Google has released a government requests tool. It’s highly illuminating and may end up being quite disruptive. That’s what surprising data visualizations can do for us.

Google receives thousands of requests from governments around the world to provide user data, mostly relating to criminal cases. It also receives many government requests to take material down from search results, ads, YouTube (non-copyright requests, because those would be coming from private parties), Blogger, etc. Google didn’t include child pornography take-downs in these numbers, because it does that on its own.

The tool allows us to see the number of requests from different countries that Google received during the last six months of 2009. More than 3600 data requests from Brazil during those six months and more than 3500 from the US. But just 40 or so from Canada and 30 from Israel.

Germany asked for removal of material more than 180 times in six months—mostly having to do with court orders related to defamation claims, according to Google. If you click on the country’s results, you’ll see what percentage of removal claims Google complied with during those six months. For Germany, it’s pretty high—94% taken down. Germany also made more than 450 data requests.

At the moment, the data is pretty coarse-grained. We don’t know how many data requests Google complied with. We don’t know what the triggers were, precisely, for any of these requests. We don’t know anything about China. (Why? Google says, essentially, “It’s complicated”: “As noted in the map, Chinese officials consider censorship demands as state secrets, so we cannot disclose that information at this time. During the period that Google’s joint venture operated google.cn, its search results were subject to censorship pursuant to demands from government agencies responsible for Internet regulation.”)

But the numbers are surprisingly high. Thousands upon thousands of requests for data about users?

Perhaps other companies will follow suit. That would be extremely helpful. It’s hard to know how meaningful the Google data is without any comparables.