Breaking the Wall of Text
Team: 1 | Duration: 12 weeks | Reading Time: 12 mins
I attended a talk by a doctoral researcher at the HCI International 2015 Conference, who presented her work on the emerging topic of contract visualization - an approach that aims at making legal contracts clearer and more user-friendly with the help of better typography, layout design and information visualization.
This topic sparked my UX Designer’s curiosity and provided inspiration for the research questions behind this study:
What changes to legal information design might convey complex information in a more logical and intuitive manner?
Can a standardized layout help demystify the legal wall of text into a clear, simple design that uses visual cues to support comprehension?
This was my capstone Project as a Masters candidate in the HCI program at Georgia Tech.
Copyright Problems in Online Communities
In the large-scale content analysis of public conversations about copyright in different online communities, one clear theme that emerged was the prevalence of problems related to copyright expressed by creators in these conversations (Fiesler, Feuston, & Bruckman, 2014, 2015). The five major types were:
Dealing with consequences
Fear of infringement
Dealing with infringement
I have used these categories as a framework to organize the copyright content.
Private Parts is an open-sourced, customizable toolkit by the mobile security company Lookout to help developers implement visual, user-friendly privacy policies.
This plug-and-play toolkit was the inspiration behind my responsive visualization framework for copyright policies.
A “Nutrition Label” for Privacy
Drawing from nutrition, warning, and energy labeling, as well as from the effort towards creating a standardized banking privacy notification, the process for constructing and refining a label tuned to privacy is presented (P. Kelley, J. Bresee, L. Cranor, and R. Reeder, 2009). Their user study methodology was particularly useful to me in understanding how a standard copyright policy compares with a restructured policy which employs visual solutions.
Cognitive Load Theory (CLT)
In order to understand written texts people form mental models (Johnson-Laird, 1983), mental representations based on the principles of causality, spatiality and temporality (Zwaan, 1995). In light of CLT, both information structure (how the content is ordered and organised) and information display (how it is visually presented) should play a key role in supporting comprehension and intellectual performance.
Structured content with a glance-able summary of the sections to promote immersion in the text
Visual cues (based on The Gestalt Principles of Visual Design) to assist readers in focussing on important items
Responsiveness across platforms
Design and Implementation
I began with restructuring the original content based on the areas of copyright problems in online communities identified previously in related research:
I consulted Dr. Casey Fiesler as well as Mr. Joshua S. Wattles, Advisor in Chief for all legal matters at DeviantArt and had them review both of my versions. Some of the high points of their feedback were:
✓ Logical division of content
✓ FAQ style layout
✓ Good use of visual cues for improved readability
× Beware of oversimplification of text, especially in explaining the license terms
× Hidden content increases liability on the company, so interaction should be intuitive
The application provides the copyright policy in the form of a single web page. Some early designs of my prototype are shown below:
Compared DeviantArt’s standard copyright policy with my restructured policy by employing the Rapid Iterative Testing and Evaluation (RITE) method. Using a within-subjects design where each participant completes 23 questions relating to information-finding tasks, perception of comprehension and usability for each prototype while following a think aloud protocol.
A 12-participant laboratory user study was conducted with all types of content creators/digesters - primarily in the student demographic. Their copyright knowledge and content creation details are shown here:
Task Structure and Analysis
The participants were given a randomized prototype to interact with, while completing an evaluation form. The usability section of evaluations was achieved by direct observations as well as interview-style questions on the interaction design. The comprehension can be gauged in two phases:
Quantitative assessments using binary Yes/No responses, along with a deeper assessment of the ternary ‘I Don’t Know’ response. The deeper assessment is tied to the ‘Phase I’ of qualitative assessments and hasn’t yet been explored for the purposes of this project.
Scope for Expansion
Content curation on collaboration with legal experts.
Introduce a social aspect with inline commentary and discussion boards.
Extend the functionality to other consumer industries like online marketplaces.