Data and Beginnings

Q. How did the Faces of Death Row project get started?

When I came to the Tribune in January, I was immediately interested in Texas’ death row. I had moved to Austin from New Mexico, where there isn’t a death penalty… It was just such a culture shock for me. I wanted to bring the issue to the forefront and show who is on Texas’ death row and how long they’ve been there. I’ve always been interested in criminal justice beats, anyway, so I was immediately drawn to that stuff. I thought it was something that definitely should be talked about more.

Q. What was it like working with this data set? Where did the data come from?

This data came mostly from the Texas Department of Criminal Justice (TDCJ).

Originally I was trying to get cost data, how much it costs to house them and put them through the legal process. But TDCJ doesn’t keep data on how much it costs per day to keep a death row inmate. (They have information on how much it costs just to house the average inmate but it’s more for a death row inmate, who has more supervision and is in more isolation.) That was where I started. It came down to, “Ok, just get me all their basic information, and their mug shots.” We decided to list all of the inmates and include their mug shots because nothing like that exists right now.

The demographic information for the inmates was fairly easy to get. [TDCJ does] have a version of it online, it’s just not very easy to search, and sometimes it’s a little out of date. But getting mug shots from the department was tough. It ended up taking a couple of months to get in touch with the right people and get the data in an easy-to-use way. I ended up getting them from the commissary department. Ultimately, it cost about $250, but we had to talk them down from an original estimate of $1,500.

On top of their data, we gathered conviction summaries for each inmate manually—which was pretty time-consuming—by using information from TDCJ crime records, court documents and news articles.

Q. Any big surprises for you? What do you think will be most surprising for users?

I expected the number of women to be a low number but I didn’t expect it to be as low as it is [six women]. The percentage of black inmates was also surprising to me because it’s so different than the percentage of black residents in the state. Also, some of these people are on death row for 30-plus years… It’s kind of shocking. I don’t think it’s something that a lot of people think about.

Building It

Q. How did you arrive at the interface/narrative for this project?

I got the idea for a filterable list with profiles when I first saw NPR’s Book Concierge. My mind was so wrapped around this death row project, that when I was looking for a new book to read I immediately started thinking, “This would be a great way to show all the inmates.” We decided to try that, and our web designer Ben Hasson gave me some wonderful mockups. It’s been pretty smooth since then.

We chose these filters because the data was consistent, and they provided the most context. There are many interesting facts you can pull out from these filters that hopefully stir conversation.

Some examples:

• 11 inmates have been on death row for more than 30 years
• 42% of all the inmates on death row are black, while black residents make up 12% of the state population
• There are 6 women out of 261 inmates on death row

Q. What is Faces of Death Row made of, in technical terms?

The app was produced using the Tribune’s App Kit, which is built on Gulp, a task runner written in Node.js. We use Gulp to power our build system, which is configured to handle our deploy process. The data is pulled into the app from a Google Sheets spreadsheet, and JavaScript/jQuery and SCSS were used to build the front end.

Q. Can you tell us more about how tech and reporting are integrated at the Texas Tribune?

There are four of us on our team–we build interactives, we build apps, we work with data. If reporters get a data set that they don’t understand, we can help them analyze it. We’re right in the middle of the newsroom, talking to reporters all the time, coming up with our own story ideas. If we hear that a reporter’s working on something, we’ll go up to them and say, “Hey, I think we should do this with that idea.”

We have two different kits that we use, an app kit and a graphics kit. They both use Node.js with Gulp. The graphics kit is what we use when we’re making a chart, and just dumping it into our CMS. The app kit is when we do stuff outside the existing CMS. It builds a completely individual site for you as opposed to a small graphic that you dump into the larger site. It is pretty standard practice when we do a larger project to use the app kit. I’ve been here six months and this is the second, larger project that I’ve used the app kit for.

Q. Were there any challenges on this project related to the ethics of presenting the information, or the weight of the subject, that informed your design and coding work?

Obviously this is a sensitive topic, so yes, there were multiple conversations to make sure we were doing this right. The biggest ethical question for us was the crime summaries. Everything else was so straightforward (this is when this person got on death row, this is when they were convicted). … The first summaries we had were pulled from TDCJ; they were written by police and went into varying degrees of detail. I looked at them one day and thought, we need to be careful to make sure this is what they’re actually convicted of instead of just what the police say happened. We decided to solely list what they were convicted of, not quote from a police report that would sometimes go into a lot of detail about a crime, from a police perspective.

A Word of Advice

Q. Any final words of advice for news apps teams approaching similar projects?

It’s great to start with huge ideas. After exploring the available data and resources, you can narrow down the project to something more realistic. It was a little difficult to stay focused on one aspect of the entire death penalty subject…. I came here and was like, “I want to get all this information on the cost, the demographics,” I wanted everything. We realized it was going to be overwhelmingly easier and better for the reader to separate these things out. We’ll see how it’s used, what people are doing with it, and then go from there to decide what we want to add on or do for future projects. If people use this tool to start a conversation, anything on this subject, that’s worth it. We wanted to provide information and let people come to their own conclusions. I am really excited to do follow-ups in the near future.