Introduction
So, let’s start with a reflection of a reading by Dimara and Perin (2020) called “What is Interaction for Data Visualization?”. First of all, Dimara and Perin (2020) take on the task of defining this ambiguous concept called “Interaction”, using inputs from the Visualization community and the Human-computer interaction.
Summarizing the Main Points
Now there were some key points that stood out to me within this reading about interaction for visualisation, such as the key components of interaction. There are key components within Interaction for Data Visualisation such as external and internal entities, external and internal actions.
What interaction must have in terms of external entities is a user and data, with the user being a person and data being an intangible information. The two can interact with each other using what Dimara and Perin (2020) call a “visualisation system”, which is computational tools basically.
External Entities
External entities also include User interaction with physical/tangible devices such as a mouse and a keyboard. Not only does it stop there, but interaction also takes into consideration human expressions, and these small things hold the key to understanding user intent and enhances the visual experience.
Internal Entities
Internal Entities: Interaction for Data Visualisation occurs only with a goal in mind (it must have an intention). According to Dimara and Perin (2020) The intention can be “data exploration, knowledge creation or social interaction”. However, whats important is that these interactions have intentions that align with a users’ needs and objective. With Internal entities, it’s also more about the mentality, skills, and abilities of users as they interact with a visualization system as it can help bring new insights into light.
External Actions
External Actions: Interaction in context of data visualisation must involve a dialogue between the user and the visualisation system, which means that the user must initiate an action and the system responds with a reaction that is visible enough so the users can gain insight from the data through the interaction. User interactions in data visualization is not just about users being able to explore the data, you can reshape it data, change how its presented and even create new knowledge from it.
internal Actions
Internal Actions: Interaction with a visualisation system is not only just about the physical interactions! It doesn’t require users to not only click buttons or manipulate data, but the user also has to engage in cognitive acts, basically a combination of both physical and mental actions. What Dimara and Perin (2020) meant is that most of the interaction unfolds within the user’s mind.
Three Approaches
According to Dimara and Perin (2020), Interaction in Visualization can be categorized into three approaches, which are “System-centric, task-centric, and human-centric approaches”. And these categories help in making us understand how interactions are designed and used basically in visualization systems, and how they provide a very exceptional user experience.
The system-centric approach focuses on how the visualization system works and not on the user or how the user interacts with the system.
Task-centric approach sees interaction as low-level tasks that relate to data types, order, user roles. These low-level interactions build the high-level visualization interactions. Example of these tasks can include selecting data points, filtering data, sorting, zooming, or even changing data visualization types.
Meanwhile The Human-centric approach only on focuses on the user’s intentions. It's about what we want and need from the data.
My Reflection on the Topic
So far, the reading by Dimara and Perin (2020) not only describes Interaction in data visualization, but it also emphasizes on its nature, it highlights the fact that with data visualization and interaction, its more than just clicking and looking at visuals for users, It’s beyond the technical aspects. Its more than that, it’s about the mental action, it’s about understanding the users’ intent, context and the interplay between human cognition and technology.
This reading provides a good understanding of the components of Interaction in Data Visualization, however it does not really go much deeper into these components. For example, like with the system-centric, user-centric, and human-centric approaches, it doesn’t really explore them, it just touches on the bases onto what they are and what they do, in terms of delving much deeper, it doesn’t really do that! However, it makes you understand what these centric approaches are so it’s good.
When reading this reading, one thing I did not like about it was how the authors started by explaining how their research took place. It was really pointless and really threw me off at first, because I did not know what they were talking about most of the time. The authors also do not get straight to the point, I feel like I would have understood the reading better if the authors had started off with the definition, then describe it afterwards with its components. They laid the definition in the middle of the document and by the time I reached the definition, I had already made up my own definition within my mind.
Or they shouldn’t have included a definition at all but let us construct our own definition from the reading since basically they described what Interaction for Visualisation is. I also sort of do not agree with their definition of Interaction for Visualization, I feel like it can be more than that. The definition they provided is more ambiguous and complicated. It felt a bit vague and complicated.
One thing that confused me was how the authors kept on saying, Interaction for Visualisation and Interaction for Data visualization, it may be similar but its not the same according to me, the word “Data” adds a lot of emphasis on the entire term.
DEFINING INTERACTION FOR VISUALISATION
The definition below combines insights from both the visualization community and Human-computer interaction literature.
According to Dimara and Perin (2020) “Interaction for visualization is the interplay between a person and a data interface involving a data-related intent, at least one action from the person and an interface reaction that is perceived as such”.
In the end, this reading defines the definition of interaction in visualization, and with that, the definition includes the following components: Interplay, Person, Data Interface, action, action-reaction, reaction perceived as such, and data-related intent. These components build this definition.
Interplay, basically refers to an interaction between a person and a visualisation system, like for example interplay involves the back-and-forth communication between a user and the system as they explore the data.
Instead of using the term user, Person is used in order to be gender-neutral, and person could refer basically to anything, like even a robot that is interacting with data visualization system can be considered a person. This was actually well thought in terms of being inclusive to all species basically. Allowing even non-human entities to participate in interaction, it shows how the definition accommodates even diverse users.
Data Interface mediates the interaction between a person and the data. For example, it could be a touchscreen display showing an interactive data visualization, and the person can interact with it to manipulate the data.
Action is basically action, any activity done by a person, like how a person can be clicking on a data point to manipulate the data on a chart, or changing how a data visualization is presented.
Action-reaction is like when a person takes an action, and the data-interface responds basically.
Reaction perceived as such was complicated to understand at first however with more research, I understood that when a person interacts with data interface, it must respond in a clear way that the person will be able to understand.
Data-related intent is basically the goal of the interaction, the intention.
CONCLUSION
However, what I think is that this framework that defines interaction using Human-computer interaction and visualization is a huge contribution, particularly on the parts that its inclusive and adaptable to diverse modalities, meaning that it can accommodate interaction through all human senses, and it does not only accommodate a single modality. In short, it can accommodate diverse users and different contexts. I feel like however, as the field of interactions will grow, this framework will be there to provide a solid foundation for understanding and designing interactions that can accommodate a wide range of users and technologies.