EveryBlock and the Definition of News
January 25th, 2008 | Published in Data, Journalism | 10 Comments
So EveryBlock has launched to various reactions around the journalism community. Some of the common threads I hear in the CAR community come from guys like Mark Schaver and Ryan McNeill, who are both impressed with and unsure of this new thing.
It’s important to say that Adrian Holovaty, the founder of EveryBlock, is both a former colleague and a friend of mine. I’m sure there are folks in the CAR community who see me as a shill for his work. That’s fine; I’m proud of his successes. So you can take what I say in this post with that in mind.
Perhaps the criticism that has most resonance among CAR folks is the one that goes something like, “EveryBlock sure is an impressive technical feat, but it’s just data without context.” It’s rather a back-handed compliment, given our own efforts at providing data on the Web, but there’s some legitimacy there. It just troubles me when newspaper folks look at something built via a programmatic process as somehow not worthy of the title “journalism.”
To me, the main differences between EveryBlock and some of the efforts at presenting data that newspapers have developed are these: First, EB aggregates in a way that others don’t, both in terms of number and variety of sources. Second, unlike most newspaper products, it seems to be designed to let the consumer make the judgment of what’s news.
That last part is really important, because it isn’t something that newspapers want to give up - and for good reasons. There are plenty of reasons why people turn to newspapers: credibility, authority and a personal connection are among them. Many people feel very strongly about their news sources, and that’s a good thing. But there are other ways to get your news, and you would think that in these times we might, as an industry, be a bit more receptive to outside ideas.
If you live anywhere but a small town, then the question of “what’s going on in my neighborhood” can be an exceedingly tricky one to answer. Certainly most mid-size newspapers rarely have achieved blanket coverage of their communities, and that’s fine. The economics of doing so make it difficult, to say nothing of the time constraints. Data is not the answer to this, but it can be part of the answer. So to say that publishing data without context has no value, or little value, is a little hypocritical in that newspapers publish many stories of the “parachute in for an afternoon” variety. We try for context, yes, and many times we succeed. But let’s be honest with ourselves: are most of us giving our communities the kind of coverage we’d like to be able to every day?
Mark says that “there’s no evidence in the past, no evidence now, nor will there be any evidence in the future, that the way ahead for the news industry is to feed the world more raw data, however skillfully deployed.” Pretty definitive, and Mark’s a smart fellow. But something nags at me here, and it’s largely because of my work with Adrian on the Congress Votes Database at washingtonpost.com.
You could argue that it’s in between what Mark is calling “raw data” and data with context - certainly we thought a lot about what information to present and how to present it, but there’s very little that’s being done to that vote data as it comes in. The data, in this case, is its own context - it’s a feature of the app. And yet, I’d like to think that it is not only fairly popular but also a public service - precisely the sort of thing that newspapers should be doing. Could it be better, or have a narrative element more closely tied to it? Sure. But I would vigorously dispute the notion that it has not provided some evidence to the theory that providing data can be beneficial for both readers and news organizations (of course, I also am biased on the votes database, too). If providing data without context is a bad thing, then why do I see so many newspapers regurgitating home sales, crime reports and other events in their pages and on their sites?
I think Mark’s statement should have a few caveats, particularly that in some cases, raw data can be both interesting and useful, and that it doesn’t always require dozens of people to make it so. The real and emerging strength of EveryBlock, imho, is its utility - that users can make of it what they want or need. You’ll get no “EveryBlock is the savior of journalism” hyperbole from me. But it’s way too early, and says way too much about our industry, that we can dismiss it as journalistically unworthy, which is pretty much what “data without context” means.
Adrian says that the site isn’t meant to compete with existing news organizations but rather to supplant supplement their efforts. And one way I think EB will do that is by giving readers some of the powers that news organizations have always kept, starting with the power to decide what is and isn’t news.
January 25th, 2008 at 7:32 pm (#)
I agree 100%. I’m really impressed with EveryBlock. My co-worker and I had a similar idea about a year ago, so of course I thought Adrian was onto something when he announced what he planned to do with the Knight money.
Nothing happens in a vacuum on the Internet. All too often, we resist sharing data because it’s ours or we don’t want to do it without context. I think doing stories related to data one posts online is a good thing but it is not always necessary either. Context can be superfluous too.
Think of it like scientific research or academia in general. Often you share data because you want to see someone else’s take on it. You can do it to make sure the results of your study can be reproduced by someone else.
I think more openness, more data is a good thing. Democratize the data. Put it on the web. Let users search it, mash it up, find trends they might know to look for that you wouldn’t. It very well might create the kind of loyal dedicated users that newspapers think they have but probably don’t.
January 26th, 2008 at 9:19 pm (#)
[...] EveryBlock and the Definition of News It seems to be designed to let the consumer make the judgment of what’s news. (tags: everyblock news journalism web) [...]
January 27th, 2008 at 2:03 pm (#)
As another of the “is it journalism?” critics, lemme respond.
You say: “It just troubles me when newspaper folks look at something built via a programmatic process as somehow not worthy of the title “journalism.””
First, I’m not sure anyone is actually saying this. I’m not. I’m pretty sure Mark isn’t. There’s a big difference between saying *anything* built via a programmatic process is not worthy of the title of journalism, and *this specific thing* that was created via a programmatic process is not worthy of the title of journalism. God knows, I’d be a pretty enormous hypocrite is I believed the former since it’s what I do for a living. Luckily, I’m saying the latter.
So, having dispensed with that, let’s focus on the subject at hand.
What is EveryBlock? It’s an information aggregator, a data browser, and a very impressive one at that. But is it news? Is it journalism? I don’t think it is, and here’s why.
Look, I had eggs this morning. They were delicious. I could put that in a feed and publish it to the four corners of the world. But what’s the nutgraph? However damn delicious those eggs were (and I assure you they were delicious), the mere fact that I enjoyed every bite is not news. The act of reporting to the world “Aron Loved His Eggs!” is not journalism.
As editors are so fond of saying… so what? Why should anyone care?
Similarly, the fact that a sprinkler was fixed on 42nd street this week, as reported by EveryBlock, is not news. Now, it could be news if that building had been operating without a working sprinkler system for years, endangering lives and violating city code. It could be news if the wealthy, influential owner of this building used his connections at city hall to keep inspectors of his back. But EveryBlock does not tell me that, and no algorithm ever will.
Merely providing data without context is not journalism. What you do with that data to put it into context IS journalism. Which brings me to the next straw man argument…
You say: “If providing data without context is a bad thing, then why do I see so many newspapers regurgitating home sales, crime reports and other events in their pages and on their sites?”
Who in the world is saying providing data without context is a bad thing? It’s most certainly not journalism, but it isn’t a bad thing. If I believed that, again, I think I’d be a pretty big hypocrite. But lucky for me I don’t.
And those newspapers you’re talking about don’t just provide data without context. I’m pretty sure you’ll find plenty of reporting about the housing market, plenty of reporting about crime and other events in those pages AS WELL AS the raw data. We can debate the quality of the reporting, or depth of the data made available — but that’s a separate issue. Newspapers are providing both data and context (if you ask me, not enough of either… but like I said, that’s a separate issue).
Which leads me to a statement I completely agree with: “The real and emerging strength of EveryBlock, imho, is its utility - that users can make of it what they want or need.”
Exactly! That is the difference between journalism and raw data — people (not just professional reporters) can take this raw, unfiltered information and by placing it into context, by filtering out the Aron’s eggs from the slum lord down the street, commit an act of journalism.
So, my objection to EveryBlock really boils down to labeling. It is an amazing technological step that can enable journalism, but it alone is not journalism. That’s it. To label these information streams as “news” is just…wrong, however delicious the eggs are.
January 27th, 2008 at 7:00 pm (#)
Just to clarify, I read Mark as saying that providing data without context is a bad thing for newspapers. Or, at the very least, not helpful to them.
Maybe we’re talking about the act of journalism and the meaning of news, which are two separate things. I just have a feeling that the determination of what qualifies as journalism matters much more to journalists than it does to people interested in information in whatever form they decide. Nobody wins a Pulitzer for publishing election results, but they could for congressional coverage. That doesn’t mean that publishing the data alone can’t constitute news.
January 28th, 2008 at 1:32 pm (#)
Actually I think if Everyblock is reporting a sprinkler got fixed on a specific street in a specific neighborhood in a city, it WOULD be news.
Would it be news to everyone? No. Would it be news to the people on that block? Yeah, especially if it was flooding or causing problems for others.
We can go huge and reach for a worldwide audience with the net. Or we can go small and be like a thousand/million tiny newspapers.
So yeah maybe this is what Doig likes to call computer-assisted publishing. But It’s really good computer-assisted publishing.
January 28th, 2008 at 6:19 pm (#)
“News” is serving as a catch-all term in this argument, which I think is part of the problem. any of us, taking a random sample of a dozen or so of our own stories, can probably see the gaps.
Take Aron’s housing story argument. On one hand, that could mean a story that puts a face on the foreclosure issue, or that quantifies and identifies how ARMs are affecting a community. Conversely, it could mean a story like this: “Average home sales prices in Doe County fell for the first time in thirty years Tuesday. ‘Terrible. Just terrible,’ random expert Mr. XXXXX said. At the same time, the average number of homes sold grew in each of the surrounding counties, by a typical margin of XX percent.”
We’re probably all thinking of the first story when we get into the “context” argument. But the truth is story B has some value, too, and that value is largely based on — you guessed it — “raw data.”
Who cares what the expert said? Sometimes information makes more sense without us getting in and mucking it up.
January 29th, 2008 at 12:49 am (#)
I feel like the critic who reviewed a new play at the community theater, and because he didn’t just offer unqualified praise, is accused of being an enemy of theater itself.
As Aron points out, you’re knocking down a series of straw men.
“It just troubles me when newspaper folks look at something built via a programmatic process as somehow not worthy of the title ‘journalism.’”
Someone may have said that. It wasn’t me. Do you have a citation?
“But there are other ways to get your news, and you would think that in these times we might, as an industry, be a bit more receptive to outside ideas.”
I think it’s unfair to equate honest criticism with being unreceptive to outside ideas. My post was an attempt to answer the question: Here’s this beautiful thing — why doesn’t it interest me very much? Why am I unlikely to come back here very often?
“Data is not the answer to this, but it can be part of the answer. So to say that publishing data without context has no value, or little value, is a little hypocritical in that newspapers publish many stories of the ‘parachute in for an afternoon’ variety. ”
I don’t see what’s hypocritical about saying something is good, but just doesn’t move me that much. I don’t see why we should only be cheerleaders.
“If providing data without context is a bad thing, then why do I see so many newspapers regurgitating home sales, crime reports and other events in their pages and on their sites?”
I don’t think it’s a bad thing . Newspapers have always published agate, and there will always be some people who want to read agate, but I’m not one of them. My plea, essentially, is that if you’re going to do something like this, then tell me something interesting, something I didn’t know before, and don’t make it hard for me to find what’s of value. (Am I always able to do that in my own work? No, unfortunately.) The Congress Votes Database is a fine example of a tool that makes it easier to find valuable information and I’m glad it’s there, although personally, I’d rather have a top-notch reporter dig in there and pull out a Holy Shit, Hey Martha story for me.
“Mark’s statement should have a few caveats, particularly that in some cases, raw data can be both interesting and useful, and that it doesn’t always require dozens of people to make it so.”
What I wrote was, “I’m ostensibly a data person, but browsing raw data, while it can be worthwhile, isn’t nearly as compelling to me as the sudden, unexplained death of a 28-year-old movie star.” My point was that we are competing for people’s time and attention, and I personally don’t think browsable, searchable lists of public record trivia will be a particularly strong draw.
“But it’s way too early, and says way too much about our industry, that we can dismiss it as journalistically unworthy, which is pretty much what ‘data without context’ means.”
No, it doesn’t. It says that we are awash in data and that what I want is help sorting the significant from the insignificant. Good journalism does that, and if wanting more of it is a sin in the church of the future, then I’m a sinner. What I don’t want to see is a world where all of our energy is devoted to telling parents what the weather will be at their kid’s little league games.
Sorry for the slow reply, but I’ve been too busy the last few days to respond.
Next time I’ll be sure to include a nut graf and some bullet points so I’m not misunderstood.
January 31st, 2008 at 8:29 am (#)
Matt: I think you’re making my point for me. You’re right: A sprinkler could be news. But someone would have to know the context — that it was causing flooding for others. Everyblock, you’ll notice, does not tell me that. Everyblock gives equal weight to a sprinkler repair, an elevator outage, a mundane building permit issued and the tragic death of a movie star. Any or all can be news, but it requires someone to take that raw data, see the context (the nut graph… the “why this matters”) and commit an act of journalism.
Everyblock is a great tool. But it’s not news.
February 13th, 2008 at 8:09 pm (#)
But isn’t this a straw argument on it’s own? I mean the idea that a newspaper would write a story about a single sprinkler in one building getting fixed is ridiculous. There’s not enough people affected by it to matter.
That’s where Everyblock comes in. It’s news to that one block. it’s news to the people in that building. Everyblock doesn’t have to provide the context cause the reader itself is providing it.
I’m not saying Everyblock replaces news or good journalism. I’m just saying it provides a type of news that probably hasn’t been around since the pamphlet was the primary form of publication. It’s the equivalent of bulletin board posts at the local community pool. Or the HOA flyer.
Just cause a journalist and an editor don’t touch it doesn’t mean it’s not news to someone.
I think Everyblock is using geography to sort significant from insignificant.
I don’t think we have to spend all our energy to telling people what the weather will be at their kid’s little league games.
I also don’t think we should scoff at doing that either.
I don’t think the average person cares that we’re wanking on about whether that’s journalism or not. Most newspapers/news sites just aren’t USEFUL enough to readers. I don’t think we should abandon big idea/big context journalism. But I think if we don’t find a way to be useful on a small-scale as well as a big one, we’ll get our lunches eaten by the dude in the garage.
Craigslist, urbanspoon.com, rottentomatoes.com, political blogs (biased and unbiased)… these are all eating away at what people used to look to newspapers for. If all we have is the big idea then we’re reducing what we can be and making it harder to see anything but red at the bottom line.
February 14th, 2008 at 9:55 am (#)
[...] EveryBlock and the Definition of News. Derek Willis [...]