In a comment, Ben asked how PolitiFact went from idea to PolitiFact.

How did that get refined into the site we see today? Was the content and feature refinement mostly the work of web people or people from the print newsroom? Any lessons learned to help the rest of us help our editors and reportorial colleges see the new dimensions web apps can bring to conventional content?

The content and feature refinement, at least at first, was the work of Bill Adair and I almost exclusively. After we first talked over the idea, I sketched out how I thought the database should be laid out based on Bill’s vision in Django. Arguably the best thing about Django is that out of the box it has a fantastic administration tool. The admin tool let me show Bill what I had done and let him enter in a few records into the database. Doing that, he’d have more ideas, or I’d think of something I wanted to put in, and I’d add them to the model code. After we got pretty happy with what we had, we split up: Bill started reporting out and writing a few items and I started building the views — Django’s query layer that takes data from the database and sends it to the templates for display. From this, both of us came up with new things we wanted and old things that didn’t work right. Out of all this work came the demo, which brought the decision to go forward, which brought on board our IT staff and a web designer, plus more input from both web and newsroom colleagues.

There’s a couple of lessons I see in our experience that may or may not work for you:

• Keep the number of people initially involved as absolutely small as possible.


• Ideally, the people involved at first should be the one with the vision and someone who can translate that vision.


• Ramp up staff slowly and minimally. More people means less decisiveness.


• Ruthlessly compartmentalize. Let designers design, reporters report, etc.


• Accept that you’ll never create the perfect app, so decide quickly what’s worth fixing, what’s not and what belongs in version 2.0.

And that brings me to the title of this post. One of the key parts of developing PolitiFact was the ability to translate a standard style of news story into something bigger, broader, relational. I’m not saying that a skilled programmer can’t do this, but in my experience, this is where the journalist part of the term “programmer/journalist” has the most value.

If you’ve been a reporter, hopefully you’ve been paying attention to how people react to your stories, what information readers remembered, how readers approached stories. Hopefully, you’ve paid attention to the weaknesses of how you’ve approached some stories in the past. Also, as a reporter, hopefully you’ve paid attention to the parts of your stories that repeat, over and over and over again. Names, places, even the structure of certain story types.

If you have paid attention, it becomes a matter of creativity to translate that reporting experience into a web application. You rely on your experience writing those stories to figure out what parts you need in the database — a byline field, a headline field, a blurb field, a body text field, etc. Having written these stories before, you know what parts repeat — candidate name, candidate party, candidate home state. Having talked to readers, you know some of them only want to know about the candidate they care about, so you set up one-stop-shopping pages for each candidate. You know some readers are single issue people, so you set up and issues database and pages for everything you have on a single issue. You know some readers are political party people, so you do the same things for party that you did for issues. Now, I’m using PolitiFact as the guide here, but you can apply this to lots of different stories, ideas, databases, applications.

The reason I think the journalist/programmer idea is so interesting is that it’s the journalist part who can put themselves into the position of reporter writing the content and reader consuming the content, and it’s the programmer part who can translate all that into a web application, taking content in one side and displaying it on the other side in as many ways as you think readers would want.

A lot of journalists I’ve talked to lately are all dazzled by the programming part of this equation. Don’t dismiss so quickly the importance of your journalism experience. If you’ve been paying attention to what you’ve been doing, that experience has prepared you well to create web apps.