Sergey Nivens - Fotolia

Doors opening for open source data visualization tools

Open source data visualization technologies have matured to the point where users say the available tools can handle large amounts of their visualization workloads.

Commercial tools currently dominate the data visualization software market, but, as in other realms of data management and analytics, open source technologies are starting to catch on. They include the D3.js visualization library and Leaflet mapping library, plus R, a widely used programming language for statistical analysis that has been augmented with interactive visualization features.

One of the criticisms aimed at open source data visualization tools is that they require too much coding knowledge and specialized training to make them work. But as their development continues, they're beginning to look and feel more like polished visualization products, according to users who are familiar with them. And in the case of R, multiple software vendors are offering commercial products for enterprise use.

"I think the reason you're seeing these tools become popular is they're available and they're mature. You can get a lot of work done before you hit the boundaries that the open source tools have," said Paul Bradley, chief data scientist at ZirMed Inc., a vendor based in Louisville, Ky., of software as a service applications for healthcare administration. Bradley is a proponent of R, saying that vendor RStudio Inc.'s namesake graphical user interface for the analytics programming language has become much more user-friendly and less coding-intensive.

[D3 graphics let users] dedicate resources to data exploration and data science rather than trying to create a fancy chart.
Charles Whittakerhead of BI, Avant Inc.

"You can go a very long way with just R," agreed Clement Brunet, director of research and analytics at The Co-operators Ltd., an insurance company based in Guelph, Ontario. Speaking at TDWI's 2015 conference in Boston, Brunet said R is particularly good for taking pilot analytics projects through the proof-of-concept stage and then scaling them up to production use. And that's only getting easier as the R environment becomes more graphical in nature, he added.

Online lending company Avant Inc. uses a combination of commercial business intelligence software and open source data visualization technology. Charles Whittaker, head of BI at Chicago-based Avant, said he's generally interested in anything that enables him to spend less time visualizing data and more time analyzing it. The prebuilt graphics of the D3 library are a perfect fit for Avant's needs, he said, adding that they let users like him "dedicate resources to data exploration and data science rather than trying to create a fancy chart."

For simplicity's sake, Stephen McDaniel, co-founder and chief data scientist at consulting services provider PowerTrip Analytics, recommends open source visualization tools that skip data discovery functionality and focus exclusively on creating charts and graphs. One tool he favors is Lyra, an open source technology being developed at the University of Washington that McDaniel described as "the Photoshop of data visualization."

Lyra takes the D3 framework, a collection of JavaScript documents which on its own requires users to do some manual coding, and automates everything through a point-and-click interface. And because the output format utilizes common Web languages like HTML, SVG or CSS , it's easy to share data visualizations throughout an organization or publish them to the Web, McDaniel said. "It's like the missing piece," he added. "It's not iterative questioning of data -- it's about building a specific graph and publishing to Web servers, merging in content and serving it up to the world."

Next Steps

Keep the focus on business value with visualization tools

Visualization tools spot the value in big data

Data scientists are all about the open source tools

Dig Deeper on Business intelligence dashboards and performance scorecards