “We succeed by committing to a that. The Peninsula Library system in Cali- solution and making it work.” The speakers fornia was another early WorldCat Local noted that their projects were accomplished adopter; however they turned off WorldCat through hard work and dedication on the Local almost immediately due to user com- part of their project teams, who overcame plaints about their inability to limit a search the obstacles, implemented the system, and to a specific branch library. Although UW’s persisted after the implementation to trouble- WorldCat Local has this same limitation, shoot issues and make the system a success. they have overcome this issue by maintain- ing two separate library catalogs: the former Usability (or, it really is all Innovative Interfaces library catalog, which about the users). offers more advanced search functionality, as Usability, for patrons and for library staff, well as WorldCat Local. was a theme throughout the conference. User feedback about UW’s WorldCat Once libraries have implemented the new Local catalog was primarily positive; nega- system, how do they determine if it meets tive feedback usually focused on specific their goals? Can users efficiently use the sys- problems rather than broader complaints. tem to locate and borrow materials? GPLS UW also made an interesting distinction created a new position focused on usability; between the types of negative complaints this person will be responsible for conduct- that they received: they classified the issues ing usability tests with staff and patrons at as transitional (a result of users’ learning member libraries. NCSU conducted focus curve with the new system), or functional groups which shaped the development of (actual problems with the system). Not all the TRLN system; Kristin Antelman noted, complaints indicate that something is wrong “too often, our opinions are arbitrary and with the system; instead some offer a teach- driven by the people involved.” Usability able moment in which users can be taught how to efficiently utilize the new system. After the implementation, and ideally on an ongoing basis, libraries need to ask their users if the system is working for them “…too often, our opinions are and how it can be improved. Users, expec- arbitrary and driven by the tations, and research methods change; the people involved.” library catalog (or any other system) needs to be responsive to this changing environ- —Kristin Antelman, ment and continue to adapt to users’ shift- Keynote speaker, ing needs and preferences. Throughout the conference, speak- 15 ACRL Menucha 2008 ers energized and engaged the audience by encouraging us to think imaginatively OREGON LIBRARY ASSOCIATION

about what the library catalog could be. References How can we jazz up the catalog interface Antelman, Kristin. “Endeca at NCSU: and offer features that our users expect and What Have We Learned?” Keynote address, utilize in other online applications? How ACRL-Oregon/Washington Fall Confer- can the back end of the system offer better ence, Corbett, OR, October 23–s24, 2008. functionality for fulfillment and delivery http://zuiprezi.com/prezi/819/view/#76 of materials? How can we incorporate the lessons learned from conference speakers. Daniels, Tim. “Georgia Library PINES.” These lessons are: Presented at ACRL-Oregon/Washington Fall Conference, Corbett, OR, October • investigate all of our options 23–24, 2008. http://www.nwcentral. org/?q=node/1122 • identify our project goals Georgia PINES Catalog. http://gapines.org/ • appoint streamlined implementation opac/en-US/skin/default/xml/index.xml teams North Carolina State University Libraries On- • seek input and gain buy-in from our line Catalog. http://www.lib.ncsu.edu/catalog/ users Reese, Terry. “Regional Issues and Options • establish clear communication methods with Next Generation Library Catalogs.” for reporting successes as well as problems Presented at ACRL-Oregon/Washington Fall Conference, Corbett, OR, October • collaborate with regional libraries 23–24, 2008. http://www.nwcentral. org/?q=node/1108 • be persistent on our path to success Shadle, Steve, Kathleen Collins and Anne- • remember, throughout the process, to Marie Davis. “WorldCat Local at the Uni- keep our focus on the user. versity of Washington.” Panel presentation at ACRL-Oregon/Washington Fall Confer- This conference provided a great start ence, Corbett, OR, October 23–24, 2008. to the conversation on how our individual libraries and our regional consortium can Triangle Research Libraries Network. shape the future of library catalogs. http://search.trln.org/

University of Washington Libraries World- Cat. http://uwashington.worldcat.org/

16 LibraryFind™: The Development of a Shared Library Platform at Oregon State University Libraries

by Terry Reese he digitization of academic resources how software and services should be devel- Gray Family Chair has been a boon for library patrons, oped. For example, O’Reilly argues for the for Innovative Library Services Tproviding library users with access idea of “perpetual beta,” treating users as to resources not available in earlier years. co-developers in the development process. If However, the advent of Internet discovery O’Reilly is right, and Web 2.0 philosophi- tools like Google and Yahoo have led us- cally represents this world where boundaries ers to expect simple yet richly interactive do not exist, then for librarians the hallmark systems that transparently facilitate access of a Web 2.0 system cannot simply be a user to information. And it has been the library interface improvement. Rather, librarians community’s challenge to develop systems need to look beyond a system’s surface im- Librarians need to look and strategies for meeting changing user provements to evaluate the system’s ability to beyond a system’s needs. This has meant a closer look at our cross boundaries and provide opportunities surface improvements own culture and a need to rebuild ourselves for interoperability. as more nimble organizations with the abil- Web 2.0 systems create new access to evaluate the ity to quickly shift directions and develop points to encourage users to exploit and system’s ability to cross more transitive services. While many create tools from available data. The Web libraries continue to talk about the need to 2.0 philosophy of data interoperability re- boundaries and provide become more like Web 2.0 organizations, flects one of the library community’s tradi- opportunities for few have actually started to move in that tional core values. When the Oregon State interoperability. direction. University (OSU) Libraries first started However, before moving forward, we developing the LibraryFind™ application, should really step back and define some developers focused on how the applica- of the vocabulary that will frame the tion would encourage and promote greater conversation—specifically how Web 2.0 interoperability with the Libraries’ informa- and metasearch are being defined. Many tion resources. While LibraryFind’s™ user of the discussions around Web 2.0 center interface (UI) reflects many of the tech- around technology and functionality, which nologies associated with Web 2.0 the aim are an outward manifestation of the Web of the LibraryFind™ was to embrace the 2.0 philosophy. The focus on this aspect Web 2.0 philosophical mantra and build a often leads to confusion as dialogue about tool that could function as a component of Web 2.0 bogs down in debating feature sets a larger unified library platform. rather than the guiding principles of the In addition to Web 2.0, the concept of Web 2.0 movement. Like many communi- metasearch is a term that is often misun- ties, libraries have become enamored with derstood or used interchangeably with much of the functionality that has come out federated search. For the purpose of this of the Web 2.0 movement (tagging, facet- dialogue, federated search will be used to ing), but have failed to fully understand describe a search done over many differ- the movement’s foundational principles. As ent resources, where the query runs on Tim O’Reilly writes in his article, “What is numerous remote servers before results are Web 2.0,” the driving force behind the Web aggregated and returned to the initiator of 2.0 concept is one of no boundaries. When the search. Metasearch, on the other hand, the Web is envisioned as the development will be defined as a search that is able to platform, traditional rules and boundaries query and aggregate content simply no longer apply. Out of that vision from local and remote 17 comes new rules and philosophies related to indexes. Metasearch’s abil- OREGON LIBRARY ASSOCIATION

ity to aggregate content comes much closer formation landscape. This enabled OSU to the classic Web search engine model, Libraries to, for the first time, offer a single where a single index is utilized to query a set of Application Programming Interfaces large cache of information. While federated (API) to departments and patrons looking search tools have no such central index and to develop connections to library resources. must rely solely on their ability to retrieve Also, LibraryFind™ provides the Libraries’ data from a wide variety of sources, metase- developers with a unified API for building arch utilizes an approach that combines other patron services. This development has both harvested and remote collections to- allowed the OSU Libraries to take a more gether. The distinction is an important one active role in defining and developing our O’Reilly argues for due to the inherent limitations present in own search infrastructure, as well as develop the idea of “perpetual federated searching. Since federated search a product that could potentially benefit queries multiple remote servers, a certain the greater library community as a whole. beta,” treating users amount of built-in latency will always be While a discussion of LibraryFind’s™ cur- as co-developers in the present as the search tool communicates rent functionality or roadmap for develop- development process. with its various remote servers. Metasearch ment is out of scope for this particular ar- applications, however, attempt to strike ticle, information about the application can a balance between locally harvesting and be found at the LibraryFind™ project Web indexing content (when possible), while page: www.libraryfind.org. The remainder still providing support for federated search- of this piece will take a closer look at why ing of resources that do not support local OSU Libraries developed LibraryFind™ harvesting and indexing. Within the library and its positive outcomes. community, the most notable resources that So why create LibraryFind™ in the first do not support local harvesting and index- place? When I get asked this question, I’m ing, are electronic journals and abstracts. often surprised to find that many people In 2006, the OSU Libraries decided to assume that the primary motivation for take a more active role in the development creating LibraryFind™ was money. I can of its information infrastructure. Unsatisfied understand this line of thinking. On average, with the current crop of “next generation” a federated search solution to meet the needs systems available to the library community, of OSU Libraries would run approximately the OSU Libraries opted to re-think what it $20,000 to $40,000+ depending on the meant to develop library services as part of vendor and functionality purchased. While a single unified library platform, and set out LibraryFind™ has reduced our total cost of to develop that vision. This move certainly ownership for a metasearch application, the was not without risk, as the OSU Libraries actual costs or potential cost savings played a moved away from a vendor-supported model very small role in the decision to develop the to one in which the Libraries would be pri- LibraryFind™ application. marily responsible for both ongoing support In 2005, the OSU Libraries were using and development of their retrieval platform. a vendor-supported federated search tool. The first of the components to be de- Yet as the Libraries became more interested veloped as part of this initiative was Library- in developing custom services for our varied Find™. LibraryFind™ is an open source audiences, it became clear that this feder- metasearch application that was developed ated search solution simply did not meet 18 to be used as the primary access point for the current or future needs of the organiza- aggregating the library’s fragmented in- tion. While the Libraries could have looked Vol 15 No 1 • Spring 2009

for a new vendor solution, Administration Find™ application was the first to utilize a saw an opportunity to reinvent the library, “perpetual beta” model. New features would adopting a much more fluid environment be continuously released and evaluated. that would encourage the rapid develop- Services that found an audience were devel- ment of new and improved library services. oped further and retained; those that did LibraryFind™ served as the test-bed for not were removed from the application. For this model of service development. The re- the Libraries, this represented a big change sponsibility for creating the application was from past projects which generally included shared between Emerging Technology and a work plan with a definable ending point. Library Services, Reference Services (who LibraryFind™ on the other hand, contin- oversaw and provided formal usability test- ues to advance as the LibraryFind™ team … the process used to ing) and Digital Access Services (formally assesses how user needs evolve. develop the application known as Technical Services). Another change emerged in how the has changed the way So how has the development of Li- Libraries evaluate the success and failure of braryFind™ changed the OSU Libraries? new projects. As tenured faculty, our success that the OSU Libraries There have been two fundamental changes for promotion and tenure is tied to project view the creation of that have grown out this experience. First success. However, the Libraries have had to user services. and foremost has been the creation of a recognize that there is value in creating proj- library-wide platform. For the first time, the ects that never succeed because they provide Libraries have a discovery platform to use as valuable feedback for the next attempt. a building block for other library services. In LibraryFind™ is an ongoing research the two years since LibraryFind™ has gone experiment at the OSU Libraries. The live, the Libraries have been able to utilize program and its development continues to LibraryFind™ to expose the Libraries’ re- allow the Libraries to test concepts, push sources through other projects, like Oregon boundaries, and share library data with the Explorer (www.oregonexplorer.info), a state- campus and its extended user community. wide natural resources portal and Library à What’s more, the application has served to la Carte (alacarte.library.oregonstate.edu), further the research mission of the Libraries a tool designed to enable rapid develop- by significantly contributing to the metase- ment of course and subject guide pages. As a arch community and providing a valuable platform component, LibraryFind™ acts as resource for other libraries interested in a conduit for information, bridging the gap developing their own search infrastructure. that had existed between externally devel- Finally, LibraryFind™ has fostered an envi- oped applications and their users. Addition- ronment where librarians are more com- ally, the application itself has been well- fortable with research and development and received and continues to be well-utilized by are actively identifying applications which the OSU user community. further the Libraries mission to provide In addition to LibraryFind™, the users with the tools they need for teaching process used to develop the application has and research endeavors. changed the way that the OSU Libraries view the creation of user services. Like many References organizations, the OSU Libraries suffered O’Reilly, Tim, “What is Web 2.0,” from a paralysis of perfection, an idea that O’Reilly, http://www.oreillynet.com/pub/a/ services needed to be perfect before they oreilly/tim/news/2005/09/30/what-is- 19 could be presented to the user. The Library- web-20.html. The New Summit: Building the Foundation for Enhanced User Services

by Alan Cornish n December 1, 2008, the Orbis • Display of post-search filters, or facets, Head, Library Systems, Cascade Alliance unveiled a new on results screens; these facets include: Washington State Summit search system (Orbis author, format, publication date, lan- University Libraries O Cascade Alliance 2008). The creation of guage, and topic. this new, much more robust, Summit catalog was motivated by several factors. • Improved relevancy ranking of search Most importantly, the Alliance Council, results. in a 2006 strategic planning effort, identi- fied next-generation discovery systems as • Book jackets displayed in many re- a priority initiative for the Alliance (Orbis cords. Cascade Alliance 2009). The most press- ing need identified by the Council was to • Support for user-created lists, tags, and improve the patron interface to meet the book reviews. expectations of our current users, whose expectations are shaped by services such as • Ability to customize and extend the Amazon, Facebook, and Flickr. catalog through a published Applica- In March 2008, the Alliance Council tion Programming Interface (API), met and decided to commit to a product which provides the information that still in development, OCLC’s World- developers need to customize results Cat Navigator. The WorldCat Navigator and record displays. solution consists of three components: a customized view of WorldCat.org, which One additional element was needed, supports discovery; the Navigator Request however, to make this migration a success- Engine (NRE) software, which is an exten- ful one: maintaining the requesting and sion of OCLC’s VDX interlibrary loan borrowing capabilities to which Summit software, to support returnable borrowing; users were accustomed. These capabilities and the Circulation Gateway program, include the ability for users to search and which enables standardized communica- seamlessly request items from Summit insti- tion between NRE and library circulation tutions; to review the status of requests; and systems (OCLC, 2009). In the envisioned to renew and cancel requests. The previ- development partnership, OCLC and the ous Summit catalog and borrowing system Alliance would work together to create an relied upon the Innovative Interfaces (III) integrated search/request/delivery solution INNReach consortial borrowing software based upon these software tools. and the III Millennium software, which The essential features of a next-genera- is used by each Alliance library as its local tion discovery system are embodied in the catalog. Because of the tight integration new Summit catalog (summit.worldcat.org) that III achieved with Millennium and IN- and are: NReach and the high requesting volume of the existing Summit service, the challenge • Retrieval of article records from Sum- proved to be formidable. mit; indexes include ArticleFirst, British When the Council approved the Library Serials, PubMed/MEDLINE, change to WorldCat Navigator, it also cre- 20 and ERIC. ated an implementation team to spearhead OREGON LIBRARY ASSOCIATION

the migration. The Implementation approach by the Implementation Team, Team managed the WorldCat Naviga- by staff at the Alliance and at member tor implementation; which included institutions, and by OCLC staff who supporting the shutdown of the III were assigned to the implementation. INNReach system; coordinating Also, the tight timeline meant that the software development tasks performed INNReach shutdown process would within the Alliance; and supporting have to begin in early November, communication with and training only seven months after the Council’s for Alliance members (Orbis Cascade decision to commit to the WorldCat Alliance 2009). To a great degree, the Navigator migration. Implementation Team’s work, and the Beyond the timeline issue, there migration as a whole, has been trans- were technological ones. During the parent. A search of the term “summit fall 2007 Northwest Innovative Users migration” in Google will retrieve Group meeting, Kyle Banerjee pre- detailed information about the mi- sented at a plenary session on next- gration from the Alliance’s Web site, generation discovery systems (Banerjee including information on the Imple- 2007). One of the points that Kyle em- mentation Team’s work (Orbis Cascade phasized was the lack of maturity of the Alliance 2008). In short, the Alliance then-available discovery solutions, with and its members are taking a leadership all of them employing techniques such role among consortia in the develop- as screen scraping, use of JavaScript ment partnership with OCLC and code hacks, and the use of software in the implementation of WorldCat packages designed for other purposes Navigator. Making information on the (such as content management systems). migration public, when possible, will As a result, in order to deploy a much enable other consortia and institutions more robust discovery system that to benefit from this work. would better meet the needs of Summit It was clear from the start that the users, the Alliance and its members had migration would be challenging. The to embrace some relatively immature deadline required that the migration technologies. Much of the early work to Navigator would have to be accom- for OCLC and staff members working plished in a very tight timeframe and throughout the Alliance went to ensur- would require an “all hands on deck” ing the best possible request experience and service for users, given the techni- cal challenges of this migration. Additionally, it was clear at the March Council meeting that the Al- Making information on liance would be assuming some local the migration public, when software development responsibili- possible, will enable other ties as a development partner with OCLC. This was quite a change from consortia and institutions to the previous software support model benefit from this work. with III’s INNReach, for which the 21 Vol 15 No 1 • Spring 2009

vendor performed essentially all develop- ment work, and improvements in the software were requested and implemented through a lengthy enhancement process. … the Alliance has implemented For example, programming had to be per- a software solution that has the formed to support pass-through searching from Millennium systems throughout the potential for truly integrating the Alliance to the new Summit catalog. The discovery and delivery processes. programmers went beyond replicating the Millennium-INNReach functionality and While the migration work is still in implemented pass-through searching that progress as of this writing, January 2009, a retains advanced search limits, such as number of important milestones have been date ranges and media types, when pass- achieved. The core functions of discovery ing a search from a Millennium catalog and returnable requesting and borrowing to the new Summit catalog. Developing are in place. All of the discovery features the OpenURL resolver was particularly listed at the beginning of the article are critical, because it enables the system to now available to Summit users, including process requests for articles. The resolver the ability to extend the discovery system is configured so that institutions can set using a published API (OCLC 2009). Most up one or more entries as needed in order importantly, by employing WorldCat Navi- to support multiple fulfillment options, gator, the Alliance has built a foundation e.g., for each campus of a single Alliance that will support the ongoing improvement institution. These local development of user services. The Navigator Request efforts are being supported by an Alli- Engine software can support the delivery of ance technology group; the programmers articles as well as returnables (books, etc.). responsible for coding specific applica- The NRE can also support requests for ma- tions for the new Summit are listed on terials held by libraries outside the Alliance, the technology group’s Web page (Orbis either through WorldCat Resource Sharing, Cascade Alliance 2009). or by transferring request information from Because the new summit runs on the NRE to ILLiad, another OCLC resource WorldCat.org platform, only OCLC records sharing product. are searchable. Therefore the Alliance and While OCLC is still developing sup- OCLC launched an immediate initiative port for article requesting in NRE, the Al- to update the accuracy and completeness liance has implemented a software solution of member library holdings in WorldCat. that has the potential for truly integrating OCLC supported the updating of holdings the discovery and delivery processes. Sum- through reclamation and batch updating mit users will have the ability to search a projects, which enabled records to be added growing and diverse set of collections as in WorldCat for holdings not previously a single unified collection, and to obtain represented in the system. Because of the materials through a unified request system. rapid implementation timeline, OCLC It’s a challenging and ambitious goal, but worked with the Alliance and member with the implementation of the new Sum- libraries to perform these updates as a high mit system, the Alliance and its member 22 priority (Orbis Cascade Alliance 2009), with institutions have taken a big step toward coordination provided by the Alliance. meeting it. OREGON LIBRARY ASSOCIATION

References Banerjee, Kyle. “Implementing next- Orbis Cascade Alliance. “About the alli- generation catalogs on Innovative systems.” ance,” updated December 9, 2008, http:// Presentation at the Northwest Innovative www.orbiscascade.org/index/index (ac- Users Group Conference, Portland, Or- cessed February 15, 2009). egon, October 19, 2007.

Orbis Cascade Alliance. “Council retreat Orbis Cascade Alliance. “Alliance Tech 2006,” updated February 11, 2009, http:// Group,” updated February 6, 2009, http:// www.orbiscascade.org/index/strategic-agen- www.orbiscascade.org/index/summit-migra- da-council-retreat-2006 (accessed January tion-tech-group (accessed January 22, 2009). 22, 2009). Orbis Cascade Alliance. “Catalog Working OCLC. “WorldCat Navigator: More intui- Group,” updated February 6, 2009, http:// tive and efficient access to local, group, and www.orbiscascade.org/index/working- global resources,” http://www.oclc.org/ser- group-catalog (accessed February 16, 2009). vices/brochures/213665usb_worldcat_navi- gator.pdf (accessed February 16, 2009). OCLC. “WorldCat Affiliate Services,” http://www.worldcat.org/affiliate/default.jsp Orbis Cascade Alliance. “Implementation (accessed January 22, 2009). Team, updated February 6, 2009, http:// www.orbiscascade.org/index/vdx-implemen- tation-team (accessed February 16, 2009).

Orbis Cascade Alliance. “Summit migra- tion news,” updated April 18, 2008, http:// www.orbiscascade.org/index/summit-mi- gration-news (accessed January 22, 2009).

23 Building Catalogs in the Sand

by Wade Guidry or the past four years, the Coastal • The ability to browse specific item types Systems Manager, Resource Sharing Network (CRSN) (DVD, music CD, audiobook CD) Coastal Resource has had an evolving public interface Sharing Network (CRSN) F for their Beachbooks library catalog in • The ability to browse movies and TV pursuit of a better search experience for shows by genre and show title patrons. CRSN, serving a population of 70,000 in Tillamook and Lincoln counties, One popular new feature has been the includes six member library systems, two addition of a “Movies and TV” page that community colleges, 13 library buildings, allows patrons to browse movies by genre and one bookmobile. CRSN’s current an- (adventure, Disney, historical, etc.), and to nual circulation stands at 850,000, with a browse TV shows by title. While CRSN collection size of 400,000 items. This article did not add this feature until June 2008, shares some of the ideas behind their evolv- some of its associated searches still reached ing catalog design. the top 50 most popular searches for the entire year. Finding the catalog Providing relevant labeling for media Users can find a library catalog one of two searching is another good idea CRSN ways: by doing a Web search, or by enter- is pursuing. Librarians know that when ing the catalog’s address in their browsers. searching music, an “author” search Prior to 2005, the CRSN catalog address, equates to a search for a composer or per- www.crsn.lib.or.us, was difficult to market former, or that “publisher” equals a music and hard to remember. So, during an ILS label, like Blue Note. The average user migration in early 2005, CRSN switched to does not necessarily think that way. Label- the much friendlier www.beachbooks.org. ing media searches with relevant terms like Such addresses double as great search terms “director/performer,” “composer/artist” to help patrons find the catalog online. For and “title/track” is a simple way to im- example, a Google search for “beachbooks” prove media discoverability. puts the CRSN catalog at the top of the resulting hit list. Cover to cover CRSN began providing cover art and reviews Media coverage on Beachbooks in late 2004. Like many In 2008, patrons of the Beachbooks cata- libraries, CRSN purchases enriched content log used the search term DVD more than through its ILS vendor, with the content any other and three times as often as the originating from Syndetic Solutions, now next most commonly used term. CRSN part of Bowker Publishing. While CRSN patrons also searched for VHS often cannot measure its impact, they do quickly enough to put it in the top 20 search terms hear from patrons if the content goes of- used. Clearly, Beachbooks patrons want fline or loads slowly. The downside to this to use the catalog to find media. Follow- enriched content has been its cost and the ing the lead of other Oregon libraries like lack of content for media like movies and Deschutes Public Library, CRSN has tried music. However, some lower-cost alternatives to incorporate media-friendly features in for enriched content are starting to appear. the Beachbooks catalog, including: Baker & Taylor now offers an enriched con- 24 tent solution called “Content Café,” which • The ability to search within broad A/V includes content for music and movies in groups (movies, music, audiobooks) addition to books. Another vendor, Book- OREGON LIBRARY ASSOCIATION

Letters, offers dynamic book-related content for newsletters and Web pages that can be leveraged for in-catalog delivery. For some patron needs, CRSN has also experimented with using their content stream in creative ways. subscription databases provide For example, one of the CRSN catalogers the perfect tool. In relation to the suggested last year that they try to mimic catalog, though, where do they the display of featured books on the Barnes & Noble Web site. Based on that sugges- go and what are they called? tion, CRSN leveraged its subscription to Syndetic Solutions content to create a dy- Web traffic, Google Analytics and AW- namically-updated display of recently added STATS. Google Analytics is by far the easier titles that appears on the Beachbooks front tool to implement and within the capabili- page. Using the same technique, Newport ties of casual techies. Public Library, one of the CRSN member Google Analytics provides interesting libraries, displays recent staff picks from the and useful information about catalog traf- Beachbooks catalog on their Web site. fic, browser screen resolutions, and connec- tion speed. It also provides usage data such Database integration as most common entry points, number of For some patron needs, subscription da- visits and unique visitors, average session tabases are the perfect tool. In relation to length, and most visited pages the catalog, though, where do they go and This data helps CRSN decide which what are they called? For EBSCOHost, the screen resolutions to test most thoroughly predominant CRSN database, CRSN has with the catalog. For example, 4 years sidestepped those issues by implementing ago 800 x 600 was still a common screen a search box that allows patrons to search resolution, but today only 10 percent of EBSCOHost directly from the front page Beachbooks users use a screen resolution of of the catalog. Patrons simply enter their 800 x 600 or lower. Google Analytics also search query and click “Search.” Libraries shows that despite growing sales of Apple subscribing to EBSCOHost can easily cre- computers, over 91 percent of Beachbooks ate such a search box using the new Search users still run Windows, and the use of Box Builder tool provided by EbscoHOST. Linux among patrons remains below one CRSN also integrates its database list percent. Although still a very small subset directly into the catalog. Embedding the of total use, mobile devices are also starting database page directly in the catalog has the to make an appearance in the list of OSes— disadvantage of relying on the dynamic, a trend worth watching. session-oriented nature of catalog URLs. To To use Google Analytics, libraries can counteract this, CRSN offers direct navi- set up an account at www.google.com/ana- gation to the database page via the URL lytics, and follow the instructions for adding http://www.beachbooks.org/databases, a simple HTML snippet to the footer of the which can be bookmarked and linked. catalog. After a few days, data is available on the Google Analytics dashboard. Undergoing analysis Usage statistics can help guide catalog Miscellany 25 design. For the Beachbooks catalog, CRSN Between major revisions, CRSN continues uses two freely available tools for analyzing to add smaller features to our catalog as Vol 15 No 1 • Spring 2009

new needs and opportunities arise. response to an unsuccessful search, our cata- Recommend it!—The “Recommend log now provides a number of search tips It!” feature allows patrons to e-mail a link to a and links to similar searches. single item in the catalog to a friend. A minor success, patrons use this feature about 20 Constant testing times per month. To keep the catalog moving forward, CRSN Comments Submission—CRSN tries likes to test it regularly. Gordon and West to make it easy for patrons to contact mem- (2008) listed three use cases they apply as a ber libraries directly through the catalog by quick litmus test of the usability of any cata- including links to comment forms within log they encounter. To paraphrase their tests: the catalog. Patrons use this feature regu- larly, with several new messages arriving • Can a patron easily find a title such as each day. Most of the comments involve “She’s not there” that includes Boolean specific account questions (“What’s my terms? PIN?” or “How to I renew?”), but patrons also make purchase recommendations and • Can a patron easily list all DVDs in the the occasional design suggestion. catalog? Direct navigation—Amazon and other commercial sites have successfully taught • Can a patron easily find titles from the users about direct site navigation with URLs New York Times bestseller lists that the such as amazon.com/movies and amazon. library owns? com/music. CRSN offers similar URLs for some pages within the catalog to help with To those tests, CRSN would add some others: navigation, as well as with bookmarking and direct linking. Such URLs include beach- • Can a patron quickly find the catalog books.org/databases, beachbooks.org/mov- via search engines like Google? ies, and beachbooks.org/library2go. No results found—Rather than simply • Can a patron quickly limit a search to 26 display a “no results found” message in fiction or nonfiction? OREGON LIBRARY ASSOCIATION

• Can a patron easily search by media Beachbooks Catalog 3.0 type? http://sandbox.beachbooks.org (in beta)

CRSN also looks to other catalogs for BookLetters new ideas. CRSN visits other SirsiDynix http://www.bookletters.com/ catalogs to find out what other libraries are accomplishing on the platform. For broader Coastal Resource Sharing Network inspiration and trend watching, CRSN http://crsn.beachbooks.org looks to catalogs of prominent Northwest libraries, especially those active in catalog EbscoHOST Search Box Builder development. Those include Seattle Public http://supportforms.epnet.com/eit/search- Library, Multnomah County Library, De- boxbuilder/ schutes Public Library, and Corvallis-Benton County Public Library. Looking even more Google Analytics broadly, WorldCat and OCLC FictionFind- http://www.google.com/analytics er incorporate new and interesting catalog features on a regular basis. Rachel Singer Gordon and Jessamyn West, “Tech Tips for Every Librarian,” Computers Conclusion in Libraries, May 2008, Vol. 28 Issue 5. Looking back over the last four years, the http://search.ebscohost.com/login.aspx?dire Beachbooks catalog has benefited greatly ct=true&db=aph&AN=31900362&site=eh from the input of staff and patrons, as well ost-live (EbscoHOST persistent link) as the anonymous contributions made by those whose work they have copied. Looking forward, the Beachbooks catalog continues to evolve. This coming year, CRSN will likely implement saved reading lists for patrons, with help from new ILS features. CRSN might also need to adapt the catalog for use on increasingly prevalent mobile devices like smart phones and incorporate new media types, as they did recently with downloadable content and Netlibrary E-books. Whatever happens, one thing is sure: the Beachbooks catalog will continue to evolve.

References Baker & Taylor Content Café http://www.btol.com/pdfs/content_cafe.pdf

Beachbooks Catalog 2.0 http://www.beachbooks.org (current release) 27 Legacy Metadata and the New Catalog

by Richard Sapon-White he nature and purpose of library The metadata for these materials Head of Cataloging, catalogs are rapidly changing. The may currently exist in paper shelflists and Oregon State University Tfuture catalog will enhance users’ catalogs awaiting retrospective conver- searching capabilities with relevance rank- sion. At Oregon State University (OSU), ing and provide access to book reviews and some of our rare books are likely only user comments. Searching the catalog may described in the special collections card be integrated with searching citation indexes catalog, although there are plans to convert and Internet databases. A whole world of these to MARC records in the near future. new possibilities for the future of the catalog The History of Atomic Energy Collection is just around the corner. demonstrates a slightly different situation. At the same time, the contents of the A catalog of the collection was published new catalog are also changing. Last year, the in print (Laudamus 1990), but in order to Library of Congress released a report calling locate and use the collection, a researcher on libraries and archives to reveal more of must first discover the printed catalog. In their hidden collections—archival and spe- WorldCat, only eleven libraries hold a copy cial collections of unique and rare materials of this title. To resolve this problem and as well as other resources not represented make the collection more visible, we have by bibliographic records in online catalogs begun cataloging books in this collection, and databases (Library of Congress Working ensuring that the individual titles are acces- Group on the Future of Bibliographic Con- sible via WorldCat. trol 2008). Meeting this challenge requires Special collections and archives also an understanding of the scope of the prob- may have backlogs of uncataloged acqui- lem and strategic planning to solve it. This sitions. The Northwest Digital Archives article explores the nature of these hidden project has opened many Pacific North- materials, how libraries can bring them to west resources to the world at large. Doing light, and what resources will be needed to their part for the project, OSU archivists add this content to the new catalog. Oregon have created digital finding aids to many State University’s efforts to meet this chal- of their collections. Some of these finding lenge will be highlighted. aids are for recently accessioned collections, Special collections and archives contain but many are for materials that have been rare and unique library materials in many stored in our archives for many years. The formats, including print books and serials, finding aids, created in Encoded Archival manuscripts, photographs, locally produced Description (EAD) format, are converted audio and video recordings, and realia. Many to MARC records using MARCEdit and of these items remain hidden from scholars then added to WorldCat and our local and researchers because the metadata describ- catalog (and therefore also to Summit, the ing them may be difficult to locate, incom- union catalog of the Orbis Cascade Al- plete, or non-existent. The challenge for these liance). By making these records widely repositories is twofold: create metadata so that available, we provide multiple avenues for scholars can search for these materials online, researchers to discover the finding aids. and digitize the resources so that scholars can Although the projects mentioned above examine them from their desktops. Digitiza- are moving OSU forward in making our tion enables discovery of the resources as well holdings more widely known, we have a long 28 as preservation of the originals. road ahead to achieve our goal of providing OREGON LIBRARY ASSOCIATION

metadata for all of our titles. Many other ma- terials owned by OSU must wait their turn for metadata creation. These include a small What will it take to reveal backlog of gift monographs (approximately a thousand titles, many of them requiring these hidden materials in the original cataloging), about a hundred mono- catalog of the future? graphs in Farsi, and thousands of serial titles awaiting retrospective conversion. Serials in the Linus Pauling Collection and the OSU metadata. We will need library staff who Integrated Plant Protection Center Library are skilled and can apply their attention to also await cataloging. detail, knowledge of cataloging rules, and The situation at other libraries is likely experience with different metadata sche- no different from OSU. Titles in backlogs mas to these valuable resources. Since these are often similar and include various types materials tend to be rare and unique, the of gray literature, such as theses and dis- trained individuals should have facility with sertations, foreign publications, and gov- languages as well as knowledge of rare book ernment documents. These materials often cataloging and multiple metadata schemas. require complex and/or original cataloging. The manual records of the past, when they With the many demands placed on the do exist, will also need to be upgraded to limited number of catalogers today, these current standards. Adequate staffing to backlogs tend to be fairly static. handle this additional work will likely be a The scope of the problem of materi- challenge for many libraries given the cur- als awaiting retrospective conversion is the rent economic situation. subject of two recent studies of the National In addition to creating or converting Union Catalog, Pre-1956 Imprints (popularly metadata, the ability to crosswalk metadata called Mansell). These studies have revealed from one schema to another will also be that 25 to 28 percent of Mansell entries have critical. Redundant metadata allows for yet to be represented in WorldCat and the multiple avenues of discovery. For example, a approximately 13 million entries in Mansell, metadata record in an institutional repository these studies indicate that as many as 3.6 using Dublin Core metadata could be con- million titles await retrospective conver- verted to MARC for inclusion in a WorldCat sion (Beall 2005; DeZelar-Tiedman 2008). record. Such work requires metadata special- Although this cataloging labor is presum- ists with the skills necessary to create and ably distributed among many libraries, the adapt macros for metadata conversion. studies highlight the degree to which many There are several obstacles to fulfill- titles are well-hidden from most researchers ing these goals. Fundraising is needed for in today’s online environment. personnel and equipment. Finding skilled What will it take to reveal these hid- catalogers has not been easy in recent years. den materials in the catalog of the future? The supply of new metadata specialists Our libraries will need the human resources coming from library schools is limited to create metadata and the equipment to while the number of catalogers currently provide quality scanned images. As with any employed is dwindling. Training parapro- project, funding will be needed to hire and fessionals is a possible solution, although train personnel to create and maintain the training takes time and close supervision. 29 Vol 15 No 1 • Spring 2009

Subject librarians will need to set priorities References to allocate financial and human resources Beall, Jeffrey and Karen Kafadar. 2005. to those materials with the greatest signifi- “The proportion of NUC pre-56 titles cance to users. represented in OCLC WorldCat.” College Despite these obstacles, OSU has & Research Libraries, vol. 66, no. 5, Sept. forged ahead on making our hidden re- 2005. p. 431–435. sources more visible. As material budgets and new serial subscriptions dwindle, staff DeZelar-Tiedman, Christine. 2008. “The have time to devote to special projects. proportion of NUC pre-1956 titles repre- With training and guidance, they can tackle sented in the RLIN and OCLC databases some of the cataloging backlogs and work compared: a follow-up to the Beall/Kafadar on our retrospective conversion project for study.” College & Research Libraries, vol. 69, serials. Special collections staff occasionally no. 5, Sept. 2008. p. 401–406. requests MARC cataloging of specific titles of importance. Library school interns have Laudamus, Leif and Ramesh S. Krishna- helped with some cataloging projects in murthy. 1999. The History of Atomic Energy the past and could be helpful in the future. Collection at Oregon State University: A Digitization projects could include funding Catalogue of Holdings. New York: James for metadata creation; such metadata could Cummins Bookseller. then be converted to MARC for inclu- sion in our catalog and WorldCat. We will Library of Congress Working Group on the continue to seek new and creative ways to Future of Bibliographic Control. 2008. On stretch our resources and serve our users. the Record: Report of the Library of Congress There are many rare and unique re- Working Group on the Future of Biblio- sources that are difficult for researchers to graphic Control. Washington, D.C.: Library locate or use in all of our collections. The of Congress. objective of including all of them in the catalog of the future will require thoughtful The National Union Catalog, Pre-1956 planning, fundraising and execution. The Imprints: A Cumulative Author List Repre- effort we make will enable our libraries to senting Library of Congress Printed Cards and share the breadth and depth of our collec- Titles Reported by Other American Libraries. tions with the wider world. 1968–1981. London: Mansell.

30 Northwest Digital Archives: Evolving Access to Archives and Special Collections in the Northwest

by Jodi Allison-Bunnell s library catalogs that offer general search across them provides researchers Program Manager, access to a wide range of materials with a level of access to information about Northwest Digital Archives, continue to evolve, so do special- collections that is not available through any Orbis Cascade Alliance A ized tools that offer more detailed access to other source. particular types of materials. Few areas have Excellent search engine exposure also en- seen as much activity and as many changes hances access. Each collection in NWDA also over the last ten or fifteen years as archives has a MARC record in WorldCat and a local and special collections. The Northwest catalog, and most of those records link to the Digital Archives (NWDA) program at the finding aid in NWDA through the 856 field. Orbis Cascade Alliance is in its seventh year Members who have digitized materials from of building effective access to archives and collections can link to those images from special collections materials in the North- item-level or folder-level descriptions in the west. The program has evolved over time finding aid in the NWDA database. to meet the changing needs of its member The finding aids are encoded in an institutions and their researchers and will XML document type definition (DTD) continue to evolve to expose these often- called Encoded Archival Description hidden unique collections. (EAD). In the past, finding aids for archival NWDA offers enhanced access to collections were most commonly created as archives and special collections materials typescript or word processing documents. in , , Oregon, and Some descriptive standards existed, but Washington through a union database local practices prevailed. Unlike library of finding aids, or descriptions of those cataloging, with its natural incentive to materials by collection. Finding aids pro- save costs by sharing and distributing copy vide varying levels of detail on collections cataloging, archives and special collections beyond the MARC record. The ability to items are unique, and nearly all cataloging is entirely original. Descriptive practices have historically been shaped by the nature of each collection, with strong arguments made for particularized rather than stan- dardized practices. With the develop- ment of new library catalogs, EAD, the World Wide Web, and a general mandate

Unlike library cataloging, with its natural incentive to save costs by sharing and distributing copy cataloging, archives and special collections items are unique, and nearly all cataloging 31 is entirely original. Example of a finding aid in the Northwest Digital Archives. Vol 15 No 1 • Spring 2009

to improve collection access to expand- climbed to over 15,000 retrievals a month. ing audiences, archivists have developed Since mid-2008, database use has been and begun to adhere to more standardized consistent with 20,000 to 22,000 finding methods for describing groups of archival aids retrieved each month. Over the same and special collections materials. In 2004, three-year period, the pathway that users the Society of American Archivists released of the NWDA site use to get to the find- Describing Archives: A Content Standard ing aids has changed considerably. While Current members of (DACS), which for the first time prescribes researchers first came in primarily through the Northwest Digital minimal and optimal elements that should the NWDA search and retrieval site, that Archives program: be present in any description of items from quickly changed once the finding aids were or groupings of archival materials (Society exposed to search engines through Google Alaska State Library of American Archivists 2004). sitemaps in mid-2006. The proportion of Central Oregon Community NWDA began in 2002 with fund- entry from search engines and referring College ing from the National Endowment for the sites quickly increased to around 50 percent Humanities (NEH) and National Histori- and since October 2007 has stayed around Central Washington cal Publications and Records Commission 90 percent. Of that 90 percent, about two- University (NHPRC). The project grew with addition- thirds are from search engines and one- Eastern Washington State al funding from those same agencies in 2005 third from referring sites, which include Historical Society/Northwest and became a program of the Orbis Cascade library catalogs, Wikipedia, and sites that Museum of Arts & Culture Alliance in 2007. focus on specific subjects. Eastern Washington The Alliance’s administrative and fiscal Google metrics give much more detail University home is the , and the on these visitors and how they behave on NWDA Program Manager is an Alliance the site. The most commonly retrieved Gonzaga University employee. NWDA’s technical infrastructure finding aids are for the Montana State Idaho State Historical Society is hosted by Washington State University, Prison Records at the Montana Histori- with Head of Systems Al Cornish serving as cal Society, the Montana Ku Klux Klan Database Administrator. A half-time student Records and the Expo ‘74 World’s Fair Lewis & Clark College programmer helps him with system main- Records at the Eastern Washington State tenance and development. Member fees Historical Society, and the Bill Bowerman Montana Historical Society support this staff, software, hardware, and Papers at the University of Oregon. Most Oregon Health & Science costs associated with telephone and in-per- visitors are from the United States, but University son meetings. NWDA gained new members there is also significant use from Canada, from the Alliance and began as a program the United Kingdom, Germany, and Oregon Historical Society with thirty-one members in July 2007. Australia. Within the United States, most Oregon Institute of NWDA uses both Google metrics visitors come from within the Northwest Technology and internal reports to track usage. Mem- (primarily Seattle, Portland, Missoula, Spo- bers can access basic usage reports, which kane, and Eugene), but visitors from New show how many of their finding aids are York appear in the top 20. Visitors view retrieved each month, on the member Web an average of two pages on each visit, and site (Northwest Digital Archives http:// stay on each page for just under two and a nwda-db.wsulibs.wsu.edu/reports/docu- half minutes. Most keyword searches were ment.pl). These reports show a consistent on personal names, followed closely by upward trend in use. In January 2006, organization names (Google Metrics run 32 only 24 finding aids were retrieved from on nwda-db.wsulibs.wsu.edu for March 1, the database, but by January 2008 use had 2008–January 15, 2009). OREGON LIBRARY ASSOCIATION

Improved discovery on the Web also leads to increased collection use. NWDA members report increases in collection use NWDA’s commitment to usability since they became consortium members. In a spring 2008 survey of the membership, principles and testing … has 50 percent of members strongly agreed that resulted in significant changes participation in NWDA has increased the to and important improvements use of their collections, with an additional 31 Current members of percent somewhat agreeing with that state- in the site’s ability to effectively the Northwest Digital ment. Nearly half the members estimated in- serve researchers. Archives program: creases in use at 1 percent to 10 percent, with Oregon State University another one-third estimating increases in use The UTWG formulated usability princi- at 11 percent to 20 percent (NWDA Pro- ples in January 2006 based on Jakob Nielsen’s Pacific Lutheran University gram Assessment 2007). Anecdotally, many work and on Web site accessibility needs for Seattle Municipal Archives members report that researchers increasingly users with disabilities (Orbis Cascade Alli- appear in their reading rooms with finding ance 2006). The group followed with its first Museum of History & aids printed out from NWDA and knowing round of usability testing at the annual meet- Industry exactly what they would like to see. ing of Northwest Archivists in Butte, Mon- University of Alaska The design and functionality of the tana, in May 2006. The test subjects were Fairbanks first NWDA search interface, made avail- other archivists not associated with NWDA. able in summer 2004, were based largely on One UTWG member interviewed subjects University of Idaho the needs and preferences of the archivists and one took notes. Test subjects were al- University of Montana working on the NWDA project. Decisions lowed to explore the site with little imposed about the site’s functionality, appearance, structure. The results of this test showed that University of Oregon and other factors were often difficult to NWDA needed to focus on keyword rather make, as they were based on opinions about than browse searching and revealed some the needs of researchers rather than direct key navigation issues that had not previously Washington State or documentable information. The original been noticed. Resulting changes included a University Web site was intended to serve both NWDA keyword search box on the home page and Western Oregon University project members and their researchers, and greater consistency in navigation on the site Western Washington the navigation and terminology reflected (Orbis Cascade Alliance 2006). University this dual audience. In 2005, with discussions Tony Kurtz at Western Washington Uni- of general usability more visible within the versity and Donna McCrea at the University library and archives profession and with a of Montana performed the second round of need for a more rational basis on which to usability testing in late 2006 and early 2007 shape the site, NWDA formed a usability with undergraduates at their institutions. Willamette University testing working group (UTWG) to guide This testing was more formal, with scripts the consortium’s work. NWDA’s commit- and Institutional Review Board approval. ment to usability principles and testing has Kurtz and McCrea recorded and fully tran- involved new skills and significant time com- scribed the interviews. The resulting report mitments from participating members on made a number of crucial recommendations behalf of the consortium, and it has resulted that included eliminating or moving much in significant changes to and important im- of the information on the NWDA page that provements in the site’s ability to effectively was intended for member institutions; clari- 33 serve researchers. fying search options; providing more context Vol 15 No 1 • Spring 2009

to help users understand what finding aids mendations for additional changes based on are; clarifying search results; providing a their results (Orbis Cascade Alliance 2008). search within the finding aid; and making Round five of testing, performed by Megan the printing option clearer. NWDA imple- Friedel of the Oregon Historical Society, mented these recommendations that same tested the search experience on the site, year (Orbis Cascade Alliance 2006). including the basic and advanced searches, Tiah Edmunson-Morton of Oregon searching within a finding aid, and entry State University performed the third round from referring sites like library catalogs and of usability testing in early 2007, focusing on search engines. She found that a number finding aid presentation rather than search- of small and larger revisions are needed ing. Like round two, these tests were scripted to make the site easier to use (Orbis Cas- and transcribed. Her tests showed that sub- cade Alliance 2008). NWDA is currently jects scrolled rather than read lengthy finding reviewing these results and considering what aids; wanted better options to search and changes to make to the site next. navigate within finding aids; were confused Usability testing data has provided the by lingering archival jargon; and had varying group with a better-supported framework opinions on whether the site satisfied their on which to base changes to the site: real us- needs for detail. Specific recommendations ers have told NWDA what does and doesn’t included modifications to the navigation meet their needs and showed us what they sidebar, addition of an expand-collapse most want from the site. Naturally, member function, changes in sequencing and fonts, institutions and working group members clearer divisions between sections, and direct also have the ability to shape improvements links to repository homepages (Orbis Cas- and changes to the site, but the focus on cade Alliance 2006). NWDA implemented users has helped NWDA get away from “de- all of these recommendations by summer signing for ourselves.” Usability testing has 2007 as part of a larger site redesign that also changed to reflect changes in use: while incorporated new graphics and a cleaner it was essential early in the program to test interface. That same summer, the Usability primarily the NWDA site and search func- Testing Working Group was renamed the tions, the fact that 90 percent of users now Usability Design Working Group (UDWG) come in through referring sites means that with a broader charge to shape the develop- it is now much more important to optimize ment of the NWDA site. that experience. In 2008, the UDWG did two addition- This stance is also shaping NWDA’s al rounds of usability testing. Round four, future program. Since 2004, NWDA has again performed by Edmunson-Morton, wanted to move beyond metadata to offer was a follow-up to round three and tested considerable access to digitized archival con- the subjects’ reactions to changes made to tent, but the planning and needs assessment the site. Their responses to some elements for such a project was beyond the project’s were positive, including repository links and capacity. The Alliance merger provided the the search within a finding aid function. impetus and opportunity for that planning. Other responses indicated that some chang- Since 2007, NWDA has been studying exist- es, including some aspects of the search ing digital programs within NWDA and within the finding aid and the expand-con- the Alliance, establishing priority researcher 34 tract function, need additional refinements audiences across institutions, and asking or are not working as expected. Both users NWDA and Alliance members about their and working group members had recom- needs and desires for programs, best prac- OREGON LIBRARY ASSOCIATION

Greene, Mark A. and Dennis Meissner. “More Product, Less Process: Revamping …the focus on users has helped Traditional Archival Processing.” American Archivist 68, no. 2 (2005): 208–263. NWDA get away from “designing for ourselves.” Northwest Digital Archives, “Northwest Digital Archives Report.” 2007. http://nw- tices, training, and services. In conjunction da-db.wsulibs.wsu.edu/reports/document.pl with Elizabeth Yakel of the University of (accessed March 25, 2009). Michigan’s School of Information, NWDA has conducted interviews with researchers Allison-Bunnell, Jodi. “Annual Northwest from priority audiences to assess their needs Digital Archives Member Assessment.” and desires for selection and presentation of 2008. Presented May 2008 at Northwest digitized archival content. The results, cur- Archivists Conference, Anchorage, Alaska. rently being compiled, will shape the direc- tion of a future program that will integrate Northwest Digital Archives Usability Test- content hosting, reformatting services, meta- ing Working Group,”Usability Principles.” search, and digital preservation. NWDA January 30, 2006. http://orbiscascade.org/ has received a Collaborative Planning Grant index/cms-filesystem-action?file=nwda/re- from the Institute for Museum and Library ports/ usabilityprinciplesutwg.pdf (accessed Services to propel the program toward reality. March 25, 2009). With its commitment to usability testing, skills developed within the group, Orbis Cascade Alliance, “Northwest Digital and looking forward to expanded digital Archives Usability Testing Working Group services in the future, NWDA will contin- Reports.” 2002-2007. http://orbiscascade. ue to build an effective program to enhance org/index/northwest-digital-archives-reports access to those unique collections that, (accessed March 25, 2009). in a world where published materials are increasingly universally accessible, are the Orbis Cascade Alliance, “Northwest Digital lifeblood and definition of the institutions Archives Usability Testing Working Group - that hold them. Usability Testing Round 4 Report.” March 12, 2008. http://orbiscascade.org/index/ References cms-filesystem-action?file=nwda/reports/ Society of American Archivists, June 5, nwda_utwg_ut4_report_20080312.pdf (ac- 2006. Development of the Encoded Archi- cessed March 25, 2009). val Description: DTD. http://www.loc.gov/ ead/eaddev.html (accessed March 5, 2009). Orbis Cascade Alliance, “Northwest Digital Archives Usability Testing Working Group - Society of American Archivists, April 16, Usability Testing Round 5 Report.” Novem- 2007. Encoded Archival Description Tag ber 6, 2008. http://orbiscascade.org/index/ Library. http://www.loc.gov/ead/tglib/ (ac- cms-filesystem-action?file=nwda/reports/ cessed March 25, 2009). udwg_ut5_executive_summary_20081106. pdf (accessed March 25, 2009). Society of American Archivists, Describing 35 Archives: A Content Standard. Chicago: Society of American Archivists, 2004. Usability Survey of Keyword Searching

by Elizabeth Ramsey his usability survey was originally at random from CU library users who were MLS student, developed during a practicum at present at those times I needed to complete Emporia University Ta small liberal arts university in the task. Although not a scientific sampling, Northeast Portland, Concordia University those surveyed represented a broad range of (CU), as part of my graduate studies at Em- CU library users, from a freshman complet- poria University. It was originally suggested ing his first semester in college to an English as a project by my practicum supervisor and language learner with conditional admission head of CU reference, Judy Anderson, as to CU to a middle-aged student returning the first step toward creating an online tuto- to complete her graduate studies after being rial on keyword searching aimed primarily out of school for over twenty years. After at distance students The goals of this project listening to a scripted introduction, each were to give me practical experience with subject completed an informed consent the usability concepts I had been studying form, answered preliminary questions, then as well as giving CU librarians new insights attempted to complete eight tasks related into the needs of their users. to keyword searches using the Concordia The first task was to identify what we University library catalog. Observations hoped to learn about CU library users and were recorded on a checklist with a num- their methods of finding resources. Judy ber assigned to each subject to help ensure Anderson and CU reference librarian Krista confidentiality. Each survey was completed Reichard helped develop seven initial ques- in less than twenty minutes. tions we wanted answered through the survey. These initial questions then guided the de- Initial Questions and velopment of the survey’s interview questions Concomittant Survey and tasks. We also looked at other library tu- Questions/Tasks torials in use by libraries in the Orbis Cascade 1. Do students understand when to use a Alliance, and usability surveys conducted at keyword search? other libraries. Resources that proved espe- cially useful are listed in the references. Interview Question: When do you Additional support for this project was think you should use a keyword search? provided by Donna Bachard, Concordia University Research Committee Representa- Task: Perform a keyword search in the tive, who helped refine the Human Subjects Concordia University library catalog Research Investigation Application and for “adult education.” ensure confidentiality procedures. Molly Lee, an intranet communication manager 2. Do students know how to modify for Daimler who is experienced in usability their search? (any field … Boolean studies, also provided initial guidance in terms …) the possible processes and procedures for usability surveys. Task: Modify your search. Look for: “adult education” AND “literacy” print Method resources available in English at Con- The survey was conducted from November cordia. How many results did you get? 11 to 18, 2008, on the Concordia Universi- 36 ty campus and in the homes of subjects. The 3. Do students know what information eight participants in the survey were selected they need to find an item on the shelf? OREGON LIBRARY ASSOCIATION

Interview Question: How would you searches of any kind. Half of the subjects did find the text “Print literacy develop- not know when to use keyword searches in a ment” in the Concordia library? library catalog, so an explanation of when to use this type of search would be an impor- 4. Do student know electronic books are tant introduction to a tutorial. available in the catalog, and how to ac- In the task requiring subjects to cess them? modify their searches, only one-quarter of the subjects used the “modify search” Interview Question: How would you button at the top of the search results page. access the resource “Review of adult Most subjects knew to add terms to their learning and literacy?” original search to narrow their results and understood the Boolean method of add- 5. Do students know where to find the ing AND between their terms. However, contents of a book? because subjects did not generally know how to get to the advanced search page Task: Find the contents of the book through the “modify search” button, they “Print literacy development: uniting missed the opportunity to narrow their cognitive and social practice theories” searches in additional ways, such as by lo- cation, language, type of resource and year. 6. Do students know how to find similar Several comments were made that the type items? in the “modify search” button was so small as to be unnoticeable. The survey results Task: Find similar items on the topic and comments point to the importance of Adult Education. drawing attention to the “modify search” and other buttons at the top of the page in 7. Do students understand how to mark a tutorial. and save items? Many of the participants were not en- tirely sure how to locate resources, whether Task: Go back to your original search they were traditional books or e-books. Be- results. Mark and save: “Adult learning cause less than half of the subjects connected method: a guide for effective instructions.” the call number with the location of an item on the shelves, this topic is also an impor- Task: Send the item you saved in a tant component in a tutorial. Only one brief display format to the following participant had experience with Netlibrary e-mail address. and Ebrary, so resources should also include pointers to tutorials for these products. Discussion Most subjects were able to eventu- During the preliminary interview all subjects ally complete the tasks that required them reported that they had conducted keyword to click on the item title and use the tabs searches of some kind. However, only two marked “More Details”, “Find Similar subjects had done so using the CU library Items” and “Full Records” (see Fig. 1). catalog (http://catalog.cu-portland.edu/ However, several expressed surprise at the search~S2). All had conducted keyword information they could access there, so searches using Google, and said Google was the uses of the tabs should also be clearly 37 the main place they conducted information explained in a tutorial. Vol 15 No 1 • Spring 2009

how to modify search results, how to find an item on the CU shelves, how to use e- books, how to find the contents of a book, how to find similar items, and why and how to save and export items. While some sub- jects were able to find work-arounds which took them outside the library catalog, they will be able to perform much quicker, more accurate searches once they have a better understanding of all of the facets of the CU library catalog interface. A clear and de- tailed tutorial that includes comprehension Fig. 1: Single item display in Concordia University catalog checks such as a brief quiz may help them The tasks which gave the subjects the achieve that understanding. most trouble were those connected with the use of the “Book Bag” button. Only one References student understood that clicking on this Feldmann, S. (1999). The key to online button was the first step in being able to save catalogs that work? Testing one, two three. and export an item from their initial search Computers in Libraries, v19 n5 p16-18,20 results, and this student had attended an in- May 1999. formational lecture presented by a CU librar- ian. Only one student clicked on the item Foott, S. (2004). A usability study of the they wanted, then used the “Save Record” library catalog at Monash University. Re- button at the top of the screen. When these trieved September, 2008 from http://www. two subjects did manage to save their records valaconf.org.au/vala2006/papers2006/89_ and view them, they complained that there Foott_Final.pdf. was no clearly identified link to e-mailing their saved item, only the “View and Export” Fried Foster S., & Gibbons, S. (Eds.). button. Another commented that when she (2007). Studying students: the undergradu- accesses articles from databases there is always ate research project at University of Roch- a clear method of saving and exporting items. ester. Chicago: Association of College and Half of the subjects also asked how such a Research Libraries. task could be useful in their future searches. These results indicate that users need clear Reeb, B. (n.d.). Lita Regional Institute: instructions for saving and exporting items, Design process and usability. Retrieved and the “Book Bag” label does not adequate- September, 2008 from http://docushare.lib. ly represent this feature. rochester.edu/docushare/dsweb/View/Col- The results of this usability survey lection-2888. appear to answer in the negative all the questions initially posed. Thus, all aspects of Usability testing at Bobst Library at New a keyword search present in the survey tasks York University (n.d.). Retrieved Sep- should be addressed to some degree in a tember, 2008 from http://www.nyu.edu/ 38 tutorial: when to initiate a keyword search, library/resources/usability/testing1.htm OLA Quarterly Spring 2009 Vol 15 • No 1

Advertising Rates

Space Size Single Issue Multiple Issue

Full page 6.375” wide x 10” high $200 $175

1/2 page (horizontal) 6.375” wide x 4.25” high $115 $100

1/4 page 3.5” wide x 4.25” high $60 $50

Contact: Margaret Dull, [email protected], 503-943-7685 (Phone), 503-943-7491 (Fax)

The Publications Committee Diane Sotak, Chair Loretta Rielly Jey Wann Margaret Dull, ex officio University of Portland Oregon State University Oregon State Library University of Portland Ad Coordinator Non-Profit Oregon Library Association Inc. Organization PO Box 2042 U.S. Postage Salem, OR 97308 PAID Eugene, OR Permit No. 462

OLA Quarterly Publication Schedule 2009

The OLA Quarterly is the official publication of the Oregon Library Association. It is indexed in Library Literature & Information Science and Library, Information Science and Technology Abstracts. Each issue of the OLA Quarterly is developed around a theme determined by the Publications Committee and Guest Editor(s). To suggest future topics for the OLA Quarterly, or to volunteer/nominate a Guest Editor, contact the Publications Committee by visiting http://www.olaweb.org/.

Vol./No. Theme Deadline Pub. Date Guest Editor

Vol 15 • No 2 President’s April 15, 2009 June 15, 2009 Mary Ginnane Summer 2009 Conference Issue Eugene Public Library

Vol 15 • No 3 Oregon’s 150th: July 15, 2009 September 15, 2009 Elizabeth Nielsen Fall 2009 Libraries Then and Now Oregon State University & Teresa Landers Corvallis Public Library

Vol 15 • No 4 Across the Great Divide October 15, 2009 December 15, 2009 Robert Schroeder Winter 2009 (a special OLA/OASL & Ruth Murray joint issue) Portland State University