status

location
Washington, DC
Subscribe to GeoRSS Subscribe to KML


WIAD DC Data Visualization panel

Published in Conference


WIAD DCThis weekend I participated in a fun panel on Data Visualization as part of World Information Architecture Day in DC. The moderator was Sean Gonzalez from Data Community DC and included from Amy Cesal from Sunlight Foundation, and Maureen Linke and Brian Price from USA Today / Gannett Digital.

You can see the video here.

There was a clearly interesting gap between our perspectives as storytellers versus tool creators. Data journalists such as Maureen and Brian focus on a story and develop or use tools that focus on that story. From one theme to the next, they reuse these tools but each output is a uniquely crafted experience in order best convey a story.

By contrast, I focus on building platforms and tools that enable anyone to develop their own story. In order to do this, I need to think about the generalization of data management and visualization capabilities to adapt to a wide range of applications. The tools need to permit customization without requiring indoctrination such that the storyteller can focus on their goal without the tool getting in the way.

Importance of Data

Common across all of our discipline, and the point that was most reiterated, was the vitality of data. At first finding quality, authoritative data, and subsequently the effort to clean it up, validate, normalize, analyze, and finally portray. The best visualization is useless if the data are suspect.

Fortunately, finding data is getting easier. Driven by open data initiatives and supported through specific and growing open data catalogs from the source means that there is reduced effort discovering relevant information to use for your visualizations.

Imperative to proper journalism, and the web, is the requirement to cite your source. Even more, linking to the source data and authority means that users can trackback to the raw data and create or validate their own findings.

Evolution of Medium

Along the concept of web links a few of the audience asked about when which type of visualization was appropriate. When is a static image sufficient and when should you use a complex interactive visualization?

Our discussion explored the idea of responsive visualization where it is important to understand the reader’s medium and situation of consumption: mobile phone on the metro, laptop in the office, or a mix of both? Personally I tend to find interesting articles on my phone and bookmark them for viewing later in full resolution on my laptop or iPad.

By developing responsive visualization, a story can provide a fast and sufficient static image on a mobile device while growing into a deeper visualization on a computer.

Resources

We shared many specific resources through the discussion. A few of the highlights include Nathan Yau’s FlowingData, replete with examples, critique and tutorials. O’Reilly has a number of books on data visualization and data science that walk through detailed methodologies and examples. Journalists should check out NICAR.

Similar Posts

Comments are closed.