Designing a better library also means designing better library systems. If our search systems are broken then that reflects badly on the rest of our operations and results in a degraded user experience. For many individuals their only interaction with the library may be through the online library catalog, the website or a specific electronic database from EBSCO or ProQuest. Even though the latter may be commercial products over which the library has no design control, they must still be considered a part of our total user experience.

I have written previously here and here about the challenges of bringing simplicity to library systems for research tasks that almost always involve some degree of complexity. For example, all of the incoming freshmen at my university take a course in which they write several research papers. One requires them to compare the media treatment of two individuals, possibly celebrities or politicians, across different countries or time periods. This is not simple. It requires a variety of knowlege at multiples levels. Yes, I’m sure the students would like to instantly find articles from which they could gather and synthesize information into an essay, but it just doesn’t work that way. Some critical thinking, some actual searching of library systems and some analysis is involved. For freshman accustomed to webpage/wikipedia cut and paste jobs, this complexity can be a rude awakening.

It is easy for librarians to acknowledge that our search systems must be simpler and easier to use. It’s an almost no-brainer type of statement, and one sees this all the time in librarian blogs and in the library literature. I’ve probably said it myself. But saying it needs to happen and actually making it happen are two different things. Even if we could simplify all of our systems would that be the right outcome? According to Don Norman, the design expert, simplicity is not necessarily the right answer or the goal. He writes:

I conclude the entire argument between features and simplicity is misguided. People might very well desire more capability and ease of use, but do not equate this to more features or to simplicity. What people want is usable devices, which translates into understandable ones. The world is complex, and so too must be the activities that we perform. But that doesn’t mean that we must live in continual frustration. No. The whole point of human-centered design is to tame complexity.

What is Norman’s solution to the simplicity-complexity conundrum? Design, of course. He states that “complexity can be managed”. I agree. He mentions that three simple design principles must be followed: (1) use modularization to take an activity and divide it into small, manageable modules; (2) mapping in order to make sure the relationship between actions and results is apparent and understood; (3) provide an understandable, cohesive conceptual model. So where does this leave librarians? What if Norman’s three priciples, particularly modularization, could be adopted for commercial database projects – and it follow more of a step-by-step approach – modularizing the search process.

Assuming the user has a reasonably well thought out search approach I envision five modules for completing a search. The modules are simple and guide the user through the process. They are:

The interface would be more user friendly and intuitive, prompting the user through each module. This might work by presenting the user with a screen for a basic search. Norman points out the importance of meeting the individual’s need to feel in control of the device. Prompts could be constructed to keep the system understandable yet allow the user to make decisions that provide that feeling of control. As each module is completed Norman would also say that a feeling of accomplishment must be delivered. I could see possibilities for that as the user moves through each module. Normal also says that the system requires “continual, informative feedback that can be pleasurable” and mentions Apple products as examples. One area where current library research products fail their users is they lack the design for letting users know where things are going wrong and what to do about it. This requires designing in expectations about problems users will encounter, and suggestions for how to improve search performance. Perhaps that requires some sort of sophisticated back end programming that isn’t available just yet.

While there are no exact answers here, one thing that should inform our thinking is that by stating library research tools need to be simpler we get no closer to making that happen. It’s a nice idea, but I agree with Norman that the focus should be on making the systems usable and understandable, and that design can help manage the inherent complexity of these tools. Norman also mentions that tools should also be enjoyable. While that’s a nice goal – and it certainly would be desirable to have library users looking forward to doing their research – I’m not anticipating that we can make the process a joy to experience any time soon. I do know that students and faculty who are passionate about their reseach are often passionate about the library research process. They don’t see it as a necessary evil, but rather as an exciting discovery experience. How do we tap into that passion, and exploit it for the majority of our users? That’s a subject for another post.