Remind me again...

...why we're supposed to trust governments rather than corporations with our private data?

Hunh

Some scientists have done a study of the forwarding patterns of Internet chain letters and come up with two interesting findings:

1) Ninety percent of the time, when someone forwards a chain letter to a group of people, only one person out of that group will forward it on.

2) At the median, when a person receives an Internet chain letter it will have been forwarded three hundred times before it reaches them.

I'm not sure what any of this means yet, but I find it intriguing. I'm particularly curious how this compares to pure information, as opposed to chain letters. I can't imagine that a piece of information would go through three hundred episodes of telephone before it reaches a person . . . would it?

Full paper here.

The Things One Can Discover...

...with enough data and the right algorithm:

Falling Coca-Cola sales in a specific region of Africa are an excellent indicator of civil unrest, famine, or some other problem in that region.

Now That's What I'm Talking About!

A system that analyzes the content of books and suggests new books for you to read based on the amounts of dialog and action, the density, the pacing, and a couple of other factors. The books currently in the beta are heavily skewed towards science fiction, which is a genre I haven't really read in since I was 16, so I can't say how well it works based on what's in there now. But kudos for the general concept!

The Slashdot story has links to more background on the project and the person behind it.

Be a Data Geek, Win a Prize!

The British government is going to give 20,000 pounds to the person who comes up with the best idea for mashing up and re-using the reams of data collected by said government.

I checked the rules, and it's not limited to British citizens. The deadline is in September. Full details here.

Skewz

It took me awhile to decide whether this new Skewz site is brilliant or one of the signs of the apocalypse, but I think I've come down on the side of "brilliant."

Basically, the point of Skewz is to use the wisdom of crowds to make explicit the bias that exists implicitly in the media, while also functioning sort of like Digg, aggregating stories that people find interesting. People submit stories and then get to vote on how much and in which direction they think the stories are skewed.

But Skewz isn't designed to let people only see the stories that they agree with ideologically—the pages are divided up into one column for "liberally skewed" stories and one for "conservatively skewed" stories, which lets readers see both sides of the news right next to each other. Unfortunately Skewz doesn't seem to actively pair conservative and liberal stories on the same topic, but still, it's useful to be able to see what both sides are talking about on any given day—especially since quite a bit of media bias isn't in how a story is covered but in what people think is newsworthy in the first place.

And, one of the best parts: it aggregates people's skew ratings for each story and uses them to create a giant chart showing the ideological skew of quite a few of the major newspapers and blogs on 20 different issues.

So of course this comes out after I quit editing the books where I had to pair opposing viewpoints and was always running around trying to find somebody arguing the other side of some mildly obscure issue....

Hat tip: Marginal Revolution (again!)

Well That's New and Interesting

While doing a search in Google today I got the following line at the bottom of the first page of results:

"In response to a complaint we received under the US Digital Millennium Copyright Act, we have removed 1 result(s) from this page. If you wish, you may read the DMCA complaint that caused the removal(s) at ChillingEffects.org."

I wonder if Google has been doing this for awhile and it just so happened that today was the first time I happened to do a search that hit one of these, or if this is a new thing?