A time bomb: why any news site without a relational back-end will eventually fail
So I was having a conversation with Brett Roegiers the other day, and he mentioned a blog post by Derek Willis on what he calls “Bomb Throwing” in the newsroom. In a general way, he talks about the new wave of online journalists dropping bombs on their colleagues in the form of Nelson from the Simpsons: “Haw haw, you’re medium is dying!” So long and good luck. Don’t worry, you can always read about what’s going on in our newsroom blog. Willis’ blog post is more specifically targeted, but I’m not going to get into that.
He has a point, though — you can’t just tell everyone that their model is wrong and flawed, and your way trumps all they know. Not only that, but a lot of the proposed ways to “fix journalism” I’ve heard don’t seem like they will stand the test of time. Live-twittering events is all well and good, but only a certain subset of the population will understand or even seek such a thing out. Pretty flash graphics, maps, videos and slideshows are great, but they’re only things that aid in reporting and understanding. They’re not enough by themselves. They’re like the toppings on your sundae.
I’ve seen a lot of preaching to the choir about online journalism, but much less real-world action in trying to get the people most involved with news organizations to understand the core issues. If nothing else, we need to remember that even though we’re nerds, the people who will most likely use the site probably aren’t.
And it really does start with re-working a news website from the bottom up. If you build with the idea of relational information at your core, you’ve transformed yourself from a daily story factory to an information powerhouse. This should include, but certainly not be limited to, tagging names and places inside of stories so that they can be tied together.
That’s what people really go to the paper for: information. Yes, the features and profiles are interesting too, but at its heart a newspaper is an information organization. There’s no need to define it by its medium, and there’s no reason to stifle ourselves as journalists — purveyors and gatekeepers of information — to the limitations of print.
People have the idea of news ingrained in their skulls as text blob stories with the occasional graphic to spruce the place up. When really, the same parts that went into that story could be re-synthesized for the Web to become a much more in-depth breakdown. What’s more, people are so fed up with “media bias” — well why don’t we remove a layer of translation and post our primary sources, interviews, etc. online? I’m all for transparency.
What I’m saying is, we have to do more than add cool APIs and tools to our journalism repertoire. We have to fundamentally re-think the news website and how we use the mountains of information we collect. Right now, a lot of good information is rotting on hard drives. Even at the Alligator, where I work as the managing online editor, we’re as much to blame as our CMS in how we use our resources. I’m trying, but as I’m sure anyone in my position knows, it’s not an easy fight.
Sphere: Related Content
May 28th, 2008 at 2:04 am
I was thinking this conversation would make a good blog post. Well done.
May 28th, 2008 at 2:20 am
Right on! It’s not middle-aged reporters who don’t care about YouTube and Twitter that’s the real big long-term problem.
“Information powerhouses” are exactly what we must become. Sadly, I don’t see owners out there yet with the foresight to invest in that new model, to trust it, to take a risk.
But if us nerds don’t remember the big picture, who else is there to whisper in King Lear’s ear?
May 28th, 2008 at 4:45 pm
I’m buying what you are selling. But I think another problem is a cavalier attitude toward throwing everything to see what sticks … with no measures for success except if other online jou folks think it works. Obviously I’m biased toward scientific research and statistical modeling.
You’ve probably seen the rumors of forthcoming NYTimes API. I love the democratization of data, but I have big concerns as we strip context out of news. Reporting is job #1, but journalists do try to provide context as well (perhaps mostly fail). Anyone can run data and create pretty charts. That data can be totally misleading. I hope that in this rush to data-fy the news, teh big papers (and hence influencing the smaller ones) don’t forget to hire or at least consult statisticians.
May 28th, 2008 at 7:29 pm
Well, since WordPress won’t let me comment on my post (wtf, wordpress?) I’ve hijacked a test comment by Brett to further the discussion a little…
@Brett - Yep. I realized I don’t post very often, yet I write enough to other people to the point where I should just start *repurposing my data!*
@Kevin - Right. I can’t stand when people say that it’s the old people who “don’t understand X-web-service” that are holding us back. We just haven’t been pitching solutions that fit the core issue. It’s not their fault they don’t understand, because I think we’ve been largely deluding ourselves into thinking we understand.
The web services are cool, and what they teach us about how conversations and interactivity evolve on the web is invaluable. But part of the draw to newspapers is the ease of use. Besides reading and page-turning, no special skills go into obtaining the information therein. We need to make information *that* accessible to our online readers as well. Not only should the info be accessible, but highly relevant. I know that this is a complex task, though.
@Dave - That’s very true. I can’t deny that I’ve been tempted to throw out a bunch of stuff to see what becomes popular in my Analytics report, but a well-reasoned approach will beat spray-and-pray any day. I learned that in Counter-Strike. I know this is a bit more nuanced than a game, but the same theory generally applies.
I agree with your point about statisticians as well. Hell, a lot of journalism should be fact-checked by statisticians anyhow.
In any case, context is something I would never want to take away from the news. In fact, the system should be very context-aware, most likely by combining human inputs (in-story tags, etc) with what the system knows (links forged within the database, acrobatic algorithms, whathaveyou).
It’s do-able. And in my next post, I’ll take you through what I’m talking about here.
May 30th, 2008 at 6:13 pm
On “tagging names and places inside of stories so that they can be tied together”: If you haven’t already, have a look at OpenCalais.
June 3rd, 2008 at 10:25 pm
[...] by the conversation occurring in my last post, and spurred by some video I don’t feel like editing, I’m going to flesh out my theory [...]