The RUSA/MARS Local Systems & Services Committee invites you to join our discussion forum 'Discovery Systems: Solutions a User Could Love?' at Midwinter.

When: Sunday, January 17, 2010 from 1:30-3:00 pm
Where: Westin Waterfront, Faneiul Room (Mezzanine Level)
425 Summer Street connected to the Boston Convention and Exhibit Center

Panelists include:

• Marshall Breeding, Director for Innovative Technology and Research, Vanderbilt University who will provide a brief "Overview of Discovery Systems."
• Cody Hanson, Technology Librarian, University of Minnesota, who will briefly discuss "User testing and feedback on Primo at the University of Minnesota."
  
• Frances McNamara, Director, Integrated Library Systems and Administrative and Desktop Systems, University of Chicago, who will briefly share experiences of the "LENS Discovery System, based on AquaBrowser." and; 
  
• Barbara DeFelice, Director Digital Resources Program, Dartmouth College who will discuss "Summon @ Dartmouth College: the User View."
  

Our panelists will highlight the experiences of libraries that have implemented "next generation discovery tools" that provide access to disparate library collections from a single search box. Panelists will focus their comments on user response and subsequent assessment of the local implementation. 

Discussion forum participants will be able to share their experiences with discovery tools and ask questions following the panelists. A summary of the key ideas gleaned will be posted on the MARS Local Systems & Services web page following Midwinter.

Photo credit: The photo "Magnified" was taken by Jake Bouma (jakebouma) on March 9, 2009 and uploaded to Flickr on March 10, 2009 with an Attribution-Noncommercial-Share Alike 2.0 Generic Creative Commons License. The photo was downloaded on January 3, 2010 from http://www.flickr.com/photos/jakebouma/3345296623/ 

Found via Stephen's Lighthouse. 

You are more likely to use Wikipedia if you are male (39%), 18-29 (44%), a college graduate (50%), or make more than $75,000 a year (42%). Wikipedia has 24% share of traffic of the top 20 educational web sites. Google Scholar comes in at number 6, Google Book Search at number 7, and the National Library of Medicine (I guess PubMed) at number 9.

I don't think we're doing a good job promoting the electronic library resources available at the state level, like the Ohio Web Library. Anyone in Ohio with a public library card can get instant access to a bunch of EBSCO databases, the Encyclopaedia Britannica, and a lot more.

As easy as that is to say, we all know what the real issue is though. The reason Wikipedia gets so much use is due to the fact that it is dead simple to discover via search. The Pew report points out that Wikipedia articles have a high number of in bound links. Therefore, they display near the top of Google results thanks to the PageRank algorithm. 

A conscious decision was made to include Google and Google Scholar as part of federated search. The single federated search box includes the phrase, “Search library databases and Google at the same time," to get student's attention. Katy shared that during instruction she informs students that Google is not able to index the deep web (like subscription databases). By using Fox Hunt, students have a better chance of finding more relevant sources for their assignments. Usage statistics show a dramatic increase in the number of full text article usage and PDF downloads.  

Image searching was an additional benefit for Marist from their implementation of federated search. The college offers a fashion design/merchandising major. It was often challenging for students to find pictures of fashion for class projects. The Library's solution to this problem was to include the NYPL Digital Gallery, NYPL Picture Collection Online, and American Memory web sites as part of the fashion design/merchandising federated search. Katy indicated that the descriptions in American Memory are very detailed and often include clothing descriptions. 

The conference paper, presentation, and demos of Fox Hunt are available on Marist's web site.  

I was most impressed by their decision to include the image collections in federated search. Image searching is not currently included in OhioLINK's federated search tool. I am not sure if it has been considered, but worth suggesting. Ohio students (and all Ohioans with a public library card) are able to search across multiple art collections within the Digital Media Center. However, it would be nice to include other open digital collections in the OhioLINK federated search tool.

technorati tag: acrl2007

Jeffrey and Jared developed a search problem and had students at three different BYU campuses (Utah, Hawaii, and Idaho) find information using regular A&I databases and BYU's federated search tool. They wanted to know if federated search saved students time and if students preferred federated searching over traditional A&I database searching. Iris has a written a more descriptive post of this session. Jeffrey and Jared indicated that they are writing a paper for publication based on their research and are still analyzing data.

The initial outcomes from their research are what you expect. In general, federated searching saved time (roughly 10% for the students in Utah) and they preferred using federated searching (70%) over traditional A&I databases. The data is different for each campus, which could be related to how the search tool has been customized and implemented locally on each campus. Their research was also focused on one discipline, biology, and other disciplines may experience other results.

Our students and faculty at Muskingum College heavily use OhioLINK's federated search tool (Metalib).  Since August 28, 2006, the beginning of our Fall semester, there have been 10,740 searches executed. Our campus is currently ranked 12th (behind Bowling Green State University, ahead of Ohio University) in usage from August 28. This is attributable to having the federated search box front and center on our web site.

I know some librarians believe we do our students a disservice by providing federated searching. I believe that it is better to give them a "Google like" tool so that they can find quality articles from journals and magazines vs. the questionable content on the open web. This is especially necessary when many of our students wait until the last possible attosecond to start their research and tend to take the path of least resistance to full-text.

What do you think? Is federated search a necessary evil?  Have you had good experiences or bad experiences? Do you teach it during instruction? 

technorati tag: acrl2007