In the last year, Graphic Hunters has moved the S-H-O-W event about data visualisation to the virtual stage. The overall topic of the latest event was ENCORE. ENCORE was a hybrid event and celebrated the fact that we need more; more inspiration, more data visualisation, and more great speakers.
But ENCORE was also about what’s next in the field of data visualisation. What are topics that are of importance? What can we expect in the years to come?
Students from the Zurich University of the Arts joined the event and made a summary of all the talks. Each summary concludes with one takeaway.
1. Storytelling through data
Sabine Devins is a Berlin-based digital content creator, journalist, and researcher who specialises in storytelling with data. In her talk she passionately emphasised the importance of context in combination with datasets. By showing the audience graphical examples of the gold price, voting rights for women, and password security and also by asking two important questions – why? and how? – Sabine demonstrated the power of context.
- Why did the price of gold increase and how did it affect the markets?
- Why was there a feminist movement at that time and how did it impact society?
- Why does the average password have eight characters and how do people come up with those exact passwords?
Looking at the patterns behind the datasets could give us a better and more holistic understanding of a situation. The story shows us the cause; the graph shows us the effect. Both are needed to get the bigger picture and convey information in a more tangible way. Giving context to a diagram or graph can help readers to not only understand it better, but also to better discuss the topic. Sabine gave simple but powerful instructions on how to deepen the understanding of data and help others understand it better too.
ENCORE: Unlock the stories in your next dataset by asking why and how!
2. Using movie-industry tools for data visualisation
Fernando Cucchietti leads the Data Visualization and Analytics Group at the Barcelona Supercomputing Center (BCS). His work focuses on scientific data visualisation and data science applied to industrial problems. The main aim of his work is to generate projects which maximise the impact on the viewer and increase memorability. To achieve this goal, he employs tools and methods widely used in the film industry.
At ENCORE, Fernando talked about the importance of images for scientific data visualisation. Complex topics and large amounts of data can be made more accessible with visual support which help increase the efficiency of communication. Since scientists are often not able to generate these visually appealing images, and designers have a hard time dealing with scientific data, Fernando suggests that scientists and designers work together to achieve an optimal result.
A problem in the collaboration between scientists and artists arises from the different tools used, which can be highly specific on the research side. To connect those tools and the data used for the visualisations, Fernando became an expert in creating working pipelines and using them.
While the gathered data is the baseline for the work he creates, an exploration phase is necessary to understand what can be interesting for the narrative. The typical pipeline for the data-driven animations looks like the following:
- Pre-production, which includes data casting
- Data analysis and conversion
- Production, consisting of modelling, animation, and rendering
ENCORE: Scientists and designers should work together and borrow from other industries to achieve greater connection with their audiences
3. Behind the scenes of the the official German COVID-19 vaccination dashboard
As an independent designer, Moritz Stefaner is constantly searching for the best way to visualise information. He focuses on the question of how one can find new and different ways to visualise and communicate data.
In collaboration with the German Ministry of Health and the Robert Koch Institute, Moritz and his team were responsible for the creation of the official German COVID vaccination dashboard, called Impfdashboard.
During his talk Moritz gave us a more in-depth overview of the website: how it was conceived and designed, what tricks and principles the team used, and what unique features they incorporated. He also mentioned the difficulties of working with constantly changing data, which necessitated frequently updating the dataset with data from the past.
An eye-catching feature on the website is the “Vaccination Clock”, which illustrates how many people are getting vaccinated per second in real time. This familiar framework gives the reader the feeling of how fast vaccination dose administration progresses. But the greatest thing about the clock is that it makes it easy for people to remember the statistics because a clock is associated with an everyday experience and is, for us humans, a natural scale.
The overall goal was to present the important vaccination statistics in a concise, appealing, widely accessible, and trustworthy way, which Moritz and his team definitely achieved.
ENCORE: Think beyond the chart! Find a familiar framework for your narrative.
4. Data visualization as an art of the future
Information designer Julie Brunet gave an insight on how designers can find their place in the quickly changing field of data visualisation. Her frustration with her former job pushed her to make an Instagram account to independently work on more and better visualisations. Early on, she was confronted with the question: what’s next?
From there, Julie dug into data visualisation’s past to see what withstood the test of time and to understand where the field is heading.
Against the backdrop of rapidly shifting technological possibilities, finding the balance between tool proficiency and holding on to only one tool that might grow redundant is a big challenge these days. Julie suggested collaborating with other designers to round out and expand the knowledge base. Her biggest tip was to figure out what you want to do and where you want to go; only then can you choose the tool most suitable for your endeavours.
Knowing what part you want to play can be one of the hardest things to decide. Just like the choice of tools, your choice of profession and identification can feel stifling. Julie suggests creating your own title to communicate what you do rather than what you are. Doing so, she said, also helps with imposter syndrome, which she faced as she embarked on data journalism.
When considering the future of data visualisation, Julie recommended looking to the pioneers, rather than the whistleblowers – folks who do not actually work in the field. There is no better way to build a future in data visualisation than by exploring its past while immersing in its present.
ENCORE: Create your own title to communicate what you do rather than what you are.
5. Voices of tomorrow — a new direction for augmented analytics
Matthew Falla is an interaction designer and co-founder of Parallel, an innovation consultancy helping product leaders develop original propositions based around simulations and synthetic environments.
Matthew presented the ongoing research project “Voices of Tomorrow.” He used the talk as Experimental Marketing to introduce the project and the idea and to collect opinions about it from the audience.
The project creates a balance between technology and humanity through algorithmic systems. In today‘s technology, systems are often cold, without a soul and without history. Matthew and his team are trying to break this perception by combining connected sensors and computer vision. They advise companies on the development of products, aiming to put the “real life” at the centre of decision-making and show the impact during a process on future people.
Questions like, “How can we bridge the gap between the physical and the digital? How can new technologies be developed without losing their soul? Can the synthetic solve the problem of the real?” are critical to Matthew’s team on these types of projects.
ENCORE: People’s lives and future lie behind a data point. Move from statistics to stories.
6. The art of transit data
Craig Taylor is a senior DataViz Design Manager at Ito World. He specializes in analysis and animation of geographic data. In his work, Craig focuses on narrative-driven, 3D data visualisation. His interest lies in manipulating data in creative and unconventional ways. The data visualisations he creates can be insight-driven, functional, or expressive of artistic freedom.
Taylor worked on the Bus Open Data Service (BODS) project for the Department for Transport and KPMG. It is an open-source platform that provides national bus timetables, fares, and live vehicle locations.
He works with a combination of internal Ito World tools: Houdini and Cinema 4D. To animate coverage over time, Craig translated the buses into glass marbles and played around with lights tails to give them directions. He is fascinated to discover unique patterns that appear through his work.
In another series called Occupancy Patterns he describes the number of people using buses. He plays with symbolism for each bus pattern on a network. Density or capacity levels are shown with sizes, colours, speed, spatial distribution and so on.
ENCORE: Find creative and unconventional ways of presenting your data (you might also delight yourself).
7. Thinking with visualisations, fast and slow
Steven Franconeri is a professor of psychology at Northwestern University and Director of the Cognitive Science Program. He is primarily involved in approaches to visual thinking, visual communication, and the psychology of data visualisation.
In the beginning of his talk, Steven addressed the human behaviour of recognizing and associating different representations. He made clear that certain visualisations can be recognized faster than others. This phenomenon is due to the fact that our brain is designed for the natural world, in which we constantly encounter objects, faces and scenes. In infographics, however, the focus is more often on the proportions between individual objects. This cognitive process is much slower than just recognizing the objects. Steven demonstrated his point with three mental tools that resulted in surprisingly mental processing differences.
Steven then offered examples of the importance of the right form of presentation to shorten thinking time. One such example was the bar chart, where the goal is to determine which of the two bars is the higher. If we as humans have to compare each pair of bars individually, the thinking time gets slower. If we however choose a different visual display and show for example the delta values of the two bars, we can compare much quicker. So every comparison of shapes we have to make has a reading direction and this can be determined by the designer when creating the visualisation.
ENCORE: Your presentation choices can encourage faster recognition.
8. Dataviz as a press freedom tool in Singapore
Rebecca Pazos is an Australian journalist working for the Singaporean The Straits Times news agency. She is the leading journalist for the digital graphics team and the newsroom advocate for data storytelling.
Rebecca started her talk with an example of her work on gerrymandering. In 2020, she and her team created a powerful visualisation of the constituency changes in Singapore using data from 1968. They showed how politics influenced those shifts in territorial units to give one political party an unfair advantage in elections. An interesting observation she noted was the feel some data can have compared to the raw numbers. Due to Singapore’s size, it felt that there were more changes happening than there actually were caused by the phenomenon of “splitting towns.”
Additionally, their visualisation revealed some unexpected places where gerrymandering took place. Readers also had the opportunity of putting their address into a search engine and receive information about how many times their constituency has changed. This gave a playful character to the visualisation and increased engagement with the reader.
Rebecca has the good fortune of witnessing the impact of her work and she realises the power visualisations wield. She stressed the importance and responsibility to get the visualisation right and to be certain about what she was presenting.
ENCORE: Be mindful of the power and responsibility you have as a data practitioner.
9. Getting lost in the world of data
Marcin Ignac is a data artist and computational designer. He talked about the opposite of the dataviz and the generative design approach. He shared that he often felt pressured in one direction or the other -evaluative versus generative- depending on the project. Dataviz is a readable form. Evaluative data is classically used; readers can see statistics and charts. Generative data is an abstract representation. It is a format that tries to describe events or statements,tangibly, experientially.
In the combination Marcin sees data art. He illustrated an interesting example by visualising data from Dropbox. He developed a block-like overview of the different data types using colour to create a perception of available data and its proportions. He sourced this overview as a basis for a 3D animation.
He also shared his project, Divergence, which explored the point where the line between virtual reality and the real world disappears. The project focus was to visualize the answers to questions as data in space and over time. Marcin sculpted the answers to help his audience perceive them as tangible.
ENCORE: Data can be experienced as well as counted.
10. Visualisation, or the need to keep society in the loop
Iskra Velitchkova is a generative artist based in Madrid. She has worked in data visualisation for 10 years. At the beginning of the pandemic she quit her job and decided to create art instead. Making use of algorithms, she creates semi-abstract natural forms and speculative species of animals. She used to work mainly with d3.js. For her current work using algorithms, she works mostly with software like Processing and p5.js.
In one of her examples, she tried to break down the process of recommendations algorithms. For her birthday, instead of doing a Google search, she decided to take her motorbike and ride though the city of Madrid, searching for a restaurant she would like. The whole process took her a week.
She compared the time spent searching for the restaurant as distance travelled. If the personalized Google search would have taken a minute, her week-long motorcycle journey would equate to a plane ride from Madrid to New York City when compared to taking a boat. Flying created for us a way of travelling which doesn’t allow us to grasp the true distance.
Today, her main inspiration and goal is to connect society to massive structures of big data, in order to fully understand the needs of people.
Iskra described her experience curating an intimate event in a flat, where she invited people from many fields: scientists, engineers, taxi drivers, artists, politicians, and more, to debate and discuss her art and to reverse engineering her nature-inspired works.
ENCORE: Visualisation is a wonderful tool to expand access and understanding.
11. An intro to accessible data visualisations
Sarah L. Fossheim is an independent developer and designer focused on creating inclusive and accessible products.
There is a massive overlap between making visualisations easily understandable and making them accessible. If you are struggling to make a design accessible, it may be time to take a step back and search for an easier way to represent your data.
Accessible products are better products, Sarah found in their work. Accessibility doesn’t just mean giving everyone access to the same data, but giving everyone equal opportunities, regardless of their ability. There are visual, hearing, cognitive, and mobility disabilities which should be taken into consideration when creating a visualisation. But there are also text size, contrast, colour, ability of keyboard navigation, and the need for screen readers to transport the same information as the visuals.
Accessibility issues should not be the problem of a single developer at the end of the process, but rather should be integrated into the process from the start and considered at every step. This practice makes accessibility easier and leads to better results.
With all that in mind, it is OK to start small. If you find something to be inaccessible, just start the process to fix it, every effort made is better than ignorance. There are a lot of useful resources and tools to help you build accessible visualizations. To get started, head over to https://github.com/dataviza11y
ENCORE: If you are struggling to make a design accessible, it may be time to take a step back and search for an easier way to represent your data.
12. Prolific plotting: quantity for quality — see more, see more clearly
Ian Johnson has a background in computational math and has worked as a UX engineer at Google. In his presentation he live-coded and showed us how to analyse the data from the Google merchandise store whilst preparing a (fictional) presentation for our boss in only 20 minutes. Ian presented a new feature from Observable called Plot. It’s an open source library.
Ian demonstrated a workflow for visualising data that emphasized on creating a greater quantity of visualisations in order to reach higher quality. With his live coding he showed the speed at which design decisions can be evaluated.
ENCORE: By coding and designing together we can rapidly gain perspective on data.
13. Spaceships and reappearing graphics
Eleanor Lutz is a New York based Illustrator focused on infographics about space-related topics. She presented her work for The New York Times on a series of three different spacecraft from different countries that were built to explore Mars. In her presentation she compared her approach to each of the three spacecraft and showed how the information from the manufacturers influenced her work.
The first spacecraft she presented was a Perseverance Rover designed by NASA from the USA; built to explore the Jezero Crater. The Rover was well documented as NASA published the full 3D-model of the Rover, which allowed Eleanor to rotate it on the NASA website and trace the screenshots with Adobe Illustrator. Once she traced the screenshots she could shade and highlight the important parts of the spacecraft. She found this the easiest of the three visualisations since all the information she needed was available.
The second spacecraft was from the Mohammed Bin Rashid Space Center in the United Arab Emirates. The dataviz design process was harder because Eleanor didn’t have a 3D-model available and had to rely on 3D-diagrams and split the spacecraft into two parts. This process required she had to animate the model and interpreting the given information, drawing the top view, and going on from there. The hardest task was creating the animation because she had to rely upon and translate the visual representation from a video.
For the third spacecraft, Eleanor used information from China National Space Administration and from China Central Television. This turned out to be the most time-consuming project because she stepped up her process by drawing the model with Blender instead of using previous techniques such as paper crafts. As she was fairly new to the software it took her additional time to adjust. Once she figured it out, she prioritized getting the model into the same perspective as the other two spacecraft so that they could be printed on the same page.
ENCORE: Invest in the details for a worthy pay off; new skills can be useful on future projects.
14. No longer a bottleneck: the evolving role of the data visualisation practitioner
Rachel Binx has a background in math, art history, and interaction design. In her work she is interested in the relationship between geolocation data and the real world.
Rachel’s talk explored three main points:
- How data visualisation tools have become more widely available.
- The application and challenges of data visualisation within organisations such as Netflix.
- How visualisation can be more visceral and memorable.
When Rachel started out with dataviz, she was using Processing. Back in those days, it was necessary to “code uphill both ways”, as she put it. When Processing became P5.js and later D3.js, a data visualisation that would have been considered a Bachelor’s-level work was now widely available and relatively accessible as a template.
Variations of such D3 templates are still widely used today, but in Rachel’s practice, she discovered that she would need a more flexible approach for visualising data.
Rachel worked for Netflix for five years. They were using data to inform content creation decisions. As the sole data visualizer, Rachel felt like a bottleneck after a while. Using tools such as Jupiter, ggplot2, Notable, and Observable however, helped her enable a faster turnaround and increased collaboration with the data engineers. She also had to learn a lot of backend development, so that her work could be shared more easily and reliably throughout the organization.
Finally, Rachel talked about how dataviz might benefit from a less minimalistic approach. She argues that a more fun and non-traditional visualisation might actually be more intuitive to understand and memorable to recollect.
“Chart junk” as it’s dismissively called by some, may actually have an advantage in tying the subject matter to the visual representation because it reduces the distance between the visualisation and the subject matter.
Rachel further argued that visualisation should be opinionated. Abstraction can rob the data of its impact. As an example she compared two graphs showing total carbon concentration in the atmosphere. But one chart situated this data in the context of past prehistoric events, thus leaving a much bigger impression on the reader.
ENCORE: Less is not always more. Create an impact on your reader.
The summaries of the talks were written by students of interaction design in Zurich. Many thanks to Guan Arobei, Sandro Beti, Nicola Bischof, Eleonora Bonorva, Micaela Brazerol, Elena Decarlo, Bin Martig, Johannes Reck, Thore Reigber, Rejane Schrago, Miguel Seabra, Svenja Steurerjene, Janosch Tillich, Daniel Treystman, Silvan Weber, Nadia Westermann, Fabrizio Willi. Thank you Jurgen Spath and Marcial Koch for the opportunity.
S-H-O-W is organised by Graphic Hunters, a training institute in the Netherlands. ENCORE was the fifth S-H-O-W in the series to date. Previous events focussed on topics like emotions in information graphics, rules in dataviz and visualising impact.