One of the problems with implementing linking standards is the “appropriate copy” problem, which is how to determine which copy of an article is the one to which a link should direct users. One solution would be context-sensitive links from databases to an institution’s resources by way of a publisher service. Another solution is to use a link resolver, like SFX. Link resolvers, as Boston & Gedeon (2008) write, were precursors of the OpenURL because they didn’t provide standards-based linking protocol. The benefits of using this approach are interoperability, resource discovery, and linking between resources in a personalized and user-friendly approach.
The primary concern when looking at all these technologies is interoperability, as Weddle (2008) points out that things like the link resolver will affect other things like the MARC record service. All are connected through a single point of data entry. Weddle (2008) does a nice job reviewing some of the tools available, two of which I list below with some notes for each:
- A-Z list: service provider responsible for changing global knowledgebase when resources added/removed, and librarian worries only about the local list, facilitate known-item searching
- OpenURL and Link Resolver: OpenURL via link resolver implementation provide crucial links among resources and is closely connected to A-Z list, allows patron to go from browsing citations and jump directly to the resource housed in another location, the source and target relationship,
Weddle (2008) then introduces the topic of federated searching, which addresses the need to search across many e-resources and helps with topic searching. Walker (2010) also discusses federated searching, or metasearching. “Federated searching represents a direct response to users’ behaviors and preferences…Through careful selection and grouping of library resources, libraries still provide value-added services—services that are not available via free Web searching,” writes Weddle (2008). And Walker (2010) agrees: “Web resources are more ‘discoverable’ when other sites link to them,” which is what CrossRef, DOIs, Link Resolvers, and OpenURL help.
The benefits of linking to the article level are that it makes citations actionable immediately and enhances efficiency of online research. In other words, the author and topic is more important than the publisher in terms of identification, so linking to the article level enables the system to go across all publishers and takes users to their websites and versions of the document. While it helps fix errors and serves as a business opportunity for publishers, libraries also benefit: “Namely, [DOIs] make linking reliable at any level of granularity possible. In the same way that links drive readers to publisher content, libraries may also see increased usage of acquired electronic resources as an additional benefit” (Brand, 2003).
Grogg (2006) takes a different route in his examination of finding content in ERMS, looking at reference linking. I pulled this excerpt from the article: “Internal linking may seem more attractive because, on the surface, there appears to be no need to address the appropriate copy. If a user has entrance to a content provider’s front door, then the user will most likely have access to whatever lies within the provider’s content database(s). That sort of reasoning is fault, however, because in addition to the large aggregator issue previously described, no one provider can hope to include all literature cited in every reference list” (Grogg, 2006). To help illustrate this point, Grogg (2006) likens internal linking to a shopping mall, and external linking as a downtown shopping district, where the difference is between closed and open environments. External linking (or intersystem) is needed when a user goes from one content provider to another.
Brand (2003) looks at concepts that reach beyond citation linking, things like
- Forward linking: I understand this to be the same as a cited reference search in a single search interface, as found in WOK.
- Parameter passing: using OpenURL syntax to send a key along with the DOI that enables extra functionality to benefit publishers and users (i.e. a parameter might be info about the source entity).
- Multiple resolution: user clicks on one link and can see a menu of publisher-supplied options
Walker (2010) also is forward thinking in a way that I found particularly appealing. Consider for a moment this excerpt:
“The behaviors and expectations of today’s users are changing, so make the effort to learn about them. For example, users take for granted, and expect, relevance ranking; they like the user experience to be intuitive, without compelling them to learn how to use the system; and they expect to be able to interact with the system by contributing, for example, reviews and tags that will help them and others find the information in future.”
It is interesting to me from a “behind the desk” experience because – with the exception of a handful of physicians I have assisted at the health sciences library at which I work – hardly ever have patrons asked me about how results are assessed in relevance ranking. But regardless of whether we think patrons should think about this, we as information professionals need to be aware of their expectations, their in-the-moment “This is what a link/library/resources should do” thinking. This might mean examining things like tagging and faceted browsing as effective search strategies, which might before have seemed like less academic, I guess you could say, methods of inquiry.
Image credit: Wikimedia by Petritap
No comments:
Post a Comment