Session A4: Bringing the Library to the User: the Practice

Saturday, July 15, 2007

Presenters: Casey Bisson, Plymouth State Univ.
David Lindahl, Univ. of Rochester Libraries
Emily Lynema, NCSU Libraries

This session was coupled with "Bringing the Library to the User: the Theory" (which I didn't attend due to conflicts) but worked very well as a standalone introduction to some of the more innovative approaches libraries are trying to bring catalog data and other information resources to "where our users are" -- which is basically on the Internet, and comfortable with Google and e-commerce search interfaces. Each of the three presenters gave the audience a tantalizingly brief overview of their projects -- it would have been great to have more time with each project and lots more time for questions.

The main takeaways I heard were:

• We need to let users search the way they already know how to search, not the way we want them to search -- and how users know how to search is with a simple ("Google") search box, relevance-ranked results, and faceted browsing within results (e.g., if you type "worms" in ebay, you can then choose to look further under "bait" or "video games"). Faceted browsing makes use of the data we already have in our records (subjects, genres, formats, regions, languages, etc.) only now presented in ways that are easier for users to work with.
• We need to employ open standards, including standard markup of our data so it can be re-used in a variety of applications (some of them not yet invented)
• We need to allow users to interact with our data -- to add their comments and to re-use it in ways that meet their needs
• We need to include more resources than just items with a MARC record

Casey Bisson (presentation is at http://maisonbisson.com/blog/post/11879/#presentation-bringing-the-library-to-the-user) is the developer behind Scriblio (formerly WPopac -- http://about.scriblio.net/about/) and provided my favorite remark of the conference: "We get one chance to prove we're not stupid." If our interfaces are too hard to use, and don't pull up relevant resources the user will leave, pronto.

David Lindahl (presentation is in the AALL Conference Handouts, http://programmaterials.aallnet.org/) explained how he is drawing on specialists in anthropology, computer science, graphic design and other fields to solve problems users have with using technology. The focus needs to be on the task the users are trying to accomplish, not the tools. Very often librarians have it backwards -- we're too focused on tools themselves, not the task the tools are supposed to be supporting. His "C4" prototype runs on top of a Voyager ILS but the plan is to also have standard APIs to integrate with home-grown systems. (See http://www.library.rochester.edu/C4 to play with the prototype.)

Emily Lynema (handouts also at http://programmaterials.aallnet.org/) was the last of the presenters, and as we were pressed for time unfortunately had to skip over many of the slides in her presentation. NCSU built their interface using Endeca, which is not a library-focused company -- it provides search technology for e-commerce sites such as Barnes and Noble, Wal-Mart, Circuit City, etc. Their implementation sits on top of their SirsiDynix Unicorn system. See http://www.lib.ncsu.edu/catalog/ for their implementation.

The Q and A session afterwards brought a lively discussion on several points:

• These projects were done without much (if any) assistance from the ILS vendors. Emily remarked they did the project without consulting with SirsiDynix beforehand, and David said that while C4 has the verbal support of Ex Libris, development is proceeding without significant help/input from them.
• While open source projects are intriguing, it may be best for the time being to "let the ILS be the ILS" but improve the front end user experience through projects such as these three instead of trying to develop an entire open source integrated system. There is also a whole culture of successfully running open source software communities and libraries are amateurs at this; we have a steep learning curve ahead to fully integrate open source into our libraries.
• Because circulation statistics are declining while use of online resources increases, we may eventually need to use "most e-mailed" (or "most linked to," or some form of citation analysis) instead of "most checked out" to indicate the popularity of items.

These three projects (along with OCLC's WorldCat Local, which I saw a demo of in another session) make it clear that libraries are no longer waiting for ILS vendors to make significant improvements in the user interface. We know what our users need and use, and are forging ahead to bring it to them as quickly as possible.