Data is decidedly intended to be either private or public. If your data is private, you want very much to keep it that way and you’ll go to great lengths to ensure the data’s privacy. Just ask the leadership team at Twitter how they feel about their strategic notes being leaked to TechCrunch and laid bare for all the world to see. The corollary is that if your data is public, you very much want as many people to see the data as possible.

We all know that viruses can “go viral” as can chain letters, YouTube videos and other memes. But what about data – can it go viral?

Yes, it can. It’s a function of how useful, valuable or interesting the data is. At Socrata, we call this the social data effect.

The social data effect contends that the farther data is replicated and propagated from its initial source, the more intrinsically useful, valuable or interesting the data is.

Let me show you the social data effect by example.

On July 1st, the White House New Media team wrote a blog post, in which they publicly shared the salaries of all White House staffers:

A blogger from Talking Points Memo (TPM) read the White House blog and wrote a short blog post of her own, adding some editorial opinion and embedding the salary data into her own story:

The next day, a writer for GOOD / blogs saw the story in Talking Points Memo and wrote his own story, adding some editorial opinion of his own. The writer attributed the TPM story via a hyperlink. He also embedded the salary data into his post:

Within a week, the salary data had been propagated to more than 20 different websites and 100,000 people saw the data – half of which saw the data on some site other than its official and initial source – The White House blog. That’s the social data effect in action.  Because this dataset was propagated it’s an intrinsically useful, valuable or interesting dataset.