We have, and so has Reddit. For any site that could be described as a community, the people sitting round the table are probably more important than the shape of the table.
E.g. we don't even have our own search (I recommend http://www.webmynd.com/html/hackernews.html if you haven't tried it), but we do crack down pretty vigorously on trolls and spammers, because they hurt a site more than lack of features.
The difference between Hacker News and Everyblock is that Everyblock needs to pay six salaries. (I know they have a parent company with cash now, but does reddit even cover its own hosting and salaries? I honestly don't know.)
Regardless, the grant will end and they will have to generate at minimum six (decent) salaries worth of revenue. I don't think the grant will last them much longer. Don't get me wrong, I know most of them in one way or another, and wish them the best ... but like I said, it isn't exactly like Hacker News or reddit in the 'survive for the fun / publicity of it' sense.
The code is free, but the aggregated data might not :) And the knowledge about how to aggregate, and how to transform this data.
For example, approach Starbucks, and ask them where do they plan to open new stores (er... I know SB specifically is closing shops, but the point remains), and then SELL to them data about the area.
Sure, lots of information are already available on the website. What's to stop Starbucks to just enter the website and do their research? That's where intimate knowledge about the data comes in. SB won't ask what the website can answer. They'll expand the questions, and ask you to convert the data to Excel, and etc..
Know what business need! And put a line on the sand: Your people will do a lot of things that are free to consume and use and mashup and all that jazz. But some of them, you will charge, specially the one-time jobs, or for those rich enough to pay...
You can charge a download fee for the source or binaries. I'd sure as hell pay $100+ to grab a copy of EveryBlock and host it for my neighbourhood or city.
Basically if you release your server-side software under that license, you can get others to release the changes they make so your own hosted copy will become better (hopefully):
"The GNU Affero General Public License is designed specifically to ensure that, in such cases, the modified source code becomes available to the community. It requires the operator of a network server to provide the source code of the modified version running there to the users of that server. Therefore, public use of a modified version, on a publicly accessible server, gives the public access to the source code of the modified version."
Well, as others have said, of course you can survive -- there's more to a site than its codebase.
And second, this isn't exactly a dilemma. They got a grant in exchange for a promise to open source. Seems like a pretty reasonable deal to me. If you want to keep it all to yourself, then you've got to pay for the development yourself.
new product to me. My first impression is that many marketing and election campaigns might make use of these tools well. Such campaigns shouldn't be spending lavishly on IT departments and could make use of consulting and other custom usage by the experts that built the tools.
So the answer to "how to make money off open source?" seems to always lead back to "consulting and support". Not bad, its just not 100x returns on licensing, its "just good revenue".
good luck, would like to see tools like this continue to evolve!!
We have enough examples now where a company open sources its product and still makes money, Redhat, Mysql to name few. You can always open up the base product and charge for additional features (the freemium model).
You have to create an ecosystem around the product it will be profitable. That ecosystem can be additional plugins or services around a free software product. Since free software is based on the concept of sharing the network effect will kick in, for good products.
If the product is good you will make money immaterial of open source or closed source nature. In my opinion free software has more stability and performance but most free softwares are packaged badly (or not packaged at all). When Free software hackers learn the art of packaging a product and start realizing that there are others who are going to use the product too, we can should lot more usable products from open source world.
In the short term, no one will likely bother to copy the code and open a competing (or even related) site. So business-wise, opening the code doesn't have any short-term impact.
In the longer term, if they were to become hugely successful, then many people would love to copy them and having access to the code would make that very tempting. But "hugely successful" is a big if. By then, they probably would have enough of a community that copycats wouldn't hurt them too much.
We have been toying with similar ideas for our startup, especially since its concept lands itself very well to community-based development. No decision yet.
Yes, it should be a paid iPhone app. And/or charge for per-query REST API access to a hosted Everyblock data access API.
Bonus points for the iPhone if we can do simple GIS mashups on the fly like 'Show all the schools in the area' or even better allow filters and drill-downs 'Show all the schools with a graduation rate greater than X%"
E.g. we don't even have our own search (I recommend http://www.webmynd.com/html/hackernews.html if you haven't tried it), but we do crack down pretty vigorously on trolls and spammers, because they hurt a site more than lack of features.