Let's talk about data

Any person who does excavation work should click or call before digging

What is open government?

Open government, according to the Government of Canada, is to “create greater transparency, accountability, increase citizen engagement, and drive innovation and economic opportunities through open data, open information, and open dialogue.”

What open government means to us

• To provide Canadians access to all NEB documents and enable transparency;
  
  
• To improve accessibility and understanding of government data;
  
  
• To showcase what we do and how we perform; and
  
  
• To engage and interact with Canadians – receive their feedback and search for ways to implement it, when applicable.

Artists and scientists both strive to see the world in new ways, and then tell others their insights. Data visualization relies on both disciplines as storytelling tools.

The National Energy Board (NEB) is using data visualization to turn massive spreadsheets of data on Canada’s energy sector into visually appealing images the mind can more easily understand.

To this end, the NEB established a three-year collaborative initiative with the Interactions lab at the University of Calgary, VizworX, a specialized software development company, as well as TheHesterView Inc., a leader in data innovation projects.

Together, the team creates visualization tools through a thorough multi-step process that helps them understand and structure the data. Here is an overview of how the process unfolds.

Step 1: Prepare data primer and data set

The development gets under way with identifying and preparing needed data and meeting with the NEB subject matter experts to build an initial understanding of the data.

Step 2: Understand the data; what questions it can answer

As the team gets familiar with the data, they gain a sense of what stories the data can tell, and identify what questions the visualizations can answer.

Step 3: Data and design workshop

Developing visualizations in an “iterative” process – which means concepts are developed, analyzed and refined repeatedly.

Early on, the group meets for an intensive data understanding workshop at the NEB, involving the technical, design and visualization teams.

At this stage, the key pieces of data to highlight are selected and preliminary ideas are sketched out and shared internally.

The NEB visualization team acts as an interpreter and coordinator for the various groups – internal data experts, design team and technical team – to ensure an end-product that delivers on all goals and requirements.

Step 4: Develop, analyze and refine preliminary ideas

After significant idea generation, design development and fine-tuning, the visualization is ready to be brought to reality.

Step 5: Interactive and accessible web app created

VizworX then works on the software coding that will turn the ideas into reality. Its team builds the visualization for the web – creating the source code and bringing the interactive features to life.

Step 6: Fine tune visualization with input from designers and the NEB

The process of turning the visual design into a full-fledged web app involves constant communication between designers, programmers, and the NEB visualization team. This collaboration also results in some final refinements to the visualization design.

Step 7 & 8: Quality checks and upload final visualization to the NEB website

After final reviews and quality checks, the visualization is uploaded to the NEB’s website and optimized so it’s easy for search engines, such as Google and Bing, to find it.

Step 9: Upload data and source code to open.canada.ca

The final step is to post it to the federal government’s open data website: open.canada.ca. The source code is also posted for other organizations that may wish to learn from the development work.

The finished product is a powerful new tool that enables Canadians to take part in informed discussions about the management of the country’s energy.

"Transparency” has become a buzzword for governments seeking to engage citizens more fully. But shining a light on information involves a lot more than just lifting the shades.

The National Energy Board (NEB) began its shift to transparency years ago by, for example, making its data on pipeline incidents open to the public. The big reveal had limited impact, however, because few of us are experts in sorting through massive Excel spreadsheets.

More was needed so that any Canadian citizen could – and would – digest the information, find the stories and engage in an informed discussion about the 73,000 kilometres of interprovincial and international pipelines the NEB regulates. How could the NEB make it easier for anyone to discover the stories hidden in the numbers?

That question led to a unique collaboration between the NEB, an information visualization research laboratory at the University of Calgary, and VizworX, a customized software developer. Together, they have developed a way to turn the NEB’s 2008 to 2017 data on pipeline incidents into visuals that are easy to create, understand and communicate.

Consultant Annette Hester was brought on board in 2015 and began a countrywide search for the best data visualization laboratory in Canada. After speaking with experts at leading institutions, like the Ontario College of Art and Design, in Toronto, and the University of British Columbia, in Vancouver, she was pointed to Dr. Sheelagh Carpendale at the Interactions lab within the University of Calgary.

The U of C lab was the only one large enough to handle the full spectrum of what the NEB was looking to achieve. The data visualization team at the Interactions Lab approach visualization research from both the design and computer science perspective. This dual perspective, coupled with the collaboration with VizworX has led to web-based data visualizations of NEB pipeline incident data.

It was a project unlike anything the Interactions lab or NEB had done before, so negotiations took months. But the collaborators agreed upon a shared goal: using research into data visualization to make the NEB’s information more understandable and accessible to members of the public.

The NEB believes data visualization helps everyone, including experts and non-experts, to better understand the raw numbers by making tools available to analyze large and complex data sets, and look for trends and patterns. Those goals dovetail with U of C labs’ research focus, which is on finding new ways for people to connect with, and better understand, data.

To test the waters, a small team undertook a pilot project, which became Exploring Canada’s Energy Future. The three-month pilot project tested the idea that visualizations could make data more understandable for everyone. The pilot was not only enthusiastically received, but it also earned the Contribution to Innovation in the Regulatory Field award from the Community of Federal Regulators, a partnership of federal departments and agencies.

Bolstered by the pilot’s success, the NEB and U of C labs undertook a full-scale project: one that would create visualizations of further NEB data, including incidents at NEB-regulated pipelines and facilities, advance research into the design behind the visualizations, and give NEB staff the tools they need to carry on once the research is done.

The three-year Data Visualization Initiative (DVI), which runs until fall 2019, will provide additional visualizations of data for other aspects of the NEB’s work. This experiment goes beyond just making data open; it contributes to the NEB’s mandate to encourage energy conversations by providing Canadians with unbiased, reliable and accessible information.

Katherine Murphy, DVI project manager, says the research being done will advance the use of visualizations for any interested organization. Making the code open source, she says, is “built right into the contracts.”

Sharing the research findings is a huge part of the project. Murphy says the team spends a lot of time talking to the data community in other government organizations who are watching the DVI project closely.

Dr. Sheelagh Carpendale, project co-lead, who heads the Innovation in Visualization research group, says the attraction for academics is the ground-breaking nature of the work. “We are a research lab,” she says. “We’re a very small group of people working on trying to answer the next big question.”

“There isn’t a template for what we’re doing,” adds Dr. Wesley Willett, project co-lead, who heads the Data Experience research group. “We’re acclimatizing the NEB to start thinking about their data in new ways.”

In the next phase of research, the U of C labs will work with various citizen groups, such as Indigenous communities, to learn more about what engagement with the visualization tool might mean. The feedback citizens provide will inevitably lead to refinements in the tool.

Resource: Energy visualization at the Interactions Lab

Five years ago, Bon Adriel Aseniero was sorting through a number of research papers for his master’s thesis in Computer Science at the University of Calgary, assessing how people were working with data, when he struck upon a novel idea.

“I had no idea what was cool or interesting in the papers,” he said. So, he devised a unique way to make sense of the data. He sorted the information visually, in interconnected columns to help him understand what he was looking at.

When the paper was published in 2014, he moved on from the clever tool he had created to sort the data.

Until 2017.

That’s when Dr. Sheelagh Carpendale, his thesis adviser, and director of the Innovation in Vizualization (InnoVis) lab at the U of C, remembered the novel way Bon chose to visually present the data. Dr. Carpendale was reminded of his idea while she was reviewing concepts prepared by Doris Kosminsky, Professor of Visual Communication at the School of Fine Arts at Federal University of Rio de Janeiro.

“I wasn’t aware I was building something totally new,” says Bon. “I was just trying to sort through the papers.” But, to Dr. Carpendale, Bon was building on “parallel coordinates,” visualizations pioneered decades ago by Alfred Inselberg, an American-Israeli computer scientist now based at Tel Aviv University.

“Doris showed me stacked bar graphs . . . and I thought we could take these one step further by including the ideas in Bon’s version of parallel coordinates,” says Dr. Carpendale. She approached him about joining a project to present the National Energy Board’s (NEB) data in an easy-to-digest visual form.

The NEB contracted with the U of C’s Interactions lab, seeking to find new ways to understand the mounds of data it has on pipeline incidents and related facilities it regulates in Canada. The incidents visualization tool is designed to make it easier for anyone, including non-experts, to spot trends by seeing the data presented in a visual form.

Now working on his PhD, Bon’s research is on how to use visualization for public engagement – exactly the purpose of the NEB project. The visualization tool, he says, allows people to generate their own stories. “I’m really into the idea of stories” from data, he says.

Bon did not begin his academic career thinking of data visualization. Born in the Philippines, he came to Calgary with his family in 2006. In high school, he was interested in computer science. His passion for computer games led him to learn computer coding. During his undergrad years, he met someone who was involved in the human interaction side of programming.

“I attended one lecture on data visualization, and that’s where it all clicked for me.”

A drawing hobbyist, he felt a personal connection to Dr. Carpendale, who also came from a visual arts background. “We understand the same things” about seeing data in a new way, he says.

Bon stresses the visualizations he created were just the start. They took on new life as team members added their own ideas to his concepts. He says the team helped him step back and critique – and therefore improve – his own work. “Sometimes you fall in love with your work too much,” he says.

Visualizations works best, he says, when it is they are aesthetically pleasing, and capture people’s curiosity. To him, anyone is capable of understanding sophisticated visualizations, even those that are much more complicated than the pie charts and bar graphs we are all familiar with.

To Bon, visualizations are a tremendous way to build public trust in the data from agencies, such as the NEB.

“We want to be able to build data provenance (validation),” he says. “We want to be able to show that this information is not biased.”

Now 28, Bon hopes to pursue his learning through teaching new media design once his PhD thesis is complete. Microsoft and Google are on his list of potential future employers.