Like this story? Get daily updates and alerts of big news events. Enter your e-mail address:
Adrian Holovaty's eagerly anticipated new Web service, EveryBlock.com launched yesterday in three major cities, Chicago, New York and San Francisco. The service seeks to answer the seemingly simple question "What's happening in my neighborhood?" by aggregating three flavors of local information: news articles from metro papers and blogs, "fun from across the web" from sites like Craigslist and Yelp and (most significantly) information from several government databases, some of which was not previously available online in machine-readable form (e.g. NYC crime data was previously only available in PDF form).
Why should I care?
Holovaty and Miner's ChicagoCrime.org was a monster hit, named one of 2005's best ideas by the New York Times.
The site's as close to a "rock star" launch as you'll see in the online news world. Holovaty and EveryBlock designer Wilson Miner are both alumni of the Rob Curley era at Lawrence.com, where Holovaty co-created Django, a popular open-source Python development framework. In 2005, they created the widely-praised Google Maps mash-up ChicagoCrime.org. I don't think I've been in a conversation about online news in the last year without someone mentioning Holovaty and asking what I thought EveryBlock was going to look like. Scoring a two-year, $1 million dollar grant from the Knight News Challenge sure raises awareness.
Basically, if you run an online news organization today, you're trying to hire "someone like Adrian Holovaty." I've heard those exact words from many an editor. The industry's been waiting with baited breath, hoping he'll crack the code and create something that'll save them. So how'd they do?
What's on EveryBlock?

In addition to all the already available data that Holovaty and developer Paul Smith were able to scrape into the site, the project's "people person," Daniel X. O'Neil had the not inconsiderable task of working with city governments to free up new datasets for the service to aggregate. He seems to have met with varying degrees of cooperation. Here's the data type breakdown on EveryBlock:
- building permits (sfc|nyc)
- business reviews from Yelp (sfc|nyc|chi)
- crime reports (sfc|nyc|chi)
- lost and found postings from Craigslist (sfc|nyc|chi)
- missed connection postings from Craigslist (sfc|nyc|chi)
- news articles from local papers, large and small (sfc|nyc|chi)
- photos from Flickr (sfc|nyc|chi)
- Liquor license info (sfc|nyc|chi)
- restaurant inspections (sfc|nyc|chi)
- San Francisco only
- New York City only
- building violations issued and resolved (New York City Department of Buildings)
- graffitti that's been cleaned up or is being cleaned (Community Assistance Unit)
- Chicago only
- buisiness licenses (Department of Business and Licensing Affairs)
- city press releases (As if Chicago would miss a chance to put out it's spin)
- filming (!) - (Chicago Film Office - entries are helpfully broken down by production type: film, tv, event, commercial)
- street closures (Chicago Department of Transportation, unsurprisingly there were more than 3,900 block party closings in 2007)
- buisiness licenses (Department of Business and Licensing Affairs)
So basically, there's a ton of information on EveryBlock--numerous custom data types, all tagged with geographic metadata and searchable by address as well as city specific human-readable formats. In Chicago, you can browse or search by zip code, ward or neighborhood, for example.
The amount of information available at launch is highly impressive. And the EveryBlock team says it's only just getting started with content types.
How's it look?
EveryBlock Chicago's content type icons.
Miner, an ex-Apple designer, has created a clean, elegant site with EveryBlock. It's as easy to browse as Craigslist (I don't find the interface "arcane" at all), but looks a heck of a lot better, without resorting to "Web 2.0" gradients and drop shadows. Content types have sleek gray icons (Update: Designed by Nathan Borror)
I'm a big fan of the crime report image, a dead on "Chester the Molester" likeness if I've ever seen one.
What surprised and delighted me most, however, were the site's maps. I think just about everyone had been expecting them to run Google maps on EveryBlock. I have no idea what they're using to generate them, but they're sure a lot easier on the eyes. To illustrate, here's a view of the latest "news" (EveryBlock uses the term loosely, in the Zuckerbergian sense if you will, to describe all content on the site) from the block where I live, in Chicago's Lincoln Park neighborhood.

I was surprised to see Google Adsense running on EveryBlock. Having scored the monster grant from the Knight Foundation, Holovaty has repeatedly stated that EveryBlock is a non-commercial project, built for good, not for gain (not that running Adsense is incompatible with that direction). Nevertheless, Miner pulls it off, incorporating AdSore skyscrapers into the design without noticeable disruption.
What's the point?
So the site's got information that's never been this accessible before and it looks great. But how useful is any of this (journalism buzzword alert!) hyperlocal information? It's neat to see all the filmings...but is it necessary? Why would I read the Craigslist postings on EveryBlock instead of just heading to Craigslist? The Techcrunch crowd posited several of these questions. Holovaty's answer? "It's all about serendipity." We shall see.
Outside.in is a competitor in the hyperlocal aggregation space (though one of their investors sees it as a collaboration). It's been aggregating local blog posts for some time now...and I've never seen a compelling use for their service. The site doesn't differentiate enough between good postings and the vast majority of local blog posts (which are derivative and unreadable). Like Outside.In, EveryBlock doesn't appear to exert editorial control over the news it displays. Posts are either geo-relevant or geo-irrelevant. If you're looking for "top stories near me," I'm not sure you'll find what you're looking for on EveryBlock.
I do think, however, that EveryBlock's launch marks a big step forward for "citizen journalism." EveryBlock's aggregation of public records will make it easier for concerned citizens looking to dig up dirt on local businesses. I predict we'll see a number of EveryBlock-driven consumer info blogs cropping up the next year (we'll also see blogs that regurgitate Chicago press releases). By putting all this government data out there and making it useable and readable via RSS, Holovaty and friends have lowered the barrier to entry into previously hard-to-enter genres of journalism. Blogger lets anyone publish a blog. Now, with EveryBlock, anyone can publish something worth reading.
Still Missing: Stronger Incentives, Better Quality Filters
As I looked through the site, I wondered for a second if "the key to all mythologies" had been created. Was there much else to be done in hyperlocal news? Of course there is.
EveryBlock puts important information into citizen's hands.
But citizen journalists are still lacking decent incentives and need better quality filters.
If you want to blog about local restaurants that fail inspection you can find some readers via national aggregators, local blogs and sites like EveryBlock and outside.in. But will you be able to earn a living? Warm, do-gooder feelings only go so far. At the end of the day, self-publishers will need ways to earn a living wage off their blogs if citizen journalism's ever going to take off. So we need new monetization tools.
I've been a faithful AdSense publisher for a few years now. It's still the best open advertising program on the Web. But it's never going to perform well enough for a local blogger. Conversion rates have plummeted as visitors have learned to spot Adsores and payouts have plummeted since the advent of Google's Smart Pricing initiative (which links payout to the quality of your site, disfavoring small sites with loyal, local readers).
We need new monetization tools that will convert readers into dollars for publishers who go local. (I'm working on some ideas in this area for the Methods Reporter. If you'd like to invest some money in what we're doing, let me know!)
And we need new editorial mechanisms, whether it be a team or editors or a wisdom of the crowds approach. There's got to be a way to push the good stuff, the stuff that matters, to the top while filtering out the chaff.
I've spent hours poring over the Chicago Blogmap (our city's best blog directory). While there are hundreds of blogs in it, for every great one like CTATattler or Gapers Block, there are 50 that haven't been updated in a year and consist of little more than text stolen from People.com. It's a crime to give blogs like that as much weight as Gapers Block, which has a team of sharp editors posting to it daily.
So EveryBlock takes us a few more steps closer to organizing and optimizing local information. but we're not there yet. Let's see who cooks up the other pieces.
What people are saying about EveryBlock online:
TechCrunch - Lost Remote: "Very cool" - Laughing Squid - Venture Beat - Journalistopia - Web Metrics Guru - ebiquity - David Dalka: "The likely future of journalism" - Curbed - Outside.In VC - Real Estate Agent - "the dataspine of a local infomation kiosk"
What do you think about EveryBlock.com? Did Holovaty and friends "crack the code"? Will it fuel a local blog boom?

Comments
[...] a new local news site for Chicago that launched yesterday over on the Methods Blog. Read that today. It’s so easy to call something a “game-changer” but this site could really lead [...]
From my experience poking around for city info, one of the big problems with data-scraping is that there's a ton of information out there that the city has to put out there but that they don't have much incentive to do so, and it's hard to find. That's going to be a big limiter for a site like EveryBlock, but the outstanding infrastructure combined with straight laborious data-entry could yield even greater results.
The most problematic part of the site is the news feed; I find it hard to separate the interesting stuff from the chaff, in part because papers still write their headlines/subheds/rubrics for print.
whetstone, that's a great point about headlines being written for print. Here on the Methods Reporter, we had a discussion the other day about our headlines. I'm pushing for editors to include at least one headline...while others want the sort of short, punchy, punny headlines that lead to click-thru's. It might be a while before people figure out that the Long Tail of search traffic will eventually overpower all but the largest live audiences.
What sort of data-entry combo are you thinking of?
Thanks for all the kind words. You're absolutely right that there's a lot still to be done on this front, which is why EveryBlock is such an exciting project to be working on. I'm of course very proud of what we've done, but I'm way more excited about all the stuff that's left to do.
Just wanted to mention something about the icons for different content types on EveryBlock, which you mentioned. Those were actually done by Nathan Borror, a friend of mine and a designer at the Lawrence Journal-World where Adrian and I used to work in Lawrence. He did an amazing job and I'm really happy with how they look in the site.
While I agree that EveryBlock is somewhat hampered by the lack of local citizen journalists writing about their neighborhoods, I think it will act as a catalyst to help create them. The site is a boon to mainstream and independent media as well as independent bloggers as a resource for information on our surroundings. For instance:
- The next time GB posts about an assault, we can cite crime statistics for the surrounding area from Everyblock (granted, we could have done that with ChicagoCrime).
- Food bloggers will be able to track new bar and restaurant openings by watching liquor license applications.
- One could look at consolidated news and blog coverage of the Montrose sinkhole, along with photos on flickr, and develop a metastory based on it.
I know that we at Gapers Block have a couple ideas for stories that would be difficult to do without a service like EveryBlock. I also know that they've got some data sources that have not yet been made live for a number of reasons, so I believe it will only improve with time.
What isn't clear from the site at current is how news sources are added to Everyblock's scan -- is there a submission process? What information is necessary within a post for it to be picked up? How often are its sources scraped?
This was a much anticipated launch and I think it helps redefine "freedom of information." If nothing else, Everyblock has opened the door for regular access to useful, relevant and important government records from a municipality (in Chicago's case) with a reputation for playing close to the vest. I hope more data will find it's way out in this or a similar fashion. Everyblock is also a wonderful model for a personal "intranet," as it seems to aggregate content that newsrooms work to complile internally and provide externally, and could help citizen journalists conceive their own record sets. Hopefully, the available data will grow and allow for more government records to exist outside of the city's own databases. I'm very impressed. With The Center for Public Integrity's recent work and efforts by other non-profits to utilize technology in new ways, the horizon might be fuzzy, but it gets clearer as we move along.
Andrew, Wilson, sorry for the delay on your comments appearing. The 'ole spam filter decided to munch on them for a while first!
@Wilson Miner, please pass my praise along to Nathan.
@Andrew Huff, I agree on the "catalyst" angle. Bloggers need quick and easy access to data and EveryBlock is serving up a brand new batch. I think coverage, particularly blog coverage is largely dictated by data availability. With EveryBlock opening up several new datasets to the at-home-in-pajamas crowd, I expect to see a surge in blog interest in these areas.
I'm excited to see what GB does with this info. I'm also curious about EB's plan for admitting news sources. Will they have quality checks? (Gosh, that sounds snobby) I'm assuming they've got an automated solution up their sleeve, if so, what on earth will that look like? For all Google's talk about the importance of automation, Google News hand-approves new publications as far as I know. Perhaps there's no way to get away from that.
[...] newspapers, blogs, Craigslist, Yelp, and government databases. Brad Flora on The Methods Blog has a great rundown of the site and the information it [...]
"... What surprised and delighted me most, however, were the site’s maps. I think just about everyone had been expecting them to run Google maps on EveryBlock. I have no idea what they’re using to generate them, but they’re sure a lot easier on the eyes ..."
EveryBlock is indeed a great new project and I'm very happy to be associated with it. To answer your question, the software used to generate the maps is Mapnik (http://mapnik.org).
BTW, I started the Mapnik project to provide a free open-source map rendering engine which could generate beautiful maps. A favourable comparison with google maps is most welcome ;)
[...] Everyblock.com: A big step for citizen journalism. Brad Flora z The Chicago Methods Reporter [...]
Post new comment