Transparency Report: Government requests on the rise



Cross-posted from the Official Google Blog 

We think it’s important to shine a light on how government actions could affect our users. When we first launched the Transparency Report in early 2010, there wasn’t much data out there about how governments sometimes hamper the free flow of information on the web. So we took our first step toward greater transparency by disclosing the number of government requests we received. At the time, we weren’t sure how things would look beyond that first snapshot, so we pledged to release numbers twice a year. Today we’re updating the Transparency Report with data about government requests from January to June 2012.

This is the sixth time we’ve released this data, and one trend has become clear: Government surveillance is on the rise. As you can see from the graph below, government demands for user data have increased steadily since we first launched the Transparency Report. In the first half of 2012, there were 20,938 inquiries from government entities around the world. Those requests were for information about 34,614 accounts.


The number of government requests to remove content from our services was largely flat from 2009 to 2011. But it’s spiked in this reporting period. In the first half of 2012, there were 1,791 requests from government officials around the world to remove 17,746 pieces of content.


You can see the country-by-country trends for requests to hand over user data and to remove content from our services in the Transparency Report itself, but in aggregate around the world, the numbers continue to go up.

As always, we continue to improve the Transparency Report with each data release. Like before, we’re including annotations for this time period with interesting facts. We’re also showing new bar graphs with data in addition to tables to better display content removal trends over time. We’ve now translated the entire Transparency Report into 40 languages, and we’ve expanded our FAQ—including one that explains how we sometimes receive falsified court orders asking us to remove content. We do our best to verify the legitimacy of the documents we receive, and if we determine that any are fake, we don’t comply.

The information we disclose is only an isolated sliver showing how governments interact with the Internet, since for the most part we don’t know what requests are made of other technology or telecommunications companies. But we’re heartened that in the past year, more companies like Dropbox, LinkedIn, Sonic.net and Twitter have begun to share their statistics too. Our hope is that over time, more data will bolster public debate about how we can best keep the Internet free and open.

Why now?

There has not been much Cuban Internet news lately, but on November 2, Granma published a statement by the Cuban Ministry of Foreign Affairs targeting the U. S. Interests Section's Internet activity. (Spanish, English)

The statement accuses the Interests Section of doing illegal training and establishing illegal internet connections and networks.

State Department spokeswoman Victoria Nuland told reporters in Washington that "We are absolutely guilty of those charges. The U.S. Interests Section in Havana does regularly offer free courses in using the Internet to Cubans who want to sign up. We also have computers available for Cubans to use. Obviously this wouldn't be necessary if the Cuban government didn't restrict access to the Internet and prevent its own citizens from getting technology training."

This activity has been going on for some time -- my question is "why publish this statement now?" Does it signal a coming crackdown on the Interests Section? Is it an attempt to influence the U. S. election somehow?

Let us know if you have any knowledge of the Interests Section classes and Internet access. Are the classes political as Cuba claims? Do they only offer access to anti-government activists?

How will the new immigration law impact the Internet?

The big news from Cuba this week has to do with the opening of travel to anyone with a current Cuban passport and a visa from the destination country.

But, everyone cannot necessarily get (or afford) a passport.

The law says a passport can be denied in the interests of defense and national security, to preserve the skilled workforce for economic, social, and scientific-technical development and to protect official information. If that is not enough, the authorities can also deny a passport for "other reasons of public interest."

It is easy to imagine these caveats being used to deny passports to people like dissident bloggers, networking professionals, and computer science students and professors. One can also imagine freer flows of information and IT goods -- particularly between Cuba and the US.

Time will tell.

Freedom House report on Internet freedom in Cuba

Freedom House's annual report Freedom on the Net 2012 is out. The report ranks nations on Internet freedom, which they compute as a function of obstacles to access, limits on content and violations of user rights. The Cuban country report is on pages 152-163 and Cuba ranks 46th out of the 47 nations surveyed.


The following are the actions Freedom House looks for. Those marked with a (C) are practiced in Cuba.

  • Web 2.0 blocked (C)
  • Notable political blocking (C)
  • Localized or nationwide ICT shut down
  • Pro-government commentators manipulate online discussions (C)
  • New law /regulation increasing censorship or punishment passed
  • New law /regulation increasing surveillance or restricting anonymity
  • Blogger/ICT user arrested for political or social writings (C)
  • Blogger/ICT user physically attacked or killed (C)
  • Technical attacks against government critics

The following figures establish some context. This world map shows the nations that Freedom House rates as not free, somewhat free and free.


Here we see all of the not free nations. Cuba is next to last, leading only Iran.


Part of the problem in Cuba is very low Internet penetration in homes, as shown below.


Bear in mind that the correlation we see here between Internet penetration and the index of freedom does not establish causality -- in fact the two variables reinforce each other.