I Don’t Get Discovery Platforms: Are We Letting Quantity Win Over Design? by TTW contributor Troy Swanson 4

Every ILS and database vendor at ALA Annual seemed to be touting their new flashy, single-search discovery tool that groups together all kinds of information sources in a list of search results. Discovery is the hot topic, and your library surely doesn’t want to be left out in the cold. The sales folks have been putting on the full-court press within higher education, and I assume also in public libraries. After leaving ALA, I just don’t get it. The hype doesn’t seem to match the impact.  I struggle to see who these tools benefit.

Who’s the Audience?
Over the years, my library has completed two formal usability studies focusing on new community college students. One resounding lesson from these studies is that students are poorly prepared to recognize differences in information sources on the screen. If new students aren’t really the target audience for discovery tools, then maybe these are really aimed at faculty members and researchers? I am skeptical. Most experts find sources through consulting the literature regularly, contacting colleagues, and attending conferences. They rarely sit down and search a topic from scratch (see Soo Young Rieh, “Judgement of Information Quality and Cognitive Authority in the Web” for an early but useful discussion).

Format as Value
When we consider search from the information literacy perspective, discovery tools also seem to be a move in the wrong direction. Lori Townsend, Korey Brunetti, Amy R. Hofer, in their work on information literacy threshold concepts,  have found that the understanding of “format as process” to be foundational to understanding research. This means that information literate individuals recognize that the format (news, peer-review, books, web pages, etc) provides an indication to the process used to create the content. This, in turn, contributes to authority and credibility of sources. Format is process, and process is value, meaning, and applicability to need (see Lori Townsend, Korey Brunetti, Amy R. Hofer, “Threshold Concepts and Information Literacy“). The idea of adding value to the research process by requiring searchers to sift through long lists of results seems problematic.

Quantity Hides Quality
Not to mention that user behavior studies indicate that quantity obscures quality. It is pretty well documented that most people rarely click past the first 10 results in a Google search despite the fact that most searches return millions of results (see Danny Goodwin, “Top Google Result Gets 36.4% of Clicks”). Yet, discovery is being sold as a benefit.

Design Thinking
I am willing to admit that discovery platforms may not be that much worse than the search interfaces we already have, but they don’t seem to be much better. They especially don’t seem much better considering the price. Why pay tens of thousands of dollars for something that is just as bad as what we already have?

I wish ILS providers thought more about user interfaces as opposed to search results. Have they really thought about how the user experience might work beyond a single search box that pukes back 1000s of results? The vendors in the ALA exhibit hall gave me the feeling that they had invented a secret weapon to win the technological arms race, but I increasingly wonder if our challenge is not about technology at all. What if this is really about design? What if the thing libraries really need is design-thinking (IDEO-style) focused on how we lay out access pages that are more than just single-search boxes? ILS vendors are missing the real market.

For example, the article by Lown, Sierra and Boyer in College & Research Libraries takes a step toward a single-search option that rethinks how results are displayed.  Perhaps breaking results down into distinct panes is a direction that warrants more exploration?

swansonphotoI know that many libraries have discovery in place so I’d love to hear about your experiences. Currently, my library staff is seriously contemplating our next steps for our ILS.  To me, bringing together these disparate tools is one the most significant challenges that we face. Who is innovating around this? What’s the next step that focuses on design?

(Thanks to Eric Phetteplace for his conversation on this topic at ALA and for reading an early version of this post.)

Troy A. Swanson is Department Chair and Teaching & Learning Librarian at Moraine Valley Community College. He is the author of the book, Managing Social Media in Libraries. You can follow him on Twitter at @t_swanson.

4 thoughts on “I Don’t Get Discovery Platforms: Are We Letting Quantity Win Over Design? by TTW contributor Troy Swanson

  • Michelle Holschuh Simmons

    I’m so glad to hear your thoughts on this topic. I have had the same response to discovery tools, and since so many libraries are moving in this direction, I thought I must be missing something. It is hard enough for students to understand the distinctions among types of information and this confusion is compounded when all of these different types appear in the same results list in the ILS. I am all for progress in the LIS world, but progress without keeping in sight the needs of our users seems short-sighted. Thank you for raising these issues here.

  • Andrew Spencer

    I agree that the number of results that discovery layers can return can be overwhelming for users. When I’m showing them how to use the system, I explain the various formats and how they differ from each other. We usually refer our undergraduate students to our discovery tool, which is both good and bad. Good because they get an idea of the resources that are available to them. Bad because it’s easy for them to get discouraged when they have to wade through pages of results. My other concern is the fact that none of the discovery layers are comprehensive – vendors don’t allow their products to be indexed by their competitor’s products. This is why I still refer researchers to individual databases for their searching, so that they don’t miss out on any content. Thanks for the thought-provoking post.

  • Charlotte Clements

    We are planning to implement a discovery tool at our research organisation in the next couple of months. I agree, a thorough literature search is best done using the search engine designed for a particular platform, however it is my experience that researchers prefer to work independently to find relevant literature as quickly as possible, and that they are generally unaware of the platforms they use (two examples just in the last week of help required because they clicked on the wrong publisher link from a Google search). Perhaps unlike university research, new subject areas are often explored by researchers here as new projects are set up and so a speedy entry to the literature is highly desirable. We are keen to expose all our subscribed material and to merge those results. It’s what happens next that is the perplexing part, as you and others note. My note to our vendor at our initial meeting was that you can search almost as well using Google, but we need a means for users to organise the results, and that is why we will pay for a discovery tool.
    It is well past time studies are done to figure out how users are best served by the search tools available to them and also how we librarians can teach the necessary skills. Now the first wave of “Google-like” search interfaces is on shore, attention can be paid to refining the tools that deliver meaningful filters for users (not librarians) – try asking your local library science researchers to pay attention to that.

Comments are closed.