:: Digital Libraries Columns


Library Journal "Digital Libraries" Columns 1997-2007, Roy Tennant

Please note: these columns are an archive of my Library Journal column from 1997-2007. They have not been altered in content, so please keep in mind some of this content will be out of date. :: Digital Libraries Columns

Evolving the Resolving


In the ubiquitous web era, it seems strange that linking well can be such a problem for libraries. But it is. The same content can be licensed from different vendors, so it's key to get the user from search results to the licensed article. But the solution could be more elegant. The development of the OpenURL specification and resolver was clearly a breakthrough. An OpenURL can indicate all the metadata components of an article such as title, author, volume, issue, date, page number, etc. The resolver software then checks its database of library licensed content and tries to find a match. If it does, then it can provide a link to that article. If it doesn't, the library can configure its resolver service to put up a menu of options such as searching the catalog to see if there is a copy in print. Recent events bring to mind the need to improve the usability of linking services. Reducing clicks At the University of California (UC), we have implemented OpenURL linking the way most libraries have--which means whenever an OpenURL icon is clicked beside a search result, our OpenURL resolver responds in a separate window with a menu of possible options. This menu sometimes includes multiple options for obtaining the full text of the article, as well as options to search our catalog, submit an interlibrary loan request, get library assistance, or report a problem. In developing a UC infrastructure for librarians to create metasearch portals, we've performed a needs assessment. We've talked to our users to find out what tasks they wish to perform and how they try to accomplish them. It's no surprise they want to know whether full text is available without first clicking on an OpenURL link. Don't make me think Therefore, we're working at resolving our OpenURLs differently. When we get search results back from our metasearch application, but before we put them on screen, we check the first ten items against our OpenURL resolver for the existence of full text. If full text is available, we put a full-text icon on the screen that links directly. This lets users know right away what they can expect and also reduces the need to go to another window, figure out which choice is best, then click through to the content. Technology is not service Another event illustrates a different problem with the way we currently do link resolution. An open web-based reference management site recently announced OpenURL support. What this means is that users of that site can now add the base URL of their institution's OpenURL resolver to their personal preferences, and OpenURL linking will be enabled for their citations. Most faculty and students would be puzzled by the term OpenURL. Even if they have heard of it, do they know why they should care? Finally, if they pass those two hurdles, how do they discover the base URL of the appropriate OpenURL resolver? Try this at home. Act like you are a library user and try to find your OpenURL resolver base URL. Reasonable strategies might be to look on your library web site and do a Google search. Neither is likely to be very productive if your library is anywhere near average. Hiding the plumbing As great as OpenURL linking is, it's only necessary plumbing. And like real plumbing, it should simply work and be hidden behind a wall. All of the necessary pieces are almost in place to make this real. For example, a directory of OpenURL resolvers is necessary so users can simply specify their institution and not have to track down the base URL of the appropriate resolver. OCLC has established an OpenURL Resolver Registry for just such a purpose. In fact, it's already possible for someone on a campus with a resolver to just go to the OpenURL Gateway service, and if the resolver is registered, the right one will be selected based on the IP address. So although we've come a long way on our quest to guide the user to the appropriate copy of an article, we still have some distance to go before it works the way it must. The user shouldn't have to think about any of this--and it's our job to make sure users don't need to.

Link List