When we last met, dear reader, I *may* have gone into a frighteningly long rant on the state of news apps. It’s a conversation that came up again this Friday on Twitter, and it just reminded me how lucky I am (I’ll say it again and again) to be in my current environment.
But I can do more than write angsty blog posts and tweets! In fact, people even pay me to make such Web projects! Isn’t that coincidental? (Not really.)
We’re trying to put some of these thoughts into practice at PBS. And as succesful (or not) as it is, we’re not doing it the way we’re doing it just because that’s what others are doing.
I’m breaking away from calling the pieces “data apps” or “data visualizations.” Visualizations can just be pretty representations of numbers that are hard to interpret. And while I like the name apps for some projects, it still sounds like a newsy tool, some sort of supplement. Or it gets confused with mobile apps.
The old me wouldn’t care what you call it, but I have learned that names matter, especially when creating a data mentality in a new culture. It frames how you do your work, and how others perceive your work. One of many things I’ve learned already in the first four months on the job. (FOUR?)
So we’re calling this “cool stuff I do” — DataStories, a name that emerged from me frustratedly and excitedly saying to my editor one day “Why can’t people recognize that what we’re trying to do here isn’t some side piece, and being pretty isn’t enough? It’s about telling stories!” Him: “So, these data stories…”
And now we have a better name than “What Keeps Michelle Up At Night”. That’s good.
Read our official explanation of this concept here on the PBS NewsBlog. It’s something that’s possible because of a remarkably open attitude at our new newsroom. If you pointed out that said attitude might come from the fact that the person at our helm is the president of the Online News Association, you would have a valid point. It’s not just Christine Montgomery though. It’s a culture that’s both already instilled by who happens to be a part of this team, and it’s also something we’re continuing to develop.
But as an extra bonus to my loyal readers, who put up with my journal of exploration/promises of how-to posts that I occasionally write/rants of frustration about the status quo, here are 13 main issues I’m trying to address with DataStories. (What, you wanted 10? If you read this blog, you should know being concise isn’t my forte.)
This isn’t some class project — this is something we’re working toward at a real, live news organization. We don’t have all the answers, but hopefully some of my key objectives are close to on target. And as with all posts on this blog, these are the thoughts of Michelle, not of PBS itself.
- Create pieces that stand on their own. DataStories are not sidecars to “the main piece” — they are their own piece. Context you need is included in intro text, or within the actual experience.
- Create pieces that teach you something. The point of interactive is to engage with what interests you. That’s got to go deeper than — here’s a lot of data, customize it by choosing your state. The numbers should help you understand something about your world.
- Give context. Context is not defined by making sure rows in a table are spaced out well enough, or going back extra-far in terms of numbers of years available. Relate the data back to events in the actual world, show how a big number compares to another big number. If this stuff were easy to understand and we could just throw it out there, there’d be less of a reason for data journalists. How are we not just making data accessible, but helping people make sense of it? Those are distinct tasks — are we addressing both?
- Have no expiration date. Don’t tie pieces to a specific event, that become irrelevant. Things stay on the Web for a long time. That means pick topics that we think will matter for at least months to come, if not years. And create the technical ability for pieces to grab data for months and years in the future, or commit to adding that data ourselves. It’s irresponsible to write articles without ever following up, so that’s not okay for interactives either.
- Make it relevant to multiple markets. Use data that is interesting to people throughout the nation (and the world, if possible), but make it feasible for people to drill in when applicable. And besides allowing the end user to drill down to their geography, local stations should be able to set the default at a certain level of detail. I don’t need a US map on the site of Chicago’s PBS station.
- Make sure the form fits the data. Don’t use maps for everything because they look pretty. Don’t use a table for everything because it’s impressive. Not every story can be told through data, not every DataStory fits a given form.
- Make it reusable. But while not every story fits one form, one form could fit multiple stories. We don’t always need to start from scratch. And these reusable forms should be able to be used by others outside our orgs, so more people can use data to get context they need.
- Help others, but not just other coders. We do a lot of sharing in the data apps community, and talk about the value of open source. That’s something I’m strongly advocating for, and we’ll keep you posted. But I’ve always said it shouldn’t be so hard. Let’s make tools that enable people who don’t know programming to be able to bring this info to their communities. Maybe it’ll pique someone’s interest enough to learn to code, so they have more flexibility in what they create. After all, program or be programmed. But in the meantime, let’s build tools to help people use the skills they already have, whatever those skills may be. We’ve got to lower the bar of entry on this, lest our work be considered esoteric and easily dismissed. That’s not okay.
- Be more visual. There’s a reason data visualization exists. Lists are good, as are tables, but to draw people in, we need data apps to be accessible to the masses as well as the geeks. Nothing wrong with using pretty colors if they’re functional, correct and don’t take away from the info. A DataStory is both an app and a visualization. Often times, those disciplines are still disparate.
- Structure our information. Comments about visualization don’t take away from the need to organize our info, allow for deep navigation and letting people search for what they want. Help them manipulate data. Suggest interesting components they might want to look at, searches to perform because of its news value, not solely their geography.
- Treat DataStories on equal footing with other pieces of content. Again, we’re not talking about supplements. Data folk need to push for more than being content with a box on the side of the main story saying “DATABASE: Search for YOUR *whatever*.
- Gain respect for our craft, inside the newsroom and out. Refuse to exist as merely a service organization, use our journalistic power to suggest ideas. And this practice of uncredited data pieces has to GO. This isn’t a 200 word news brief, so small that doesn’t merit a byline. And even a credit line isn’t good enough. That’s where you put people’s names who helped with a story, but didn’t bear the brunt of the work. They made, like, one phone call. And the real credit goes to the person with the byline on top. How is finding, organizing, analyzing and presenting data like making one phone call? I know I’m new-ish, but why don’t we get the real credit again?
- Speak up! Recognize that whatever effort we put into our work is often a perspective that’s rare in the newsroom. Don’t use that as a reason to shut up, but to speak up. No one else will do it for you, and no one will complain if you just let life happen to you. But we’ve got to keep talking about it and trying new things, if we want to make the most of our journalism. Letting the status quo win, that’s why we’re here in 2011 with a long way to go.
I, for one, will be doing more of that talking at NICAR in less than two weeks. I want to talk with YOU about these ideas, and how we can make them better. Together, we’ll make it happen.