
<blockquote>Revamping Google Calendar through co-design with a graduate student with ADD to make scheduling your life more rewarding, fun and balanced.</blockquote>
<figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*ogocLJq3XlEXmv9-1gJA-w.jpeg"><figcaption>A person looking at a computer showing Google Calendar. Photo by <a href="https://unsplash.com/@gainingvisuals?utm_source=unsplash&amp;utm_medium=referral&amp;utm_content=creditCopyText">Gaining Visuals</a> on <a href="https://unsplash.com/s/photos/google-calendar?utm_source=unsplash&amp;utm_medium=referral&amp;utm_content=creditCopyText">Unsplash</a></figcaption></figure><p><em>Saransh Grover, Crystal Marte, Sashank Mullapudi</em></p>
<p><em>This was a participatory design project done for the INST704 Inclusive design course, as part of the HCIM Masters program at the University of Maryland College Park.</em></p>
<h3>Introduction</h3>
<p>For our inclusive design course, we, 3 Human-Computer Interaction Master’s students teamed up with a 23 year old male graduate student with ADD and sensory processing disorder on a participatory design project that sought to collaboratively identify, understand, and resolve the inherent barriers that exist within tools that are essential to one’s day-to-day.</p>
<p>Typically, when designing products targeted at Individuals with ADD, or the general disabled population, researchers and designers tend to approach this from the deficit perspective and create something to fill a “gap,” rather than understanding the individual’s strengths and designing for that instead [<a href="https://dsq-sds.org/article/view/5236/4475">1</a>]. This deficit perspective does more harm than good, and can ultimately affect a person’s self-esteem and lead to feelings of low self-worth and defeat. For example, when viewing the symptoms of ADHD inattentive type, the symptoms include “trouble paying attention to details, are easily distracted, often have trouble organizing or finishing tasks and often forget routine chores.” [<a href="https://my.clevelandclinic.org/health/diseases/15253-attention-deficit-disorder-without-hyperactivity-add-in-adults">2</a>]. However, when speaking with our participant, he was proactively taking steps to manage his workload (e.g., using task and time management tools), but<strong> the tools themselves</strong> were insufficient and impeded his ability to plan, assess, and execute his tasks. Additionally, there was little positive reinforcement after an event or task was completed, which further exacerbated negative feelings.</p>
<h3>Goals of the Project</h3>
<p>By engaging in a participatory design process, our goal was to mitigate the typical power dynamics that exist between researchers/designers &amp; participants by affording a sense of agency to the participant through co-design and collaboration. We aimed to collectively understand the context of our participant’s daily life to determine the utility of his everyday tools and ideate possible design solutions. We approached our research process using a social model of disability, which is critical in preventing perpetuation of harmful stereotypes and stigmas of people with ADHD and disabled people as a whole.</p>
<p>We conducted three sessions with our participants over the course of October — November 2022. The first two sessions were in-person and the final session was done virtually. All sessions were video recorded with the consent of the participant.</p>
<h3>Phase 1: Problem Understanding Interview</h3>
<p>The first session was an introduction and problem understanding interview between the participant and the team. to briefly introduce the participant to our team as well as the goals and structure of the project. In particular, we focused our questions on his experiences with digital tools and what strategies he’s implemented when a tool doesn’t work for him. We ensured our questions did not frame his disability from a “deficit-model” perspective by utilizing neutral language and understanding his strengths.</p>
<p>We began with an icebreaking activity called “Circles of Me” which is a method that visually captures an individual’s needs and desires in a hierarchical manner. Fashioned as a bullseye, participants include what they value the most in the innermost circle in descending order. By collectively sharing our values, it balanced the power dynamics and helped us understand our participants as well as fueled collaboration/co-design.</p>
<p>Our questions focused on tools that our participant found frustrating, and we poked further as he mentioned tools to focus on three key things: his feelings during the tools usage, why he thinks the tool is frustrating and changes he suggests to the tool. All our questions and follow-ups were framed around these. At the end of the interview, our participants sketched a calendar interface, including adding new features to improve from his current tool of choice, Google Calendar.</p>
<p>After the first session, we decided on working to address challenges faced by the participant on Google Calendar. We chose this because he constantly expressed Google Calendar to be the main source of frustration, and given his constant usage as well as expressed necessity of the tool, we felt this to be the most pressing problem.</p>
<h3>Key Findings</h3>
<p>Our participant consistently takes steps proactively to assess and manage his workload utilizing tools like Google Calendar, but the interaction can elicit feelings of being overwhelmed and lead to avoidance because of the extensive burden on him; ironically, the tools built for efficiency and streamlining are inherently inefficient and stress-inducing.</p>
<p>Simultaneously these negative feelings that arose during this interaction produced a negative self-outlook, where the participant consistently blamed themselves for the mishaps or inadequacies of the tool itself; for example, when utilizing an specific website for his work as a teacher’s assistant, he wrote an important announcement that was then deleted because the text was entered in the student mode, rather than instructor mode. Although the state of the system was <strong>not</strong> obvious, the participant blamed himself for this mistake.</p>
<h3>Phase 2: Co-Design Session</h3>
<p>Prior to the session we identified three key tasks based on our findings and the participants’ interest. The tasks are:</p>
<ol>
<li>
<strong>Prepare a schedule: </strong>This included responding to invitations, scheduling breaks, prioritizing events and tasks, and feeling prepared with his event scheduling.</li>
<li>
<strong>Balancing time: </strong>This task focused on reducing the feeling of being overwhelmed by making him take breaks during his day, and scheduling breaks to his interest and availability.</li>
<li>
<strong>Assess what’s upcoming: </strong>This task focused on checking the next events, as well as positive reinforcements and breaks on the successful completion of these events.</li>
</ol>
<p>Based on these tasks, we brainstormed a few potential solutions and sketched them through crazy-eight activities. Some initial solutions we had were:</p>
<ul>
<li>Reward system for calendar</li>
<li>“Fun Time” — app for collecting ideas for breaks and scheduling as appropriate.</li>
<li>Asking Mood after completion of task and adding positive reinforcement notifications</li>
<li>Visually declutter the google calendar interfaces by shading and using less blocks + color.</li>
</ul>
<p>We presented these three tasks as well as the initial solutions to our participant during the co-design session. In addition, we conducted a likes/dislikes activity where we showed him different tools for scheduling like <a href="https://flexibits.com/fantastical">Fantastical</a>, <a href="http://todoist.com/">Todoist</a>, and <a href="http://reclaim.ai/">Reclaim.ai</a> and asked him to mention features of each he liked and disliked.</p>
<p>Finally, we asked our participant to brainstorm solutions on paper. We printed out several interface components to build upon the first session’s sketches and facilitate the physical prototyping process. For example, we printed out multiple calendar interfaces, blocks of time, widgets of tasks and calendars, icons and layouts of different screen sizes.</p>
<figure><img alt="A paper sheet with a calendar interface drawn on it through sketches and components inluding graphs, color blocks, widgets and positive reinforcement messages" src="https://cdn-images-1.medium.com/max/1024/1*s2tU_oe_K6c7ojJ6Uh93wg.jpeg"><figcaption>Sketch created by our participant after the co-design session</figcaption></figure><p>After he created different solutions through these components, we asked him to rank the solutions based on importance. The top three ranked solutions from our participants were:</p>
<ol>
<li>Reward system</li>
<li>Scheduling breaks</li>
<li>Bulk actions &amp; tentative events</li>
</ol>
<h3>Prototype</h3>
<p>The final solution is very close to what our participant envisioned in the first participatory session- An upgraded version of Google Calendar that helps make his daily or weekly planning and execution seem less overwhelming, through elements such as</p>
<ol>
<li>Dark Theme in Calendar</li>
<li>“Plan Your Week” option to select interests that would be prompted at various times during the week</li>
<li>Micro-interactions using elements of interests (Weather and Space)</li>
<li>Tentative slots on Present Day that could potentially be used as Planned Breaks</li>
<li>Positive Affirmations through visual and verbal language</li>
</ol>
<p>We chose to make an augmentation to the existing Google Calendar over creating a new application, keeping in mind the familiarity for the participant.</p>
<p>As the participant was comfortable with technology, and familiar with Google and Outlook calendars, the proposed solution was received as a better version of existing ones with many of his requirements (wishes) met.</p>
<p>To convey the final solution best to our participant, we have broken it down to 3 scenarios which cover the most of his desired features</p>
<h3><strong>Scenario 1: Planning Your Week</strong></h3>
<p>At the beginning of every week, the participant is provided with an analysis and recommendations of what could be some additions to the weekly planner on Calendar. Machine Learning and Automation would help suggest most appropriate and apt utilizations of time keeping in mind the interests and timelines.</p>
<a href="https://medium.com/media/1f39681dc3faf80054fa120c1df36b50/href">https://medium.com/media/1f39681dc3faf80054fa120c1df36b50/href</a><h3>Scenario 2: Staring Your Day</h3>
<p>As our participant is used to checking the calendar in the morning, we redesigned the morning experience to provide positive affirmation through micro interactions, such as 1) clouds washing away the colors from previous day to signify the completion, 2) Statistics denoting which aspects the participant performed at in the last week, and 3) Notifications assuring a tone of encouragement.</p>
<a href="https://medium.com/media/e7339b9c967da9fc0370847c2ebf2363/href">https://medium.com/media/e7339b9c967da9fc0370847c2ebf2363/href</a><h3><strong>Scenario 3: Using tentative slots for Breaks</strong></h3>
<p>On completion of an event, a potential break period is denoted as a tentative slot, which would bring back recommendations set by the participant at the beginning of week. Elements such as confetti are used to reinforce the importance of a healthy work-break time balance</p>
<a href="https://medium.com/media/baf5ab4d59e5fcbf8f0183327c05d8bd/href">https://medium.com/media/baf5ab4d59e5fcbf8f0183327c05d8bd/href</a><p>The aspects we propose in this redesign could help better the Calendar experience for a large group of people, especially the young adults (students and employers) who prefer to be organized and aware of their time utilization.</p>
<p>The basic elements such as Dark Theme, micro-interactions are currently available in the market in the form of extensions and plugins. However, our market research has shown that not many are holistic in terms of what our participant would need in an ideal calendar.</p>
<h3>Prototype Feedback</h3>
<p>During the final session, we presented our prototypes to the participant to gather feedback and any opinions he had. Overall, he really enjoyed the three scenarios we redesigned for but also provided some constructive feedback.</p>
<ul>
<li>He provided positive feedback on the approaches to reduce information overload visually by fading away colors, the dark theme, and adding micro-interactions related to his interests. However, he actually did not prefer Calendar blocks without colors filled in — which was surprising to us given it made the calendar much less visually overwhelming. However, in hindsight, our participant really enjoyed using color as a mechanism to categorize, and he mentioned that he wanted the ability to “laterally view colors.”</li>
<li>With the “Plan Your Week” flow, he found the most benefit in the bulk actions, the prioritizing and the breaks. In particular, he mentioned that he would use the flow for more than planning your week — for example being able to go in the future and block out holidays or planning further in advance of the week through bulk actions. We found that he enjoyed <strong>as much control as possible</strong> when planning, and while the burden was on him to plan, he felt that the visual appeals of the plan your week flow, as well as the simplicity of a single step-by-step structure made the process less overwhelming and more likely for him to proactively complete.</li>
</ul>
<h3>Future Work</h3>
<p>There were some key considerations we did not have time to explore during this project that were discussed with the participant.</p>
<p>First, our participant mentioned how he would like the calendar to self-track his tasks based on multiple different modalities like location, screen usage, and other self-tracking abilities. This would significantly reduce the burden on the participant, but would require some level of machine learning to understand what the participant is doing at a given time based on these different modalities and the history.</p>
<p>Secondly, the system used to suggest breaks must be extremely good. It should not only suggest breaks that the participant is interested in, but also at times when the participant is available to do them. This would also require a complex algorithm with some AI and prediction techniques built in.</p>
<img src="https://medium.com/_/stat?event=post.clientViewed&amp;referrerSource=full_rss&amp;postId=84d34c611528" width="1" height="1" alt="">


Shifting Burden From the User for Scheduling through Online Calendars


<figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*S-oJXNjccMfo-o1_epkeFg.jpeg"><figcaption>Photo by <a href="https://unsplash.com/@romankraft">Roman Kraft</a> on <a href="https://unsplash.com/photos/_Zua2hyvTBk">Unsplash</a>.</figcaption></figure><h4>Untangling a new societal phenomenon using visualization.</h4>
<p>By <a href="http://saranshgrover.com/">Saransh Grover</a> and <a href="https://agnesjoseph.me/">Agnes Joseph</a></p>
<p>The term “fake news”, which first gained popularity during the 2016 U.S. Elections, has now become a part of our daily language. However, “Fake News” does not properly represent the various types of false, inaccurate, or misleading information on the internet. Instead, it would be better to consider the term misinformation, which is any piece of inaccurate information that is shared without meaning to cause harm.</p>
<p>Though misinformation may be shared without any bad intentions, its spread still has harmful effects on society. For example, during the pandemic, the multitude of misinformation on social media platforms resulted in several dangerous misconceptions among the general public. These Covid “myths” were left to propagate unchecked and accepted as factual knowledge by several people on the internet. The results of these myths had serious consequences ranging from dangerous cures to racism against the Chinese population. The widespread panic, fear, and disorder resulting from these rumors led WHO to declare an infodemic, that is, the spread of too much information including false or misleading information in digital and physical environments during a disease outbreak.</p>
<p>Now you may say, great, misinformation is a problem but there is no way I would fall for it.</p>
<p>Well, studies have shown that more than 80,000 occurrences of misinformation on Twitter that were retweeted millions of times over a wide array of topics with just a sample set of 100,000 data points.</p>
<p>Our project specifically wanted to focus on COVID-19 misinformation, as not only is it a topic of global concern, but also it has a clear scientific divide over what is misinformation and what is true.</p>
<p>When creating our tool, our goals with the visualization project were to create a medium to let users better understand how misinformation is spread on Twitter. We felt like the Twitter timeline user interface did not necessarily provide a good visualization of identifying and proving how misinformation may spread over time. Our project had two important questions:</p>
<ol>
<li>Can we show tweets over time classified by fake and real in an interactive way while still maintaining other engagement metrics (likes &amp; retweets)</li>
<li>How can we better visualize a tweet and its subtweets such that we can see the sentiment linked with it?</li>
</ol>
<p>As is well known in research, and data visualization, you must stand on the shoulders of giants. Therefore, we were lucky enough to find <a href="https://github.com/cuilimeng/CoAID">CoAID</a>, a dataset of over 120,000 tweets that were pre-classified into real and fake news, as well as real and fake claims. Using this data, we decided to begin working on the biggest challenge — <strong>how do we visualize it?</strong></p>
<h3>Sketching</h3>
<p>Since we knew that our tool should be interactive and since our dataset was so large, we initially began by exploring how we could categorize it. We decided on choosing hashtags that are provided by Twitter to be the most natural form of categorizing tweets. Hence, we designed our first page with the goal of selecting a hashtag . This page included a set of pre-determined hashtags as well as a search bar (Figure 1a).</p>
<p>Now, we decided to tackle the challenge of how to visualize a Twitter timeline. Since part of our goals were to see the engagement of fake news over time, we decided to plot our data according to when the tweet was published. However, the struggle we faced was to plot thousands of individual data points (tweets) over time without clustering and confusing the user. This is why we decided to take a “drill down” approach. We created a histogram to show counts of tweets over time. Each “bin” in our histogram represents a month, and each bar for that month is the number of fake and real tweets — color coded, green for true claims and red for false. This color scheme is consistent throughout the dashboard.</p>
<p>Once a user interacted with the histogram, they were taken to a bubble chart which now only showed the tweets for the particular month they clicked on . This leaves us with much lesser data that can be visualized without overwhelming the user. Here, users could also see information about the tweet on hover like the content, the likes and retweets. Finally, clicking here would take the user to a representation of the individual tweet and a tree representing the retweets and replies of that particular tweet.The root of the tree is the selected quote tweet, and it is displayed with its contents on the screen, as well as a classification between true and false. The children of this root node are the top 5 quote tweets for the given tweet. Each child node is initially represented by cards that contain the profile picture, the engagement metrics, as well as a color coded sentiment analysis category. At any given time, the user is able to see 2 levels deep into the tree. Upon clicking on another node, the node is expanded into a full card view, thereby collapsing the remaining nodes.</p>
<figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/0*NlCVkYthA0IYxGD6"></figure><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/0*egZquttHVHhG81sb"></figure><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/0*vLCK7v4TNdZlNmks"><figcaption>Figure 1 a) Sketches for Landing Page, Hashtag dropdown box, b) yearly visualization of the number of tweets (as bar graph and line graph), and the branching tree diagram for the quote retweets, c) Sketches of the Tree system</figcaption></figure><p>Having hierarchy in our visualizations such that the data is drilled down from yearly to monthly to singular views allows the structure of each visualization to be more digestible and comprehensible. In addition to this, we employ consistent use of color throughout all parts of our dashboard, in order for users to have higher recognition of claims. Through the hierarchy, usage of color and sparing use of visual channels we are able to reduce cognitive load on users and help them make sense of the visualizations.</p>
<p>A great strength of the tree visualization is that it gives a great representation of the branching effect of tweets, which accurately conveys how information is perceived and then shared on social media, replicating the “word of mouth” effect in the spread of information online.</p>
<figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/0*WlJSKWpzkkz2_I-c"></figure><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/0*GFozY_j68ujTXtcy"></figure><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/0*JwWMoLQsAtcef2uG"></figure><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/0*DCfpCBxq28zblccg"><figcaption>Final prototypes of our application workflow</figcaption></figure><h3>Presenting — <a href="https://misinformation-roots.netlify.app/">The Roots of Misinformation</a>
</h3>
<p>Our application provides a way for users to interact with over 120,000 tweets categorized by the relevant hashtags.</p>
<figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*xjZa5P8HQ9fzRAanLswIaA.png"><figcaption>Home Page — Allows users to filter and search by hashtags</figcaption></figure><p>While our design sketches included a Histogram as well as a Bubble Chart, during development we found that separating this into 2 stages did not make much sense. Instead, we decided to combine the two into one visualization called the Swarm Plot. This visualization conveys the purpose of both previous visualizations into one, and we used filtering techniques to ensure that the data is not too congested.</p>
<figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*lfDwlZhoBFPtszjvBeusKQ.png"><figcaption>Swarm Plot for the hashtag #Coronavirus</figcaption></figure><h3>So, What did we find out?</h3>
<p>Well, displaying misinformation is a HARD problem. While our applications takes a stab at a potential way to do so, it is FAR from perfect. Here were some challenges we faced</p>
<ul>
<li>Even if something is classified as misinformation, <strong>understanding why it has high engagement is extremely difficult</strong>, and largely up to Machine Learning algorithms on social media.</li>
<li><strong>There are WAY more true posts than fake posts, and they tend to dilute the outliers of fake news visually.</strong></li>
<li>Hashtags are not a great way to categorize tweets. This is because <strong>only 16.7% of tweets in our dataset even used hashtags</strong>! If we had more time, we believe that using NLP to categorize data, or even keyword searching to better categorize tweets by relevance.</li>
</ul>
<figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*-ptVCZUAaDCzIdyP5jAl3w.png"></figure><p>While we faced challenges, we also found some key takeaways from our visualizations:</p>
<h3>Fake News Centered Around Early Pandemic Months</h3>
<p>We found that the most engaged tweets that were classified as fake news occurred during the month of March 2020 and subsequently went down over time. While this may be due to data collection, or many other reasons, we believe that it also shows how vulnerable fake news can be during the onset of events. For example, when a catastrophic event occurs, the first couple days are when there is the least amount of credible information available — leading to speculation and thereby misinformation spreading.</p>
<figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/0*WeB7Su_D7RaFDAey"><figcaption>Top Fake Tweets from the CoAID dataset</figcaption></figure><h3><strong>#Coronavirus: More “True” but Higher Engagement</strong></h3>
<p>Looking at a specific hashtag — <a href="https://misinformation-roots.netlify.app/hashtag/coronavirus">#Coronavirus</a>, we can see the trend of overall having more tweets that are classified as real, but fake ones have much higher engagement and are outliers. While this is not true with all hashtags, it can certainly show a trend in many, and falls along with <a href="https://www.science.org/doi/10.1126/science.aap9559">other research</a> done that suggests that fake news is much quicker to get 200 retweets than real news.</p>
<figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/0*CbWKMEGZFEeB_S6d"><figcaption>Swarmplot of tweets with the #Coronavirus</figcaption></figure><h3>So, what next?</h3>
<p>Our goal with this project was to visualize misinformation on social media, and while we made a strong attempt towards it, we recognize that it is a complicated, yet important issue. We hope that our work sparks conversations and interest amongst other data visualization experts to explore avenues towards visualizing misinformation. <strong>We’d love to hear your thoughts, how do you think we can visualize misinformation?</strong></p>
<img src="https://medium.com/_/stat?event=post.clientViewed&amp;referrerSource=full_rss&amp;postId=f8025bd959cf" width="1" height="1" alt=""><hr>
<p><a href="https://medium.com/visumd/look-thats-fake-news-f8025bd959cf">Look! That’s Fake News!</a> was originally published in <a href="https://medium.com/visumd">VisUMD</a> on Medium, where people are continuing the conversation by highlighting and responding to this story.</p>


Look! That’s Fake News!


<h4>Best practices for text in geographic visualizations.</h4>
<figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*jkWmlpRuedX5xOMOWgqnug.jpeg"><figcaption>Photo by <a href="https://unsplash.com/@geojango_maps?utm_source=unsplash&amp;utm_medium=referral&amp;utm_content=creditCopyText">GeoJango Maps</a> on <a href="https://unsplash.com/s/photos/geography?utm_source=unsplash&amp;utm_medium=referral&amp;utm_content=creditCopyText">Unsplash</a></figcaption></figure><p>Whenever we create geographic visualizations, text plays a significant role in creating a narrative over the visualization. In a recent paper, <a href="https://onlinelibrary.wiley.com/doi/epdf/10.1111/cgf.14309">Latif et al</a> studied best practices of implementing text with geographic visualizations focusing on how they interact with space and time.</p>
<p>Prior work has proven that a good interplay between text and visualizations plays an important role in keeping readers engaged. In order to find current and best practices to understand this, the researchers performed analysis on 22 news stories from 10 well known digital-journalistic sources like the New York Times and BBC. They created 2 collections — one a diverse set of news and the other specifically related to the COVID-19 pandemic. Below are the top eight best practices to keep in mind.</p>
<h3>1. Represent Location &amp; Time with Text</h3>
<p>First, be sure to add location (eg Boston, New York, etc) and time (Oct 2019, 1970–1972, etc) to geographic visualizations using text. Most commonly this was done within the visualizations itself at scale, but also in a textual narrative.</p>
<h3>2. Use Text to Add Context</h3>
<p>All stories included a textual background to introduce the reader to the concepts and contextual cues that will be used in the visualization. This is especially important when the visualizations and topics are of a complex manner — such as organ donations and seismic waves. Through this, we can provide a sense of domain knowledge to the reader letting them better interpret the visualization.</p>
<h3>3. Include a Descriptive Caption</h3>
<p>86% of visualizations had descriptive captions which better help explain the findings. Generally — captions must communicate the main takeaway or insight from the visualization. This makes it much easier for creators and researchers to analyze and interpret the insights than it is for general users.</p>
<figure><img alt="" src="https://cdn-images-1.medium.com/max/952/0*7WsrD2RAzuSWGJaV.jpg"><figcaption>Caption from FiveThirtyEight stating that “Many rural counties and doing OK”</figcaption></figure><h3>4. Embed Visualizations into the Narrative</h3>
<p>Successful stories embed visualizations throughout the narration of the story rather than putting them all in a single repository either at the beginning or end. This is important because each visualization must have its own set of contexts, captions, background, and all other textual features described here.</p>
<h3>5. Use Cultural &amp; Regionally Relevant Language</h3>
<p>Using language that is regionally recognized to the geographic area that is being referred can help minimize confusion and create a broader sense of abstraction that is also understood by the reader.</p>
<h3>6. Provide Sufficient Background</h3>
<p>In tandem with #4, providing background to increase domain knowledge and dataset descriptions is helpful to add context. It is important though to make sure that the story does not excessively provide background as that may sway readers away — instead, focus on what is essential to understand the visualization.</p>
<figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/0*r0DiqH0fWx0gO0jq"><figcaption>This visualization from the New York Times provides an overview of the story, not focusing on details but rather broader conclusions.</figcaption></figure><h3>7. Overview First</h3>
<p>The visual information seeking mantra originally formulated by <a href="http://www.cs.umd.edu/~ben/">Ben Shneiderman</a> — “<em>overview first, zoom and filter, then details-on-demand</em>”— is also relevant for text. When adding text on visualizations, make sure to provide an overview first and then add more details as the user zooms and filters. This is also relevant to the structure of storytelling. The first visualization in a story should always be an overview visualization, followed by a detail and then a comparison.</p>
<h3>8. Position Matters</h3>
<p>Successful stories pack visualizations and texts closely together. The visualization put next to the text helps readers better understand the descriptions. They also avoid large amounts of scrolling. A good interaction between placement of text and visualizations is imperative to a good user experience.</p>
<p>Use this list as a way to perform heuristic evaluations of visualizations that you create — and ensure that they are followed throughout the story as efficiently as possible, but note that these are heuristics, and actual needs may wary based on contextual cues and target audiences.</p>
<img src="https://medium.com/_/stat?event=post.clientViewed&amp;referrerSource=full_rss&amp;postId=3d369d383d32" width="1" height="1" alt=""><hr>
<p><a href="https://medium.com/visumd/text-on-maps-3d369d383d32">Text on Maps</a> was originally published in <a href="https://medium.com/visumd">VisUMD</a> on Medium, where people are continuing the conversation by highlighting and responding to this story.</p>


Shifting Burden From the User for Scheduling through Online Calendars

Look! That’s Fake News!

Text on Maps

New Site!

Designed & Developed by Saransh Grover © 2024