www.roanoke.com

(Originally published as a DataSphere column in The Roanoke Times, August 3, 2008)

This may be a new low in my descent into geekdom.

I gave my wife a database as an early birthday present.

Worse, I regifted it to her. Still worse, I didn't just give it to her. Anybody can use it.

On the plus side, she loved it.

It's data on all 45,000 real estate parcels in the city of Roanoke, searchable in all kinds of ways. Plus, your results are mapped. Find it right now in the DataSphere.

See, Ellen, she's real estate obsessed. Watches HGTV all the time, talks about "staging" our house for sale, disappears for hours studying home sale listings on the Web. She sneaks off to open houses on Sundays. Real estate agents tell her she knows the market better than they do.

I pegged her for just the kind of "user" -- sorry, honey -- who would love this kind of data, even if it's regifted.

I -- along with anybody else who cares to have it -- can get the data from the city of Roanoke. It's really the data behind the city's real estate GIS, or (nerd-word alert) Graphical Information System. This is the Web site where you can search and get all this great public information on every parcel of land in the city, including a map and, in most cases, a photo.

It's a terrific public service and a valuable tool that many local governments offer on their Web sites now.

But there are some things you can't do on the city's Web site. You can't search for all the houses that sold in a certain time frame, for example, or all the houses sold in a certain price range.

Real estate geeks like my wife, or anyone thinking of buying or selling a house, or people who are just nosy and want to know what their neighbor or ex-wife paid for their new houses, want these kinds of searches.

Now, you can have them in the DataSphere.

This is a great example of something I do often in this job -- I take data that's already out there and accessible, and I re-bundle it in a way that I think is more useful to the average Roanoke Times/roanoke.com reader.

Here, I've taken the same data and repackaged it with a different set of search tools. You can search by buyer's name, seller's name, neighborhood, street, street address, sale price range, and tax assessment value range.

So, you can see, for instance, that 250 properties in Roanoke changed hands in May, and that the highest priced home sale was a place on Crystal Spring Avenue that went for $790,000. Or that the piece of property with the highest tax assessment in the city is a little medical facility on Belleview Avenue valued at $201 million.

The database isn't a perfect machine. But as my boss is fond of saying, don't let the perfect be the enemy of the good.

It doesn't have every bit of data in the city GIS. The file available on the city's Web site, until two months ago, contained all of the GIS data. Since then, some data have been omitted. If that is restored, I can pass it along in our database so you can see things such as square footage in a house, when it was built, how many bedrooms and bathrooms it has, and the kind of details that will really help you analyze the market for yourself.

Sure, that house down the street sold for $30,000 more than you thought you could get for yours, but does it have 500 more square feet and two more bedrooms?

I love that you can search the data on our site by neighborhood, but the neighborhood search includes only single-family homes. You can limit your search to commercial buildings or condominiums, but you can't limit it to those kinds of properties in a certain part of the city.

There's also some lag-time for when the data are available. The current data available were posted by the city on July 14, and the most recent sale dates in it are from June 26.

The city file gets updated monthly, so that's how often I'll update the data in the DataSphere.

And while it's only Roanoke properties right now, my next step is to see if other governments would be willing to share their real estate transfer data, too.

In the meantime, Ellen is very happy with her free, repackaged, regifted database.

She's so easy to please. Love that about her.