EveryBlock.com, a new Web site from
Adrian Holovaty that allows you to search for news stories and other types of information by address, zip code or neighborhood, launches this afternoon. I believe this is the beginning of something big.
This screen shot is the first public look at the site that so many of us have been waiting to see. Here's
my first take on some of the reporting you can pursue with a little help from EveryBlock.
You may know Holovaty from his remarkable
ChicagoCrime.org Web site that maps incidents of crime daily. He has developed innovative, award-winning Web applications for washingtonpost.com, Lawrence.com and LJWorld.com. This latest effort was funded by a
Knight News Challenge grant.
I should disclose that Holovaty is on The Poynter Institute's
National Advisory Board. I interviewed Holovaty via e-mail and asked him to explain how the site works and how newsrooms can adapt what he did.
Tompkins: What does EveryBlock do?
Holovaty: EveryBlock filters an assortment of local news by location so you can keep track of what's happening on your block, in your neighborhood and all over your city. We compile news, we classify it by location/geography, and we present a beautiful, easy-to-use interface that lets people view news in specific locations.
Tompkins: How does EveryBlock work?
Holovaty: There are two main ways of reading news on EveryBlock -- by location and by type. You can search for any address, neighborhood or zip code in the city (more on the city list in a bit), or you can browse by type of information: restaurant inspections, mainstream media articles/blog entries, crimes, building permits, etc.
Tompkins: How does the data gathering/classification work?
Holovaty: We have a sophisticated collection of computer programs that crawl news and information from all around the Web. We've written some algorithms that are able to detect locations in free-form text with a reasonable degree of certainty, and we also manually tag information in cases where the computers don't cut it. This is an area of ongoing experimentation.
Tompkins: What cities did you include?
Holovaty: We're aiming high and launching in three of the largest U.S. cities -- Chicago, New York and San Francisco. We'll be adding more cities over time.
Tompkins: You have said that you didn't consider EveryBlock to be a competitor to traditional media. Why do you say that when everybody is competing for eyeballs and time?
Holovaty: Well, under that definition, YouTube, MySpace and, heck, all Web sites, are competitors to traditional media. I don't consider EveryBlock a competitor to traditional news outlets because we only include news that has to do with specific, granular locations -- not citywide, statewide or nationwide news.
On EveryBlock, you'll find out when your local pizza place is inspected, but you won't find an analysis of the mayoral budget or Chicago's bid for the 2016 Olympics (unless they plan to build a stadium near your house).
With this in mind, I see EveryBlock as being quite complementary to traditional news organizations.
Tompkins: How do you hope newsrooms will adapt your ideas and even your code to their own work?
Holovaty: We're interested in spreading the concept of "geocoding" news -- that is, classifying news articles by location. Currently, we do that by crawling news sites and applying algorithms and human editing efforts, but it'd be best for everybody if news organizations did this on their own. We're interested in developing some sort of specification/standard for designating granular locations in news stories -- look for more about that from us soon.
Tompkins: Why, in your opinion, is "hyperlocal" so important?
Holovaty: To be honest, I prefer to avoid using that word, as it has become meaningless. Some people use it to refer to neighborhoods, while others use it to refer to entire suburban areas. But I think the concept of address-specific news is important because, well, people tend to be more interested in news that happens near them. It's as simple as that!
Can't wait to spend some time with this. We also...
