Paper Reading #3: Augmenting the Scope of Interactions with Implicit and Explicit Graphical Structures

Introduction:
In this blog post, I will discuss a CHI scholarly journal looking to improve group based graphical editing. “Augmenting the Scope of Interactions with Implicit and Explicit Graphical Structures,” is a scholarly research paper published by Raphael Hoarau and Stephane Conversy at the University of Toulouse in Toulouse, France. Raphael is a PhD student, studying Human-Computer Interaction at the Laboratory for Interactive Computing at the University of Toulouse. Stephane is an associate professor at the Laboratory for Interactive Computing at the University of Toulouse, focusing his research on Human-Computer Interaction. In this paper Raphael and Stephane discuss a software program, ManySpector, they created to improve group based graphical editing and a research study to test user interactions and acceptability of their software program. In this blog, I will discuss the overview of Raphael and Stephane’s research paper, related work, how they evaluated ManySpector, and my opinion regarding their research, findings, and presentation.

Summary:
Before the research team could test human perception of group based graphical editing, they first had to develop a software program capable of manipulating grouped objects the way that they would like to. The idea behind their software program is to allow users to group objects multiple times, display attributes of group objects, and then edit those attributes all at once. For example, if I have five shapes on the screen grouped together and three text fields grouped together, and I want to change the color of all of the shapes to green, I would simply have to select the group, which will display the group’s attributes and values, and then change the color attribute to green. 

There are already systems that allow group editing to be done, but Raphael and Stephane believe that these systems do not provide enough editing capability. For example, word allows you to group objects together, but the only options it then provides for editing is ungroup, scale, and rotate, which adjust the group as if it were one object. Raphael and Stephane’s point is some users may want to group a set of objects and then rotate each individual object to a designated, uniform angle independently. Manipulations such as these cannot be done in any graphical editor currently on the market.

Evaluation:
In order to evaluate the usability of their software program, Raphael and Stephane conducted a research study consisting of a tutorial and two exercises without help. The tutorial lasted fifteen minutes and consisted of basic grouping and editing skills. Then, they asked the participants to create a certain scene or object. They also asked the participants to think out loud so that the study administrators could follow their train of thought of what they are attempting to do at any given moment. At the end of the study, Raphael and Stephane asked the study participants to fill out a subjective qualitative questionnaire, that utilized the Likert scale in order to rate their interactions with the software program. The majority of the participants stated that they would need more training to be comfortable using the tool and that they would not feel proficient in such little time.

Discussion:
I believe that group based graphical editing is an area that is prime for improvements, but Raphael and Stephane’s ManySpector is not the final product to do it. There needs to be a way of making the grouping relationships more intuitive and requiring less training in order to grasp the concepts. However, when it comes to effort involved in creating such a software program, I am very impressed. Rahael and Stephane created a rather robust application for testing and proof of concept. I am definitely interested in trying out a piece of software like this.