Introduction

As humans, we have always had the urge to chart the world around us. This urge has pushed us to improve the way we collect, process, and communicate information throughout history.

Our progress on building and implementing more structured methods over the last centuries allows us to distill patterns and trends much better as our data collections keep growing. We can see this, for example, in the structured long-term weather data collected since as far back as 1657. Eventually, consistent weather tracking evolved into how we record temperatures since 1880, data we still use today, enabling us to visualize global temperature changes over time.

Earth Observatory, world of changes: global temperature anomalies (graphic source: NASA)

Technological advancements have undoubtedly contributed to improved quality of data recording, analysis, and presentation. Particularly nowadays, in the information age, where the web had a great impact on how we create, process, consume, and interact with data. These developments now allow us to gain, with the help of data, better knowledge and wisdom about our universe, planet, society, and ourselves.

Over the course of time, different developments and influences –from technological to user research advancements– have brought us to today’s state of online interactive data visualization. To understand this, we have to go back to the development of Atari, Apple, and Personal Computers in the 1970s.

PART 1 - Interactive data in the pre-Web era

Before 1994

Map example: SYMAP output images, the first version of this automated computer mapping system was created in 1963 by Howard Fisher and was the first system that included spatial-analytic capabilities (Source: Wikipedia and Esri)

The start of digital information
It was back in the seventies when information was starting to leave books, press, and television, and moving into Personal Computers in the offices and homes of those who had the money, time and skills to own and operate them. Those early domestic machines, even if with great potential, still lacked usability and were restricted to a relatively small global community.

Interactive scatter plot example: Demo by John Tukey & Marian Fisherkeller about Prim9, a first for computer-assisted data visualization, recorded in 1973.

Limited data
During this period, technological development was promising and exciting. However, most data was only locally stored and not accessible to the general public unless (occasionally) translated to existing communication channels, like newspapers and tv shows.

Exploring navigation
Hardware-wise, this was the time when exploration of how we humans can interact with the computer and the data it can present was taking place. First attempts of touchscreens, joysticks, and input tablets were developed, built, and tested in business, educational, and research fields.

Interface navigation example: on the left the Rand input tablet in use since 1963 (Graphic source: History of Information), and on the right the PLATO IV one of the first touch screen devices using a 16x16 grid of infrared lights to track what the users touched created in 1972 (Graphic source: Pen Magnet)

Complex software
The first attempts at interactive data visualizations were created in university settings, leading to the first full-color computer graphics. Working with these, however, required solid expertise in how to code them.
Software was developed with experts on specific sectors in mind, and therefore the user base was limited. By this time, the focus was on advancing technology, not thinking much of usability, and the questioning of the ethics of software was barely acknowledged by society.

Advertisements to promote personal computers (Apple II and Atari 800) showing bar and line charts, published in 1979 (Graphic source: Geek Lord & The Mothership)

New ways of interaction
As it kept evolving, more sophisticated software required new ways of interacting with computers. As demoed in 1968, we started using more and more basic mouse behavior like pointing, clicking, and scrolling. These turned out to be great ways of interacting with data and other information on screens, but the hardware this required (e.g. mouse) was not commonly used yet.

Interactivity example: software to create interactive flowcharts for the Lisa computer by Apple (approximately 1983)

Easier to use software
The development of easier-to-use specialized software meant that we could not only work with text and digits, but we could also work with graphics. We started using computers for data visualization thanks to the charting software first released in the late 70s and beginning of the 80s, such as VisiCalc, Matlab, and Lotus 1–2–3. These tools still required training to be able to use them properly, though this became acceptable for the first group of educational, business, and even home users.

Chart example: VisiCalc software in 1980 (Graphic source: Atariwiki)

Hard to share
The amount of digital information started growing considerably. However, even if we found ways of building and manipulating digital information, sharing it was still restricted in most parts of the physical world (we either printed it or stored it in drives that we could then carry to other computers).

Visualization example: Information visualization tool, showing how we can interact with huge information sets, made Christopher Ahlberg and Ben Shneiderman named Film Finder made in 1994

Our future visions back then
By the 1980s, the possibilities and rapid advancement of technology made us dream. The idea of interactive screens and smart boards that would allow us to touch and use speech to interact with data visualization and other graphic information felt more physical and human than using a cursor on a screen, and seemed like a possible achievement in a matter of a decade or two.

1988 future vision:
This video shows how Apple envisioned how people could use the computer in the year 2000:

“A computer requires users, not watchers”: Even if the technology wasn’t there yet, we already acknowledged that sharing information (in educational, leisure, and business contexts) had a greater impact on effective interaction.
There was a big desire to not just consume content but also have the ability to access the underlying data and sources.

The potential of computers felt even too restricted when compared to futures described in novels and movies like Tron, which were suggesting us gateways into the actual digital world and interactions with virtual intelligences.

Movie example: TRON movie shots made by Disney in 1982 (Graphic source: Filmgrab)

Growth of users
We successfully pushed and showed what technology could do, as computers evolved and became more powerful and affordable. We had growing data, improving software, and better processors, and the mouse became a standard part of a PC setup. Thanks to the many advances and improved affordability, using a home computer became popular and led to millions of new users.

Number of personal computers (PCs) sales in the world (Source: Jeremy Reimer, Gartner via Wikipedia)

The growing number of computers and users started outlining the nodes of a network that did not exist yet. But soon enough, we came up with new ways of sharing and visualizing data and information beyond disks, and this eventually led to the birth of ‘the web’.

PART 2 - Web 1.0 - The early days of the internet

Approx. 1994–2004

Example: Network visualizations by R. A. Becker, S. Eick, A. R. Wilks made with SeeNet published in 1995 (Source: Penn State University)

The start of the internet
The internet appeared as a great way of disclosing and consuming information. Web 1.0 was mostly providing ‘a lot of information’ and was also mostly used by people that knew how to publish and access information. For those not technically skilled, it was known as the “Read-Only Web”, built around brochure-like pages that were updated manually allowed limited interactivity from users.

State of data visualization
Back then, data visualization on the web was reduced to static images produced with third-party software and was often targeted to niche audiences. Yet, the third-party software did allow us to generate visualizations faster and in new ways. Stand-alone programs allowed better interaction, motion and new ways to experience data.

Web visualization examples: Weather.com screenshots of static images, with left current weather overview in 1996 and current heat index in 2000 (Graphic source: Way Back Machine 1, 2)

Discovering interactions
The early internet was not necessarily user-friendly but its potential was evident. To interact with all the available information that the internet had to offer, many started to try and create a new interface to access this information. This led to a range of solutions differing from very functional to more engaging, and from abstract to very descriptive.

Interface example: Weather.com weather information main navigation in 1996 (Graphic source: Way Back Machine)

The adoption of the web might have taken longer than expected, however, those who were working on the internet were convinced it would eventually be used by many. But before that happened, its many limitations needed to be addressed. Internet connectivity needed to be improved, economically viable, and widely available.

The need to make computers more user-friendly and accessible to broad audiences contributed to the definition of User Experience Design in 1995. The acknowledgment that every stage of the human-computer interaction needed to facilitate online experiences turned design into a fundamental tool in the development of new hardware, software, and interfaces.

It was around the same time that programming languages that allowed for more complex web interaction were developed. The first version of JavaScript became standardized, and Adobe Flash was introduced and widely used. As user needs started becoming more central in hardware and software development processes, the tools that supported greater web interactivity became common.

Visualization example: Keyhole Earthviewer software was first released in 2001, the software would eventually be acquired by Google and laid the foundation for Google Earth

Available data increased
Web 1.0 allowed us to exchange data in certain cases easier and faster. The web provided ways to centralize and gather existing data in online repositories, and also to disclose it. Online open data catalogs were in their infancy, and it would take some time to make them more discoverable, accessible, and better populated.

Data portal example: OECD DAC data in 1997 was made available online in Excel documents, World Bank data in 2000 was available online in tables provided in the hard to use pdf format (source: Way Back Machine 1, 2)

Our future visions back then
The potential of all these new technologies, data management methods, and connectivity made us imagine their use in the future. We envisioned home and office interactive screens, as well as portable devices and wearable gadgets that could be used outdoors to message others and gather data, images, and videos. These visions came with a call to action to all those involved in the development of new tools: “Technology does not drive change at all, technology enables change”. It was the time to imagine the future and put technology to its service.

1996 future vision:
This video shows how Philips envisioned how people could use the computer in the year 2000:

Computers were destined to leave desktops and become smaller and even wearable. This made us imagine new ways of interacting with them, as well as collecting, consuming, and sharing data.

As the new millennium started, we imagined new ways of interacting with screens and visual content. We envisioned a Precrime Chief Tom Cruise loading, dragging, playing, and zooming into high-resolution criminal memories in Minority Report (2001). Multi-gesture, fluid, and intuitive interactions seemed like a plausible reality by 2054 by simply using glove-wearables and a glass-screen projection.

Movie example: Johnny Mnemonic movie is based on a character by William Gibson. Gibson started the sci-fi subgenre called ‘Cyberpunk’ that had a big impact on sci-fi movies in the 80s and 90s and ultimately on how technology and the internet itself evolved). This Johnny Mnemonic movie was published in 1995, and shows the potential use of Virtual Reality in the future and features ‘Data gloves’ that shows gesture-based interactions to navigate information. This would inspire movies like the The Matrix and Minority Report.

Better user experiences
Technology still had a long way to go until reaching the potential and user base envisioned by future visions by media and the private sector, but we were making progress. The web would eventually get more traction by the end of the 1990s as home network access improved and the web became more user-friendly and richer in content. We also saw the internet experience improve with the broader adoption of new browsers (Internet Explorer in 1995), search engines (Google in 1997), and other popular web portals (eBay, Yahoo, and Amazon in 1995). By the year 2000, over 400 million people had access to the internet, and they weren’t looking to simply consume information; instead, wanted to actively participate in it.

Global internet users (source: Our world in data)

PART 3 - Web 2.0 - The rise of user-generated content and interactivity

Approx. 2004–2014

The start of connected and fully interactive internet
The web 2.0 movement in 2004 became an inflexion point on how we interact with the web. We went from being information consumers to becoming active online participants, sharing our opinions and reacting to content. This was only possible by new browser technologies that introduced interactivity combined with the rise of social media platforms like Facebook, YouTube, or Twitter that placed users in a more active online role.

Computers became more affordable, the internet attracted more users, and so did social platforms.

By The Bureau of Labor Statistics (Graphic source: Free by 50)

In addition to becoming more social, the web also became more sophisticated. The interactivity that was possible thanks to new ways of web development, including new JavaScript frameworks, allowed users to explore online resources that helped them understand the world, including real world challenges that we needed to address with urgency.

The available data, javascript frameworks and browser plug-ins like Flash let any user interact with large datasets and distill their own stories, enabling them to create their own perspectives on any topic. New online data visualization tools like Google Maps, Visual Thesaurus, and Gapminder were released and quickly became popular. These portals and their underlying data led to the development of new tools. News websites added interactivity to weather reports or elections results, and started to include the first interactive graphs to articles. We also kept testing new abilities to interact with huge datasets.

Visualization example: Google Consumer Barometer made in 2011 by CLEVER°FRANKE

Interactivity did not only improve thanks to out better understanding on how to make engaging with online visual data more intuitive, it also improved as more complex Javascript libraries, like D3.js or Highcharts became available. These libraries allowed for more elaborate interactions and more sophisticated visuals that could make users want to explore further.

During this time, we also saw improvements in the way we create better online experiences around geospatial data. OpenStreetMap appeared in 2004 as an open and community-built geographic database that provides quality and regularly-updated data for basemaps and data layers that can be used in webmaps. Better and more available data together with new online services like Mapbox allowed the building and integration of performant webmaps into websites and apps.

Interaction conventions
In the early days of the internet, interface exploration and experimentation led us to eventually come up with conventions of how to build user-friendly websites. A similar process occurred in the 2000s with interactive data visualization. Over time, we were able to identify the most intuitive ways of interacting with data visualization, and users familiarized themselves with these new gestures: hover on a bar chart or scatter plot to display the y-value in a tooltip, click and drag to pan on a map, scroll to continuously zoom in and out, or double-click to zoom in in steps. And while we continued to define and respect these rules, we also discovered that we could break them in more experimental and playful visualizations.

Online data visualization example: Hans Rosling Ted talk using the online Gapminder tool to tell an engaging story in 2006

Outside of the web browser, we kept testing new hardware setups to navigate data, some of those inspired by movies released in the 90s and early 2000s.

Interface example: The G-Speak gesture based interface made by Oblong in 2008

Interface example: Google Glass a wearable, a voice- and motion-controlled interface, explained by Sergey Brin in 2013

Rapid increase of data
At the same time, we saw how data availability greatly improved. Firstly, by releasing public datasets and making them readily available. After that, by private parties and internet users that started sharing their own datasets and linked them to other existing datasets. Finally, a more active internet resulted in vast amounts of user data that quickly became valuable and attractive to many parties.

Visualization example: In 2006 the Dutch website Buienradar.nl (rainradar) started to provide free rainfall radar images without delay, using data from a radar operated by the the Dutch meteorological institute, the KNMI

The collection of vast amounts of quickly generated data from multiple sources brought us the buzzword “Big Data”. This challenged the ways we could store, process and analyze information, and we also started working on making it more accessible and interesting for all users.

Visualization example: Overview of linked open data cloud in August 2014 (Graphic source: Wikipedia)

This led to the ability of many to extend their existing datasets with other datasets to enhance their analysis with more measurements from more locations, more groups and historical data. More data became available with greater quality and granularity.

Visualization example: Flight patterns network visualization based FAA flight data available since 2007, visualization created in 2009 by Aaron Koblin

‘Data porn’
With the vast amounts of data available, technologists, data analysts and designers explored new interactive ways of providing insights from these huge datasets. This led to all kinds of graphic and data rich visualizations that eventually would also lead to a massive increase in popularity of data visualization. Some started to phrase these extremely data rich depictions as ‘data porn’, where for some it was not anymore about conveying information, but about creating something that was purely there for aesthetic reasons.

Visualization example: Weather chart of the 2011 Dutch online sentiment scraped from ±1.000 sources about the weather vs. the real weather data publicly available via the KNMI made by CLEVER°FRANKE

New devices
With the new technology, improved interaction, and greater data collection and availability, online interactive data visualization became a powerful tool to track the world around us and all the changes that were occurring in the digital world. The new wave of interactive data visualization broke out of the boundaries of technical or specialized fields, and became a concept that the general public could engage and work with. Eventually, interactive data visualization went mainstream.

Interface example: Introduction of the iPhone smartphone by Apple in 2007

This interactive data visualization hype followed a major breakthrough in the history of communications. The release of the first iPhone in 2007 was very positively and quickly received by the public as an improved model of other commercially available smartphones, with a touchscreen with double the resolution of other models that allowed multitouch and with access to the web similar to that of desktop computers. Smartphones only became more powerful, functional, affordable, and widely adopted since then.

Global phone sales (source: Gartner and others; via Wikipedia)

Personal data collection
These new ‘smart’-phones also became personal data collection devices. With all the sensors embedded in them we could, for example, keep track of our running & cycling exercises, and also check our heart rate daily. It introduced a way for us to gather a lot of new data that was just about ourselves. Adding to this, we could continuously exchange and enrich this data through the day (due to the smartphones’ ability to continuously connect to the internet), and data visualization became a new relevant way for users to decide what to do and adjust their behavior.

Interface example: Sensors and other active components in iPhones.

Quantified self and playful data
With the power of smartphones and other data collection devices, our urge to keep track of our habits grew, and also the quantified self movement attracted a bigger crowd. Data collection was no longer a thing for just business people and scientists, but also for the general public. We even started to see the joy in collecting personal data and manipulating these collections, often with playful results.

Playful data visualization example: Using running tracking app and the path it creates to create playful graphics by Leah Culver in 2013

Small touchscreens
With the new smartphones, screens reduced their size, and we started interacting with touchscreens. “Clicking” became “tapping”, and icons gained more importance and significance as reduced space needed them to become more intuitive.

Touchscreens made user interfaces more flexible, and with more gadgets that could communicate among them, user experience became device agnostic, jumping between smartphone to desktop and, in the end, smart TVs. Everything became connected, creating one big connected digital universe.

In just a decade, the number of global internet users grew threefold, and passed the 2 billion mark. We experienced by this time how user adoption was not only a response to improved technologies that allowed users to share their opinions, but also to new intuitive and user-friendly interfaces and devices like smartphones that eventually became more affordable. In addition to this, network connectivity went wireless and saw its speed and global coverage greatly improved.

2010–2014 future visions:
If you look to future visions movies produced by HP, Intel, Microsoft and Nokia in this era everything becomes a screen, everything is connected and everything can become content that is published somewhere. What is remarkable is that these kind of movies also don’t predict the future of 20 years ahead, instead, they imagined a future that was just a few years ahead.

Movie example: Guardians Of The Galaxy interface overview made by Territory Studio in 2014

Movies about the future also showed interfaces and visualizations that are very close to what was already feasible to produce (only the widespread transparent screens we did not manage to make ubiquitous yet). To some extent it is even using the retro visual language and low-tech display techniques we saw in old movies.

Using all data
In this era, we were growing more and more connected and our internet engagement started being translated into datasets that reflected our personal online habits, years before we questioned and regulated its implications. Back then, connectivity data was intriguing and made us curious about our collective habits, and made us want to bring physical and digital spaces much closer. For example, we visualized how Amsterdam celebrated the 2007 New Year’s Eve through changes in the volume of text messages over space and time. This visualization later led to the questioning of the use of consumer data and, in fact, the disclosure of personal data in the telecommunications sector would have been more restricted a few years later. Also with the introduction of the Google Glass the question rose if a device that can see everything your eyes can see should be allowed to continuously record camera footage of its surroundings, particularly in public spaces.

Eventually there was more legislative action that strengthened data privacy laws and favored transparency and accountability; but users continue to demand protection. These demands for privacy and transparency, combined with improved technical developments as well as machine learning and artificial intelligence analysis, contributed to a new shift in the web.

PART 4 - Web 3.0 - The start of the decentralized, user-owned and immersive web

Approx. 2014–2024

The next step of digital reality
By the year 2014, we realized we needed more than just sharing our opinions and content through the existing platforms as the intermediaries, so Web 3.0 and other initiatives like the Fediverse were born as a more decentralized internet. The new web allows us to not just read, and write on the internet, but also to own what we do there. A web built around decentralized governance, and also around new and improved technologies that pave new ways of processing information and building more personalized and immersive experiences: machine learning (ML), artificial intelligence (AI), virtual reality (VR), and augmented reality (AR).

Keeping track of physical life in our digital space
With better coverage of internet connection and improved transfer speeds, devices became smaller and less energy-consuming. A whole new type of data capturing devices became popular. This was the birth of the smart movement, which resulted in smart watches, smart homes, smart industries, and smart cities. These devices captured data using different types of sensors and enabled us to keep track of what was happening in the physical space. All this slowly contributed to making our physical lives part of our digital life and vice versa.

Interface example: Apple’s Smartwatch, first released in 2015, allows users to visualize their heart rate activity, burnt calories, minutes of brisk activity, stand-up time, and movement (Graphic source: Apple)

The flip side of these new tracking devices also became apparent: digital data has consequences in the physical world. For example, an open fitness tracking app that brings us together, once revealed the location of secret US army bases in Afghanistan.

Visualization example: Fitness tracking app Strava shows an outline of secret US army bases publicly in 2018 (source: The Guardian)

Easier tools to create interactive visualizations
But beyond the growing amounts of user-generated data and their implications, the new web also started hosting new tools that allow to keep track of organization data in a structured ways, and other easy-to-use tools that allow to create interactive data visualizations.

Overview of data visualization tools (Graphic source: part of book ‘Better Data Visualizations’ by Jonathan Schwabish).

By this time, the technologies to compile and chart quality data have improved to the point of allowing for the integration of interactive data visualization across platforms and targeted towards different users. On one hand, we find interactive data visualizations for the general user, like those integrated in personal devices that we use in our regular daily lives (sports, leisure, weather, health…), or as part of visual narratives and storylines that keep us informed or learning about the state of the world and major events (like scrollytelling articles in news sites). On the other hand, we also find interactive data visualizations designed for more targeted users, like industry-specific business intelligence tools and dashboards used for internal strategizing and planning, or power user-oriented and custom visualizations for research purposes.

Data amounts skyrockets
We didn’t only find new and more convenient methods to chart data for specific user groups and their preferred interfaces. Over the course of time, the costs to store data plummeted.

Historical cost of computer memory and storage (Data source: John C. McCallum (2023), graphic source: Our world in Data)

With all the data generated by companies, governments and people, the amount of data we store globally rocketed. Besides that, new currents of data are continuously added to the existing streams of data. ‘Big data’ becomes super big.

Annual size of the global datasphere (calculated and predicted by IDC Global DataSphere in 2018. Graphic source: IDC)

We now increasingly rely on the help of algorithms based on machine learning and artificial intelligence to automatically detect what information has priority. These algorithms, controlled by ourselves or others, have the power to consciously or unconsciously influence what the end user consumes.

In this sense, in the new web, we don’t keep track of everything anymore; but we or others determine more and more the rules of what we see instead. Data visualization becomes a tool to give us an insight into all available data and help us to define the rules that provide us with information, and with that our view of the world.

Controversy
Data is now captured, exchanged, and stored everywhere and is analyzed by many. In some cases, for good causes and in other cases, for illegitimate purposes. Sometimes, it can also unintentionally turn into a negative societal impact.

We now also talk about ‘Big Tech’ (companies like Alphabet (Google), Amazon, Apple, Meta (Facebook), and Microsoft) and the big influence that these technologies, software and social media platforms have on our society.

2014–2024 future visions:
More and more books are published also pointing out examples where things went wrong with the use of big data and data visualization. Books like the Weapons of Math Destruction and New Dark Age paint a sinister future of our digital and physical life. The fiction George Orwell described in his famous 1984 book seems to become partially reality. Also series like Black Mirror and movies like The Circle focus on the negative impact data can have for humanity.

Movie example: The Circle movie, released in 2017, based on the novel by Dave Eggers

The bright future of tech and data full of potential, turns into a dark future vision, full of threats.

Redefining the internet
New initiatives of the decentralized Web3 and others like the Fediverse were born as a movement that advocates for less influence by the Big Tech and a decentralized internet owned and governed by its users. This can also lead to new ways of keeping track of our data, such as not storing and accessing it via the Big Tech companies but saving it ourselves on our own (collectively owned) infrastructure.

Interactivity for new dimensions
Data visualization can also become a more immersive experience in the metaverse through the use of AR and VR technologies (developed by Meta and other tech actors like Microsoft, Nvidia, or Roblox). Even in this new technological space, data visualization will continue to make the most of advancements in other fields. For example, Unity, originally developed for the video game sector, is now being implemented in the data visualization and animation movie industry.

AR visualization example: EnARgy experiment by CLEVER°FRANKE in 2017 showcasing the amount of energy used by each device in a room using object recognition

However, as it has happened throughout history, not every innovation has been well received by users. Google Glass was released in 2013 as an early prototype of AR. This head-up display, running on natural language voice commands and connected to the internet, offered a new way of finding and representing visual information that did not do well in the market (and, in fact, it is no longer commercially available).

Data display example: Microsoft Hololens 2, released in 2019, showing a digital twins and holoportation demo

Similarly, the first affordable and crowdsourcing-funded Occulus VR set still doesn’t have the amount of users as many predicted when it was first introduced. In fact, it is estimated that VR headset sales worldwide declined 12% in 2022.

Regardless of their reception, these have been, however, bold attempts at challenging the traditional WIMP (windows, icons, medium, pointer) interfaces by allowing interactions through voice prompts that take advantage of the continuously progressing natural language processing capabilities.

Data visualization example: Mashrooi platform made in 2017, to showcasing to progress of real estate projects in Dubai, designed by CLEVER°FRANKE

AI morphs data into new forms
At the end of the Web 3.0 era, AI tools like Dall-e, Midjourney, and ChatGPT emerged and took the internet by storm. The tools showcase how huge amounts of text and images can be interpreted by AI and used to generate realistic visual and textual results based on the user’s desires. The results seem to be generated by humans, while they aren’t. The tools are built upon massive datasets, consisting of mainly human-generated content. The analysis of those datasets using powerful algorithms and models that consume enormous computing power result in these new AI models and tools.

The impact on our data and visual culture is yet unknown, but is already leading us to many questions like: What is the role of a creator in a world powered by AI models? Is the owner of the data used to train an AI also (partly) the owner of the AI model? Can we distinguish AI-generated content from real-life situations and human-crafted content? Will ‘the best’ data visualization eventually be generated automatically by an AI? What will be the implications of AI for privacy and accuracy?

Generative AI tools overview by Antler

The AI tool landscape is growing rapidly, and also includes tools that help us easily analyze and visualize data to extract insights. These AI-powered analysis methods range from simple tools, like enhanced Google Sheets, to more complex ones, enabling us to turn a dataset into digestible information automatically.

AI tool example: Google Sheets Explore Tool by Tom Mullaney (note: currently unavailable in Google Sheets)

In addition to allowing for data analysis, the new AI tools also give us new means to visualize data in new and unexpected ways. Like we at CLEVER°FRANKE did with this experiment using AI tools and the paintings of Hieronymus Bosch to showcase deforestation in the world since the date Bosch was born:

AI dataviz experiment: Using data & AI (with the help of the tools Deforum on Stable Diffusion and ChatGPT) to visualize deforestation, using the paint style of Hieronymus Bosch as input.

User needs become society needs
As it happened in previous versions of the web, we now see how technological developments force us to understand how users can make the most of them, and also how addressing user ethical demands becomes a core factor of the internet experience.

The new ways of processing and interpreting information as well as the new formats, interfaces, and devices used to visualize it are still in exploration and refinement. But we’ve seen over history how pioneering through technology and design eventually builds a larger user base on a longer term, one that will also be more private, personal, fair, sustainable, and egalitarian.

Even if we are still exploring the opportunities, capabilities, and limitations of Web 3.0, data visualization becomes both a tool to understand and track the technologies, but also an active participant of it, taking advantage of the developments in data processing and communication.

PART 5 — The future of interactive data visualization

Predictions from 2024 onward

Progress is built upon learnings from the past. In just three decades we’ve made major advances in interactive data visualization as a result of progress in four influential drivers of change:

1. Technology development
   Technology development resulted, on one hand, in improved processing power that gives way for more complex and faster analysis. On the other hand, we gained new interaction and (portable) display possibilities with new hardware and software tools. Extensive and more reliable networks were built and improved, transfer speeds increased, and so did the ways in which we collect and store data, which allowed us to create, share, and access data anywhere.
2. Available data
   There is more and more data available, in better quality and more open to everyone. We moved from collecting data on a global/ country/organization level to very personalized datasets. The goal of data collection shifted from keeping track of our world on a meta-level to keeping track of our world on a micro-level.
3. Usability and user base
   By the increased availability and affordability of hardware and software the user base grew rapidly. More users started creating and using interactive data visualizations. This also led to the demand for easier-to-use tools to create and interact with data visualization, and along the way, conventions on how to interact with these tools were settled.
4. Societal values
   Ultimately, all these developments resulted in users asking for greater alignment with societal values, which are our common shared ethical demands on the impact of technology on society, like privacy, governance, equality, or sustainability.

The progress and knowledge generation across these four drivers varied over time. The technological sector and society have assigned different weights and priorities to them, which has brought us to today’s state of the web and online interactive data visualization.

The more recent inclusion of societal values in data collection and communication could normalize the widespread adoption of data visualization; following the consequences of the hype we experienced 5 years ago. However, we could also expect that with technology ever improving, new ways of collecting and consuming data will bring us new opportunities for data visualization adoption across fields.

‘The long nose of innovation’ model by Bill Buxton published in 2008

But this could also be the end of interactive data visualization development, and its currently fully adapted state could bethe best that it will ever be. If we look at Gartner Hype Cycle (which explains the “maturity, adoption, and social application of specific technologies”), we could expect data visualization to have already plateaued years ago. And if we look at “the long nose of innovation” model by Bill Buxton (which suggests that there’s a lag of 10 years between technology development and adoption), we already surpassed the 1 billion market size of data visualization by a mile. Did we really reach the end of the development of interactive data visualization?

Where is our vision?
In the 1980s, we dreamed of interactive screens and real-time analysis. Tron painted us a future where we could become part of a virtual reality, and literature hinted at immersive digital worlds. Today we seem to have stopped dreaming or, at least, we have stopped having a hopeful and optimistic lens. Movies portray future visions as how we already thought the future would look like years ago, or predict a dark and somber future. And no real long-term future visions seem to be shared by corporations anymore.

Have we stopped dreaming about the benefits that information/data visualization can bring us?

As we look at the progress from the past three decades, we can’t help but wonder what the new ways of consuming, creating, and sharing information will be. We believe that we, as designers and creators of interactive visualizations, can still push it to a new level.

Paralyzed patient using Neuralink to play chess

Do we still need data visualization in the future?
These ambitions to continue progress are already resulting in new and promising technologies. With tools like Neuralink, that allow users to have a direct connections from their brain with their computer, we can even question if we still need ways to visualize data, now that we are almost able to connect basically our brain to every (open) data center in the world. By incorporating AI analytical techniques and the power of super computers into these products, we could potentially derive conclusions from that data, extend our brains, and use all the available knowledge in the world in real time. Will all the world’s data eventually become instant wisdom for everyone, skipping the steps of transforming data into information (by analyzing and visualizing it), and turning it into knowledge (by creating insights / stories)?

Our vision of the the next major driver for the future
We don’t think it is reasonable to expect from new AI techniques and greater data access to instantly derive the ‘right’ wisdom at the ‘right’ moment. Interfaces will continue to be essential to manage the information that makes up our thoughts.

“19th century culture was defined by the novel, the 20th century culture by cinema, the culture of the 21st century will be defined by the interface.”

By Lev Manovich author of books on digital culture and new media, and founder of the Cultural Analytics Lab.

After exploring the drivers that have shaped today’s interactive data visualization, we expect these three main pillars to have a significant impact on how we will visualize data in the coming 10 years:

• Adaptive personalized insights
  Democratizing access to quality-data-derived knowledge will require the implementation of different new methodologies and AI techniques to control our digital devices (i.e. using natural language in voice commands or through gestures). This will not only allow us to adapt the content of the message but also the form, level of understanding, etc. Progress in this field will make data insights more easily available for all types of media consumers across cultural and educational backgrounds, experience degrees, skillsets, physical conditions, or available time to digest the insight.
  
• The meta interface
  As our digital and physical worlds become more intertwined and complex, it is imperative that we maintain control over our data-driven world. Fortunately, AI tools allow us to structure data automatically, detect relationships, and extract valuable knowledge. In fact, we can now derive insights without manually exploring the data. However, to make informed decisions based on our available data, information, and knowledge, we require an overarching visual overview that can help us determine the most insightful message within data. This is why some experts refer to this era as ‘the age of the interface’, where interactive data visualization could potentially serve as the key interface for ‘directing’ our data-driven world.
  
• Checking data quality
  There is a growing need to keep track of the online information subjected to fake news, data privacy concerns, a decentralized web, and black-box AI solutions. We seem to be slowly shifting from a “more-is-better” attitude (big data, big tech…) towards a “higher-quality-is-better” attitude ( fact checking, crowd owned, …). How do we keep track of data quality? Will elaborate interactive data visualizations allow us to better understand how information is created, check sources and define what the quality of information is?

Conclusion

Although it now seems we could posses all wisdom in the world at a glance, our belief is that data visualization still has much potential to play a crucial part in addressing current and future societal challenges, empowering us to maintain oversight over emerging technological advancements, and ultimately equipping us with the insight to improve our own well-being, that of our environment, and the of digital realm. How this will look is for us yet unclear, and so there lies a huge challenge for designers, developers and information specialists to define this. But we believe that ‘the real interface to our data-driven’ world is yet to be discovered.