Tuesday, March 15, 2011

A Digital-First Dissemination Model

by Jeff Stanger

Data Gatherers themselves should use digital-native techniques in the first telling of the stories in their data. In addition to injecting subject matter expertise into the online discussion of complex issues, it will create the welcome byproduct of better, cleaner data to support subsequent interactive development.

In a post this summer, I explored the difference between "data" and "information," arguing that in our digital era, effective presentation in new media is what transforms the former into the latter. This theme, as it relates to government data, was discussed by Jeremiah Akin and Michael Castellon at South by Southwest Interactive, as reported by O'Reilly Media's Alex Howard.

Howard: The key here, emphasized Akin, is that is not just enough to simply dump data. You need ways to visualize it and make it meaningful as information. "There's a lot of resistance - people have been there, and that's not how they've done things," he said.

I certainly agree with the point that it is important to go beyond raw data to meaningful information. I wasn't at the SXSW session, but in citing "resistance" from "people" who "have been there," Akin highlights a critical gap: the one between Data Gatherers (i.e., researchers, subject matter experts) and Data Interfacers (i.e., interactive developers, technologists).

This gap presents an enormous obstacle to creating meaningful information for today's digital citizens. The Gatherer-Interfacer divide is marked by differences in each group's aptitudes, objectives, vocabulary and m.o. Data Gatherers value context and interpretation. They express those in familiar and controllable forms such as reports, articles and white papers. "Data," in the way developers think of data, are not a product. Data Interfacers on the other hand value the ability to create engaging interactive presentations that tap the new possibilities of digital media (Web and mobile applications, interactive information graphics and visualizations, etc.). To them, clean structured data are the thing. Simply put, researchers and developers are on separate paths to different end products.

Aligning these end points is the key to smoothing the flow of usable information to citizens and policymakers in a digital society. What's required of Data Gatherers is to step out of their role as report writers to become effective communicators of information in a digital setting. This means adopting a "digital-first" dissemination model that includes interactive presentations alongside traditional documents. What's required of Data Interfacers is to help them get there — not by demanding "open data" after the fact, not by complaining about data quality, and not by assuming that shiny interactive apps can replace all other narrative context — but by creating common data building blocks as an integrated part of the research and dissemination process.

The first diagram below shows the prevailing condition. The product of Data Gathering is a document. The underlying data are organized for internal analysis and write-up, not structured from the outset as the raw material of online presentations. Interactive development takes place apart from data gathering, after the fact, and most often completely outside the organization doing the gathering (e.g., data journalists interfacing government data).

Prevailing "Document-First" Dissemination Model

Diagram: Document-First Dissemination Model

The second diagram illustrates a "digital-first" model of research dissemination. Both Data Gatherers and Data Interfacers are headed toward a newly conceived package of native digital supplements — visualizations, information graphics, dynamic data-driven Web applications, and the like. Direct collaboration between researchers and interactive developers on the original presentation of the findings gives both groups what they value: researchers can inject their subject matter expertise; developers can use the power of modern digital presentation methods. In this model, Web-friendly structured data are a common building block between the Data Gatherer's initial reporting and subsequent interactive development. We need to promote this digital-first approach among Data Gatherers.

"Digital-First" Dissemination Model

Diagram: Digital-First Dissemination Model

Howard concludes:

In this narrative, it's up to governments to release better, clean data in consumable formats and the evolving art of data journalism to make stories from it that give citizens, businesses and elected officials insight into the complexities of modern life.

I agree with this summation and would extend it to all Data Gatherers, not just governments. I believe the clearest path to releasing better, cleaner data is by creating digital-native information in the first place. Data Gatherers (i.e., "governments" and others) shouldn't only see themselves as raw data providers, and digital journalists shouldn't alone carry the burden of telling stories from data. Data Gatherers themselves should use digital-native techniques in the first telling of the stories in their data. In addition to injecting subject matter expertise into the online discussion of complex issues, it will create a welcome byproduct — clean data leading to subsequent interactive development that extends and challenges the original interpretation.

Comments welcome...