A Brief Primer on OpenURL, Link Resolvers, and Their Use

For a month before school started I interned at a corporate library in the Chicago suburbs. Part of the job I was tasked with was researching and writing summaries of technologies the library was thinking about implementing. With a few minor edits, here's what I put together for my boss about Open URL and link resolvers:

---

This presentation, given by Cindy Trainor at Computers in Libraries 2005, is a comprehensive introduction to what Open URL and link resolvers are and how they work.

Here’s a quick summary from her presentation and other sources:

Why use OpenURL?

Without OpenURL:
A patron searches a source, which may or may not contain links to the target (or the target itself, e.g. fulltext); the patron takes that citation, then searches for the target in the library catalog, ejournal A-Z list, other databases, google, etc.

With OpenURL:
A patron searches a source, usually an Abstracting & Indexing database, and finds citations for items that are needed (targets). In between the sources and targets sits software called a Link Resolver or Link Server that accepts links (openURLs) sent by the sources and presents the user with links to individual targets.

What does an OpenURL look like?

http://ry6af4uu9w.search.serialssolutions.com/?SS_Source=3 &genre=article&sid=ProQ:&atitle=TechnologyTrendsforIntranetLibrarians &title=Online&issn=01465422&date=11/01/2004&volume=28&issue=6 &spage=45&SS_docid=000000735316291&author=DarleneFichter

• BaseURL: http://ry6af4uu9w.search.serialssolutions.com (This means that the library is using Serials Solutions as their link resolver. This could also point to EBSCO’s service, or Elsevier’s, or whichever one the library chose to contract with.)
• Article Title: Technology Trends for Intranet Librarians
• (Journal) Title: Online
• ISSN: 0146-5422
• Date: 11-01-2004
• Vol. 28, Issue 6
• Starting page 45
• Author: Darlene Fichter
  

Once the OpenURL is sent from the search page, what happens next?

The link resolver now checks the library’s knowledgebase (where you’ve told it what subscriptions you have and how far back they go). Most link resolver products will have a web interface where you add the subscription packages and individual journals that you have in your library’s collection. If the data in the knowledgebase is incorrect or incomplete, your users might not be able to get to journals you subscribe to, or they might be presented with a link to a journal that isn’t in your holdings. Most link resolvers will allow you to add your print holdings to the knowledgebase too, so that users will know when you have a print copy in your collection.

Next the link resolver figures out if the OpenURL provided enough information for it to link directly to the article. For example, some journals may require an OpenURL to provide the journal title, article title, volume number, and ISSN in order to create a direct link. Some journals may need all this plus the author’s name to be able to link directly. So, depending on the requirements of a particular journal, users may see some results that link directly to the article level, while some link to the issue level.

How to choose a vendor:

Before you start looking at the specific features the vendors provide, think about two issues:

1. Longevity: Does the vendor have enough customers that they’re going to be in this for the long haul, or will they drop their link resolver as an unprofitable side venture?
2. Accuracy and depth of knowledge base: The vendor with the most customers can probably put the most effort towards keeping the knowledgebase up to date, and the more libraries that use the product, the more people will be reporting errors.

(Summarized from a useful post on the Web4Lib mailing list: http://lists.webjunction.org/wjlists/web4lib/2005-June/037368.html)

Comparing vendors:

This article from the October 2004 issue of Computers in Libraries provides a comparison chart of all the major link resolver vendors. Several other sites pointed to this as the most thorough comparison out there. The consensus seems to be that ExLibris’s SFX is the most used, followed by Serials Solutions’ ArticleLinker and Endeavor’s LinkFinder.