Rebook | Research (Reading)

Master's Capstone Project, Spring 2015

Teammates: Ian MacFarland, Paul Son

 Participant reading a scrolling chapter

Participant reading a scrolling chapter

Challenge

  • Perform usability tests for ebook app reading experience
  • Recruit appropriate participants
  • Influence design decisions with concisely presented findings

Role

  • Write scripts for usability tests
  • Moderate and take notes for usability tests
  • Identify and present key findings

Summary of Research

  • Competitive analysis
  • Two rounds of usability testing for reading app

Competitive Analysis

Introduction

We split our analysis into four parts. The first was a comparison of the libraries, reading customizations, basic features, and progress visualizations of the four main ebook competitors’ apps in the US: Amazon.com’s Kindle, Barnes & Noble’s Nook, Kobo, and Google Play Books. In the second part, we focused on platforms for reading long-form nonfiction content. This covered the blogging platform Medium, the journalism reader Longform, the article saving app Pocket, and the publishing platform The Atavist. We separately considered the digital comic reading app Madefire. Finally, we looked at the writing app Scrivener to understand its main features. How does it help authors create their books beyond the usual word processing features?

Common Features Across Reading Platforms

Reading customization. All of the platforms, including the article-based apps, included some form of customization for text presentation. Common customizations were the ability to change the font, adjust the screen colors (sepia, white text on black screen), and switching between a one-column and two-column layout.

Progress bar. All of the book reading apps used some kind of progress slider at the bottom of the page to indicate current location and allow users to skip forward.

The concept of “pages.” All of the book apps had a concept of the screen being a page. Users swipe between “pages.” There was some mixing on page counts here—some platforms used the number of screens to calculate the number of “pages,” while others used page numbers from the print-oriented text.

Left: Comparison analysis of ebook platforms. Right: Comparison of long-form article platforms.


Passage Retrieval and Progress Usability Report

Introduction

The goal of this round of testing was to test updated designs for progress and wayfinding, as well as two new designs related to progress. The first new design was a bookshelf design that shows the reader’s progress within each book of their collection. The second new design was an interactive map for tracking progress in fantasy novels. The aim in all of these was to see if participants understood the progress tracking and wayfinding features, and their sentiments towards each feature—positive, negative, or neutral.

Method and Participants

We recruited five participants for our second round of usability testing. We showed them high-fidelity mockups on an iPad. We had them perform a series of tasks such as reading passages from Peter Pan and navigating using static mocks.

 

 

 

 

Major Findings

Gradient rails were ineffective. None of the users noticed the gradient rails until the feature was pointed out. Even then, they still didn’t notice the change in color until it was pointed out. Its use was still ambiguous to readers. The one who did correctly guess was neutral towards the feature.

Recommendation: Eliminate the gradient rails. Although it might be effective as a subtle indicator of progress, it would require a very precise implementation that’s not in our scope for the project.

 

Pictorial icons for passage retrieval had an unclear purpose. The pictorial icons for passage retrieval received mixed reactions. Readers didn’t know what it was for or thought it was informational, related to the content they had read.

   Fig 1 (left). Dog icon next to text referring to a dog.      Fig 2 (right). Forest icon next to text referring to a walk in the woods

Fig 1 (left). Dog icon next to text referring to a dog.

Fig 2 (right). Forest icon next to text referring to a walk in the woods

Recommendation: Authors should work with illustrators to incorporate images. The default icon set should be the donuts.

 

No passage retrieval with either set of icons. Responses were divided on the usefulness of both sets of icons: some readers didn’t notice the pictorial icon or the donut progress icon, or they noticed it and felt it was distracting (in one case it startled the reader). None of the readers used it to retrieve content. Four out of five did find the donut chart for in-chapter progress useful, but there was no consensus on how it should be presented (stationary or relative).

   Fig 3. “Donut” icon visualizing progress within the scrolling chapter

Fig 3. “Donut” icon visualizing progress within the scrolling chapter

   Fig 4. The library view, with bar length representing the book’s word count and transparency representing reader progress in the book.

Fig 4. The library view, with bar length representing the book’s word count and transparency representing reader progress in the book.

Bookshelf design was clear, but needs refinement. Readers understood that the width of the bars corresponded to length, and that the transparency corresponded to the user’s reading progress. However, a few edits need to be made. One participant expressed concern about how to visualize very large works.

Recommendation: Decide how long the “longest” book should be for the visualization. At how many words are we willing to accept some data loss with the visualization? Alternatively, consider representing the length of the book with height instead, which would allow more space (since it’s on the scrolling axis.)

 

 

 

 

Map was well received as an idea, but the interactions were unclear. The map was well received, and readers felt it useful for reading fantasy novel. Users understood that points on the map appeared as characters progressed across the landscape. However, interactions with the map and what information it provides were ambiguous. The current interaction was that tapping on a dot takes the reader to that chapter. Readers were consistently surprised when they tapped the dot and were taken to another location in the text.

We showed readers two options for the map: one where the map was a popup, and another where it was in-line with the text at the end of each chapter. There was also no consensus on how it should be presented.

Recommendation: Offer a popup with information about what chapter the reader is tapping on. Then offer them the option of tapping again to be taken to that chapter or offered some kind of summary.