Feb
Carl Grant indirectly led me to a presentation by Jane Burke, Senior VP at ProQuest, at the Charleston Conference. Adam at the K-State Libraries Blog provided a nice summary of Burke’s talk:
Burke presented on web scale searching solutions for libraries. The presentation was founded on some firsthand research by Serials Solutions into the research habits of students. This research basically showed that library pages were essentially impenetrable and unusable with their siloed databases and indexes, lingo, and high learning curve for users. The web scale solutions Burke proposed generally include full content indexing of library holdings (catalog, aggregated content, hosted content, IR, etcetera) and searching with a single box interface ala Google. Watch for these solutions which are hitting the market.
Here are two telling quotes from a recent research study cited in the presentation:
“It is also clear from teacher and student responses in the study that the library is seen as an intimidating and inconvenient place, especially and interestingly in its primary purpose - supporting student research and often assisting students in the identification, location, and evaluation of sources.”
And:
“We also concede it is understandable that students are drawn to using search engines on the Internet to conduct academic research. These engines are easy to use, available to anyone with an Internet connection, and quick and bountiful in their returns.”
So, students would prefer to use Internet search engines over the overly-complex library search environment. But, Burke notes that library content users acknowledge that library resources are far superior for credible content than web searches and prefer them for coursework. Burke summarizes the problem well:
Even though the library is perceived to be better… The use of Google and other internet search engines continues to rise
Burke’s research shows that roughly 65% of users find Google to be “the easiest place to start the research process” vs. roughly 20% of users finding library databases the easiest place to begin research.
Burke’s solution to the problem is to give users a discovery service with a Google-like interface so they can search their library’s holdings plus tons of publisher content. I imagine Serials Solutions’ Summon would be high on her list since Burke’s company ProQuest owns Serials Solutions. Burke’s solution makes sense with the caveat that discovery services will need to integrate with federated search engines to fill the content gap that is inevitable when a library commits to only providing access to scholarly material from sources available in a particular discovery service.
Tags: federated search
2 Responses so far to "Jane Burke on how to “sell” to library users"
February 14th, 2010 at 9:30 pm
But the lesson of that is that if including broadcast fed search in your ‘discovery service’ impacts it’s responsiveness and ease-of-use… use of it by your users is going to go WAY down.
The challenge is narrowing the ‘content gap’ while still providing as high ease of use as a Google-like service. Something like Summon certainly has the promise to increase the content while decreasing the usability problems. I’d only want to include broadcast search in it if it could be done without having any negative impact on ease of use and responsiveness, otherwise you’re right back where you started.
Don’t forget, as I’ve suggested on this blog before, it’s disingenous to suggest that there’s no “content gap” with broadcast fed search too. We have MANY library (licensed) databases which content which are not going to be available through a broadcast fed search tool either. (While it may be theoretically possible to include all of them, it’s an expensive proposition). So we’ve got to find solutions which find the ‘sweet spot’ of maximizing appropriate content inclusion while minimizing usage problems. It’s not at all self-evident to me that broadcast fed search necessarily has to be involved in that sweet spot for scholarly research use cases.
February 23rd, 2010 at 1:54 pm
“with the caveat that discovery services will need to integrate with federated search engines to fill the content gap that is inevitable when a library commits to only providing access to scholarly material from sources available in a particular discovery service”
I’m sure Jane would say (and I concur) introducing federated search lessens the ‘google-like’ experience. It slows down result retrieval, moves deduplication into the retrieval process, and leaves you at the mercy of individual vendor interface availability and response time.
Better we should help publisher’s understand that licensed users of their metadata should be able to ingest it into discovery tools as it’s a win-win, libraries get the RFI on their content, and publishers don’t loose our money because we’re aren’t getting sufficient RFI.