PolitiFact Pulitzer Validates Journalism-Technology Collaborations

If the survival of journalism depends on technology innovation, one or more of three things will have to happen:

1. Journalists will learn technology development;
2. Technology developers will learn journalism;
3. Journalists and technology professionals will learn to collaborate.

The Pulitzer Prize awarded last week to the St. Petersburg Times for PolitiFact, a database-powered website assessing the truth of political statements, is proof that journalists can learn computer programming. The idea behind PolitiFact came from Times reporter Bill Adair; the database and software development under the hood was built by reporter-turned-developer Matt Waite, whose job title is news technologist.

The Knight News Challenge grant that financed scholarships for technology developers to study journalism at the Medill School at Northwestern University was a bet that programmers can learn journalism. At Medill, the first two scholarship winners helped develop News Mixer, a widely praised web application demonstrating new ideas for enhancing conversations around news.

The third option — building communication and collaboration between journalists and technologists — is being tested during the current academic term at Medill. The third Knight "programmer-journalist" scholarship winner, Nick Allen, is part of the newest experiment, in which a Medill class is collaborating with a concurrent class of computer science students from Northwestern’s McCormick School of Engineering.

Before I tell you more about this new educational experiment, I’d like to celebrate PolitiFact. After all, who would have thought a few years ago that it would be possible to win a Pulitzer — journalism’s most prestigious prize — for software development?

PolitiFact, of course, is software and a whole lot more. The core of the idea is not new: for years, news organizations have included regular features that assess the veracity of statements by political candidates. The St. Petersburg Times team that developed PolitiFact simply turned that idea into a database, backed it up with thorough reporting, wrapped it all together with an appealing design, and added a dash of entertainment (labeling the most egregious political fabrications as "pants on fire.")

Aron Pilhofer, who leads the New York Times technology team that produces software-backed journalism projects, nicely summarized the significance of the Pulitzer award as a "Color of Money" moment. He was referring to the Atlanta Journal-Constitution’s groundbreaking 1988 investigative series written by Bill Dedman and documenting systematic discrimination against minorities in real estate lending. The project, and the prize, validated a growing movement among journalists to learn analytical techniques — and software applications — to find meaning in data in order to produce better journalism. In the years that followed, thousands of journalists learned these techniques, which journalists labeled "computer-assisted reporting" or "precision journalism."

In 1989, when the project won a Pulitzer, computer-assisted reporting was a novelty. By 2009, major investigative projects that didn’t include data analysis were the exception rather than the rule. And the data analysis techniques used for that project had become widely used in day-to-day reporting as well.

Twenty years from now, I hope we’ll see thousands of journalists developing online software applications that inform, engage and enlighten the way PolitiFact does. The question is how we’ll get there.

This term’s Medill "innovation project" — directed by my colleague Jeremy Gilbert and me — is trying to figure that out. We are collaborating with Kris Hammond and Larry Birnbaum, who run Northwestern’s Intelligent Information Laboratory. For years now, they and their colleagues have focused on projects that deliver "frictionless information" — that give people exactly the information they want, when and where they want it.

This quarter, they are running an Engineering School class seeking to apply their approaches to challenges and opportunities facing journalism and media. Meanwhile, Jeremy and I are directing a class of Medill master’s students exploring new ways of bringing journalism and technology closer together. You can follow the journalism students’ progress on their class blog, writeclick.org.

Student teams made up of journalism students and technology students are working on five discrete projects:

• A "smart editor" that helps journalists research and check facts as they are working on a story;
• An application that generates personalized sports articles from baseball box scores;
• A tool that would allow a person to request a personalized news feed based on exactly the amount of time he or she has to consume news;
• Two projects that leverage Twitter — one to serve "tweets" relevant to an article you are reading, the other to deliver relevant news based on a user’s "tweets."

These projects are experimental and exploratory. Whatever their direct results, I’m hoping that this combined class will turn out a small cadre of journalists who understand technology and the process of technology development, and a comparable group of programmer-developers who understand journalism and media.