Greane Tree Technology is the successful bidder for the redesign project. In advance of the first meeting with them, A. Downing held focus groups with students to gather data to inform the initial discussion. Here are some of the findings:
What students already go to the web site to do (and we should make very simple to accomplish):
1. Find Books – Textbooks for courses and books recommended by instructors (i.e., known items), but also items that they are seeking on their own to support their class work. Students do not think that we should give equal attention to other formats (DVDs, CDs, etc.) on the home page. They also need to check the status of what they have found and perform functions such as renewal. Students want an easy way to search for e-books. It was also clear that they tend to believe that all e-books require an e-reader device that is apart from their laptop or desktop computer. They want to know what books have been recommended by their peers (especially previous students in a course) more often than a librarian or instructor.
2. Search Databases -They use only a small fraction of our total titles. Federated search interests them greatly.
3. They look for tutorials, but think of them as help sessions. They look for help in areas that may not be associated directly with the library (e.g., Excel).
Other factors to keep in mind:
1. Keep text to a minimum. Whenever possible, use images and icons to convey information. They do not linger on the home page long enough to read lists.
2. Locate search boxes at the top of the home page.
3. Locate library hours at the top of the home page.
4. Place a reference chat button on pages where assistance may be required, such as at the results screen of a federated search.
5. Arrange services by those most heavily used. Order all lists by frequency of use or in some cases alphabetical.
Next Steps:
Greane Tree will supply a map of the current site so that we can make decisions about organization. We will provide them with a description of how we intend to handle the functions in the current site that rely on custom code. BCTC has identified superior substitutes. We will give them access to a collection of images that we would like incorporated into the site. We will identify the first 10 announcements (graphics with text) that will occupy the main slide show on the home page.
no mention of any kind of directional information needed? Phone numbers? address? lists of people? maps? collection locations? Do we care about any other users besides students? the most common thing I go for when I go to another library’s website is a staff directory and services numbers. I wonder if some of this stuff would be a “help session”. I like that name.
In the student groups so far those items were not mentioned. This was only the first set of meetings to identify the items at the top of their lists for the vendor. Throughout the redesign process we will be going back to students to add, refine, and test the elements of the site. Other user constituencies will be involved, but the student perspective will dominate, especially on the home page.
I’m intrigued by the recommender systems they seem to want. I wonder if we can extend bX to spaces beyond the SFX menu. I wonder if there are discovery layers we could invest in that might help us do this. FWIW, there is an academic library in the UK that reuses circulation data for a recommender system for items in the catalog. Dave Pattern, the librarian who set up the system for the library at the University of Huddersfield has a nice blog post explaining how it works and providing links to examples from the catalog.
I thought it was interesting that students seem to want to be able to segregate resources by format. I think that runs counter to the format-neutral approach libraries have been going for (i.e., present everything then let them choose the format they want).
bX Recommender is article based only but others have either linked the catalog with sites like LibraryThing or modified the catalog to allow for tagging. It may be wishful thinking on my part because I can’t find it now but I seem to remember that user tagging was one of the eventual goals with IDS Search.
Dave Pattern’s article is great but I am mostly impressed at the number of circulations his collection seems to have – the only books that get such high circulation here tend to be reserve ones. This makes me wonder about the quality of the results we’d get.
This brings also an interesting dilemma – the students surveyed wanted to know what people who took their class in the past have used, but that assumes those students did take out books themselves. If they did that at a very low level (each student took one book out for example), the recommendations might in the end not be very valuable and we might run into privacy issues (it might not be very difficult to figure out that the student who took out the book on abortion wrote the paper on that topic).