OpenURL, Federated Search Tools, One-Stop Shop and
Why Libraries are Crazy about them.
By Rey P. Racelis, Assistant University Librarian for Systems Integration

Sometime ago in the early part of spring, a Pace student commended, with some reservations, the very useful features of the interlibrary loan service offered by the library namely, the ILLIAD (Interlibrary Loan Internet Accessible Database). The service allows users to have a better control of their interlibrary transaction, follow-through and monitor the status of each request and even access the full-text of the requested articles online--anywhere in the world, anytime (theoretically a 24x7 service unless the server is undergoing maintenance or has been shut down for one reason or another). The service is intended to empower interlibrary loan users and render them in better control of all their borrowing transactions.

Under ordinary circumstances, a user goes to a database, initiates a search and checks the online catalog--the web opac--to see if the journal title containing the article is available in the library. If not, the user refers to the Illiad service, calls up the required form and starts filling-up the required bibliographic citation information. To do this, in the easiest way possible, a user has to keep a minimum of 2 windows open concurrently--the window containing the form and the window containing the bibliographic citations from whilst he copies the citation information either through "copy and paste" or retyping the information on the form. It was a long way, indeed, since a couple of years ago when printed forms had to be filled in by hand. Or so we thought.

Not good enough! according to the student.

Can the library come up with technology that allows transposing the citation information into the form instantly without necessarily copy pasting or retyping the information? In other words, automatic completion of forms without as much as a click or double click of the mouse!

Requests such as this, when received by the systems staff, do not invite mere scoffing at the idea. Nor is it relegated to those files of undoable, or unnecessary requests received daily from the researching public. Rather, these are occasions when staff re-evaluate services and scour available technology to find out if such suggestions are doable, and if so, if it can be implemented in the library given its resources---staff skills and facilities to implement or maintain among others.

It was opportune that such suggestion was put forward at a time when a new information processing technology has just been evolving and being perfected. A library systems administrator and former doctoral student from Belgium by the name of Herbert Van de Sompel, wrote his PhD dissertation studying and developing tools that allow for the fluid exchange and linking of related information through the use of linking servers and interoperability specifications which later came to be known as OpenURL. In layman’s parlance, the technology enables any pieces of bibliographic information -- the metadata--to be transferred from a database or information source to another service component which can utilize such passed on information (the passed metadata) and process it further to produce another form of service.

Thus, in the case of the Illiad, a piece or several pieces of metadata (such as author, title, ISBN, ISSN) are passed from a given bibliographic source into the Illiad service. The information becomes the data elements processed by Illiad to determine what or which material is being requested. In this case, the passed elements are parsed, analyzed and interpreted following bibliographic standards to determine what material is being requested by the person who initiated the request. This is all made possible using an OpenURL syntax which, as I understand it, is completely web and http based. This passing of data is done instantaneously with the click of a mouse.

It is not within the context of this short article to analyze the details of the technology. Suffice it to say that "OpenURL allows for interoperability by providing a simple and consistent way to identify where any item is found and how any item is described". [David Stern, "Automating Enhanced Discovery and Delivery: The Open URL Possibilities" in Online, March 2001, online copy]

In layman's parlance, OpenURL therefore allows for interoperability. In the case of Pace library experience, OpenURL has been implemented quite rudimentarily between the ILLiad service and a number of databases from 3 different vendors and/or information service providers, notably First Search (about 24 subscribed databases), EBSCO (about 14 databases) and Proquest-UMI (about 6 subscribed databases). Of the 3 database providers, the interface between FirstSearch and Illiad works quite smoothly (understandably facilitated by the fact that both First Search and Illiad are 2 products either maintained or supported by OCLC). At the moment, Proquest, while capable of passing on the information to Illiad, is still trying to resolve the issue of date format, the Illiad being programmed to receive any passed date information in the format of yyyy/dd/mm whereas Proquest is passing information in the format of dd/mm/yyyy. Other than this specific problem, the Proquest-UMI databases appear to be also OpenURL compliant.

Despite the limited number of information providers that are OpenURL compliant, gauging from what has been so far implemented, the feedback is rather positive. The process of automatic passing of information helps speed up research and provides for easy submission of interlibrary loan forms.

The possibilities in the use and implementation of OpenURL are, however, endless and remain to be exhausted. The Illiad example is just one of them. As David Stern puts it: the technology can make it possible to have a "result page containing links to hundreds of related items, regardless of location" and "media format" [ibid.] thus providing for a rather rich list of provenance of sources and services, and actual data as well.

It is in this context that such rather forward looking information facilitators and providers have utilized the technology to push the process of information research and provision to new heights of enhanced results contents.

So far, Ex Libris, with their product called SFX (special effects), seems to be the frontrunner in its implementation. Using OpenURL, it strives to link disparate sources and services all for the purpose of providing an easy "one-stop shop" in one's quest for information. It makes it possible to create a "hook" or link for an information item towards a number of services such as interlibrary loan, bibliographic enhancing tools such as Syndetics solutions, Informata (on a standstill at the moment?) and RealRead, the online catalog, the national catalog such as OCLC or the Library of Congress collection, or any of the existing document delivery services such as Ingenta, Infotrieve, or CISTI (Canada Institute for Scientific and Technical Information).

Below is an example of a screen where through the use of SFX technology, various other services are linked and called into single screen from a hook attached to bibliographic information as shown in the top box:

To see an interactive sample, you can also go to the Ex Libris’s SFX site: http://www.sfxit.com/

The process will show the versatility of an OpenURL technology and how it facilitates further the process of research.

EBSCO has a comparable service called LinkSource (still being demoed in the EBSCO subscribed databases for Pace Libraries.) and Serials Solutions has also its own comparable product in their package suite called Data Management Services.

Speaking of "one-stop" shop, it seems that OpenURL technology is a step advancing one more notch the achievements gained by implementing a parallel or perhaps an earlier technology referred to commonly as federated searching.

It is indeed this concept of federated search tool that gave way to the concept of one stop shop, albeit not in such diverse manner as made possible by the implementation of OpenURL. At best, what federated search technology made possible is to allow a single search query to be deployed across a number of databases, whether belonging to the same protocol or platform or not. Results are returned either cited several times, merged or de-duped. In short, it has made possible the integration or simplification of access to various pieces of information. This is rather a significant feature in itself. But the results generated could not be further parsed, to my knowledge, into further tiny pieces of data that can be fed into other forms of services, by way of seamless and continuous passing of data from service to service until the advent of OpenURL technology.

Among the different services that provide for this kind of federated searches, WEBFEAT is one of them (and Pace libraries one of the first implementers). Other players are Endeavor, Ex Libris, Fretwell-Downing, and MuseGlobal. Noteworthy is the fact that Endeavor is also a systems integrator providing for such library technology as integrated online system. In the case of Pace libraries, its own integrated online system, the Innopac System from Innovative Interfaces, Inc, Emeryville, CA, has also started offering a product that purports to offer this sort of federated searching feature in its own MetaFind webpac tool (part of its digital product package called MetaSource: Digital Collections Management product). These same vendors are also involved, one way or the other, in the delivery of products that are OpenURL compliant. This also shows the close relationships between federated searching and OpenURL technology.

At any rate, one of the more insightful discussions of federated search tools is the one written by Roy Tennant [see Roy Tennant, "The right solution: Federated search tools" in Library Journal, New York, June 15, 2003, online copy].

To step back a bit further, even this concept of federated searching is antedated by another library implementation that has been around for sometime but has not caught on due to the complexity of implementation. The technology referred to is z39.50 protocol. See the following website for a detailed explanation of the technology:

http://www.cni.org/pub/NISO/docs/Z39.50-brochure/50.brochure.toc.html

This said searching protocol deploys a search query to a number of z39.50 compliant databases, (called a broadcast search) including online catalog, and brings forth results in batches, painted on the screen ( that is delivered to the screen as x number of postings). One caveat in z39.50 searching is that it can only search z39.50 compliant servers. Lately, the other federated search engines and tools, deploy searches beyond a single protocol. In the case of webfeat, it can mine data both from z39.50 compliant databases as well as others that use other forms of protocol.

So this concept of one stop search and shop that seems to culminate for the moment with the use of OpenURL traces its development way back to other technological endeavors that build, incrementally, to produce the present state of the art in searching technology, information processing and delivery that is getting a firm foothold in many libraries and other research institutions and services.

The popularity of the different technologies cited above among libraries owes much to the growing awareness that with the advent of digital technologies, more and more information is being made available on the web. The growing number of information available online, the varying provenance of sources, the multifarious formats of data, make it imperative that somehow the search engines and tools that are used to mine such online wealth of information and then transform them into a much more organized and comprehensible fashion, be made available. They can help give shape and sense to all these collated data in a more integrated fashion, possibly de-duped, logically merged when possible, and perhaps made portable from one platform to another, thereby providing the user with some sensible or comprehensible data that can be easily used, mentally "digested", speedily processed, and when useless, easily discarded.

This "one-stop-shop" technology therefore traces a development continuum that rests on the simple premise that the digital world, while allowing for increased access to a whole lot more of data, has also precipitated an information avalanche, necessitating the development of tools to sort, sift, and find meaning among this mass of knowledge, thus making it more useful to the information consumers while at the same time facilitating the exchange, transport and delivery of information.