Beyond Open-washing: Are Narratives the Future of Open Data Portals?
There is growing consensus that the mere release of open data is not enough to realize the full potential of openness. What’s next?
There is growing consensus in the open data community that the mere release of open data — that is data that can be freely accessed, remixed, and redistributed — is not enough to realize the full potential of openness. Successful open data initiatives don’t simply tick the ‘open’ box but produce data that actually gets used. Open data portals, in particular, are prone to the risk of becoming “data dumps”, where the number of published datasets counts more than their quality or utility.
This is why, when Sheldon.Studio was hired by the Matera | European Capital of Culture 2019 foundation to design their open-data portal, we felt we were in front of a unique challenge. How do we create an open data portal that empowers the audience, and how do we avoid an open data dump 🤪? One month into the project, here is what we learned in the process 😎.
Knowing the audience is the first step to an audience-centered data portal.
Last year was a big one for the city of Matera, a city in Southern Italy of 60K souls whose history dates back to the Palaeolithic, as it became European Capital of Culture 2019 and witnessed the arrival of more than half a million visitors. Not only tourists but also artists, cultural workers, and social operators swarmed through the city and actively participated in more than 2400 events, many of which spanned multiple days.
Can you imagine the amount of data visitors and citizens generated during the year? We can tell you about what we received: dozens and dozens of spreadsheets, some handcrafted, some software-generated; textual reports; photo galleries and video interviews. We could simply upload it online in some repo and be done with it. Yet, since the beginning of the collaboration, we embraced the idea of conceiving something beyond the usual. We wanted to give the data back to the people who helped produce it. This meant focusing on what the audience needed to understand.
Saul Wurman, who coined the concept of information architecture in the mid of the 70s, often said: “You only understand something relative to something you already understand.” This simple, yet timeless, statement represents an essential lens through which we design information experiences at Sheldon.studio. In practice, it means that we should design upon the past experiences of our audience in order to explain something novel. So, knowing our audience was the first building block of our design process.
Another key ingredient of our human-centered design approach is the preference for simple visualizations over flamboyant charts, especially when the fancier design would entail a compromise on clarity. Other than the complexity of the data visualizations, we instead leveraged colors and animations to keep our chart designs fresh and engaging, facilitating the audience in the comprehension of hidden data patterns.
From a design perspective, we rooted our visualizations around the central theme of showing the liveliness and the humanity that characterized the cultural programme of Matera. For this reason, we privileged rounded shapes and a profusion of dots swarming everywhere, a metaphor of humanity as seen from a bird’s eye view and we decided to present some visualization using the metaphor of a pack of many separate units/bubbles forming bigger clusters. We feel that this makes the numbers interesting and more intuitive to grasp also for audiences with lower data viz literacy.
In line with our endeavour to keep the data visualization accessible and easy to parse, we devised an innovative way to efficiently integrate legends, charts and text. Readers usually struggle as their eyes ping-pong back and forth between the chart and its legend to understand what’s what, so we tried to intertwine the legend in the descriptive text above it, highlighting keywords with the corresponding colours in the chart. The idea is to spark curiosity in the readers as they note that some words in the text are highlighted, or, the other way around, to prompt somebody to read the text while seeking for the legend.
Next? Plan for different data literacy levels.
The co-design sessions with our client, the Matera Foundation, surfaced the need to plan for multiple entry points and different levels of data literacy, to suit the needs of the different types of people that would visit the portal.
A first step in this direction was to include qualitative data alongside the numbers and statistics. We strongly believe that quantitative data is just one possible ingredient to the story, especially when we are discussing social issues, and moreover if it’s important to include a broader audience. For this reason, we combined traditional data visualizations with original texts, and we intertwined the data stories with photos and statements by the participants.
In its final version, the project unfolds across 8 thematic sections and 6 in-depth micro-stories. We opted for these two different content formats, sections, and stories, to offer two different ways of looking at the data. The thematic sections stand as metaphorical chapters that disclose the main narrative of what Matera 2019 has represented, providing a birds-eye view on the core values of its organization. The micro-stories, on the other hand, drill down on specific events or issues of particular importance. So, for instance, while the Cultural vibrancy introduces and visualizes the amount and diversity of the cultural program, the connected Open Design School micro-story unveils how the project brought talented youngsters from all over Europe during the year (see pic below).
The way we decided to publish the open data in the portal is itself an attempt at suiting the different data literacy levels and needs the website’s visitors may have. All the data is published in three places, each designed with a specific type of audience in mind.
- 🤓🤓🤓 A dedicated GitHub repo that provides the CSV and JSON files (as a data geek would expect them).
- 🤓🤓 An “Open Data Centre” on the website, which is essentially a traditional open data portal, listing all the raw data files along with their metadata.
- 🤓 An “Open Data Corner” at the end of each thematic section or micro-story, which includes only the data referred to the specific section or story. In each open data corner, we decided to publish not only the raw data but also the aggregated and processed data files that we used to produce each visualization that is on the page.
We believe that the latter, the “Open Data Corner” is a core innovation in the way we designed the portal, as it empowers people who might have a lower data literacy than a typical open data user, like concerned citizens, activists, as well as journalists, to access and play with the data in a beginner-friendly manner.
Most of all, think of data as a means to an end, not the end in itself 😏.
The more we figured out how to translate the principles of human-centered design into the practice of creating the Matera 2019 data portal, the more we realized we were shifting the role data has in a traditional open data portal.
Open data portals are typically all about the data: how many datasets, how many formats, which open licenses. Open data portals are typically all about the data: how many datasets, how many formats, which open licenses. In Matera 2019 this hierarchy is flipped: the stories come first which narrate the data and illustrate what can be done with the data, then we provide the downloadable open datasets.
In addition, a standard open data portal will include mainly quantitative, machine-readable datasets. In the Matera 2019 open data portal, CSVs and machine-readable datasets are just one of the many components of a multi-modal narration, together with texts, videos, pictures, etc. The datasets are not stand-alone elements, but parts of an informative ecosystem covering the many facets and the complexity of what Matera European Capital of Culture 2019 has represented.
Finally, our hope is to give rise to a recursive process that sees data as a means to an end, not an end itself. Publishing the data online was not the ultimate goal of the Matera 2019 Open data portal. It is humans and their actions that generated the datasets behind the stories of the portal. And now that the data is published, it should serve this community. We want to see the data used as a tool to foster new human interactions and to inform new processes aimed at improving the conditions of Matera’s society.
With this goal in mind, an integral part of designing the Open Data Portal has been that of planning for its legacy. In the autumn 🍂, we are supporting the organization of a DataSchool, together with the Matera Foundation and with the participation of the open-data guru Maurizio Napolitano. The School will bring in the city a colourful variety of data-people, from data-activists, to design students, journalists, or social scientists, to design new forms of communication and services based on the data.
Through the design of the platform, we aimed to turn open data into commons, public goods generated and maintained by the community for its wealth and awareness. Our hope, indeed, is to contribute to a more inclusive data practice, which embraces a broader audience, provides diverse and faceted entry points for personal explorations, and constitutes a stepping stone towards new forms of information, knowledge, awareness, and social care.
Matteo Moretti is a designer and cofounder of Sheldon.studio & Alice Corona, data-journalist and founder of Batjo.eu
