What users expect
We thought it would be fun to emulate some of our favorite sites in a lightweight concept discovery layer we call Libre.
Users have high expectations these days. The hours spent in elegant web apps like Netflix and Spotify seem to be sharpening the collective sense of design. What was once the pinnacle is now the convention, and as Don Norman said, “Conventions are slow to be adopted and, once adopted, slow to go away.” So we thought it would be fun to emulate some of our favorite sites in a lightweight concept discovery layer we call Libre. Below are some of the expectations we prioritized in the design. 1
#1: Things worth doing also look cool
First, we wanted to elevate books to the same “cool status” of other media. Thanks to Netflix and Spotify, that meant choosing a dark theme with white lettering and neon trim. Because of the ready association with the national library symbol, we chose blue for the secondary color.
#2: The most useful things are also the most visible
The intent in a known-item search (33–60% of all queries 2) is rapid visual confirmation, so we highlighted title, author, and cover image. In more serendipitous browsing, the intent is evaluation, so average rating and a synopsis are prioritized second. 3
#3: All the answers are here
Several friends of mine have revealed, at one point or another, that they didn’t know the library was free. While this can seem shocking, it’s bad design to assume that the user knows everything they need: immigrants may never have had access to a public library before, and the less tech-savvy might need to know that borrowing ebooks is legal. Hence, we avoided jargon like “Place Hold,” list requirements, and explained the basic premise of a library in fine print beneath the main call-to-action.
#4: Browsing is always assisted
Other sites deliver personalized recommendations by capturing reams of personal data. Content-based recommendations like “Nebula Award Winners” or “NYT Bestsellers — Fiction” assist users in a similar way, though. Offering a compelling alternative is more important at the library than anywhere else online, since the title a user came looking for could be out on loan already. We wanted to keep our users from leaving in frustration if they encountered an unavailable title.
#5: I can bring friends
A site without sharing is a city without roads. Even if the features aren’t used too often, we decided that it was important to offer up multiple options for users to save, share, and otherwise show off their discoveries. We distinguish subtly between casual users, who might know to post or tweet, and the power user, who may want to embed a free link on his book review blog, for instance.
What other expectations are important? What features did we miss? Feel free to play around with the Libre prototype and let us know on Twitter at @readlibre.
Endnotes
1: Our work in this article focuses on a popular reading use case, and will therefore seem more applicable to public libraries. Still, we hope our friends in academics get something out of it too.
2: EBSCO and Ex Libris are at odds over this figure. EBSCO says “Just under 30” and Ex Libris “over 50.” Both of them exclude author searches from their definition of “known-item” entirely, which seems to me a mistake. Often an author search is an easier route to a known item: for instance, when the title is so long as to be annoying to type or so short as to be ambiguous. Therefore, I inflate their estimates by about 5%.
3: Notably absent are Format and Availability. These are currently displayed after the user clicks “See at the Library.” A more robust implementation might have them both appear on the page.
