Transparency reports on trial: New front for free speech?

The latest high profile free speech fight isn’t over a book, a movie or even a video game. Instead, the court case is over a corporate report, and has led media companies to join Twitter in an unusual First Amendment challenge of government gag orders.

The court case highlights the growing significance of the so-called “transparency reports” that Twitter and a growing number of other companies, are using to inform users about government demands and other trends that affect the internet.

Since they began appearing five years ago, the reports have served as an important measure of free speech and privacy. But they can also be a PR tool for the companies that publish them. As a federal judge in California gets ready to hear Twitter’s case, it’s time to reflect what these reports are — and are not — all about.

Five years of reports on snooping, and Snowden

Google published the first Transparency Report in 2010. It showed how often the government came calling for information about users (such as Gmail messages), and also explained how many times copyright owners and others asked Google to remove content from the web.

Subsequent Google reports evolved to include data from other countries, and occasionally included droll details about specific government requests — like the time the Mounties demanded to know which Canadian had peed on his passport in a YouTube video (Google refused to say). More seriously, the reports show how governments are using Google as a back-door way to collect information about their citizens, and how this phenomenon is increasing.

After 2010, other tech companies came to follow Google’s example: Twitter, Yahoo, Microsoft and others have now issued several reports of their own. The reports vary in detail and sophistication, but all point to a similar drumbeat of demands by governments for information about users’ identities and personal data.

In 2013, the reports took on a new importance in the wake of Edward Snowden, the former NSA contractor who revealed how U.S. and other intelligence agencies had created a massive surveillance apparatus on the back of data gleaned from tech and phone companies. Civil libertarian groups like the Electronic Frontier Foundation also describe them as essential to pressing for surveillance reform.

“This is how we hold countries accountable for action … Governments behave better when they know they’re being watched and tracked,” said Eva Galperin of the EFF in an interview last year.

Edward Snowden in his first public speaking appearance at SXSW on March 10, 2014.

Edward Snowden in his first public speaking appearance at SXSW on March 10, 2014.

Post-Snowden, the tech industry redoubled legal challenges against government gag orders that preclude them from disclosing the very existence of the stealthy procedures (such as National Security Letters and “FISA Orders”)  that agencies like the FBI and NSA use to demand data.

This pushback includes Twitter’s legal challenge, which last week led a range of media companies — from the Guardian to NPR to BuzzFeed — to file a supporting brief.

“Twitter’s proposed transparency report is no less entitled to free speech protections than ‘literature’ or ‘movies,'” claimed the companies.

But even as the ongoing legal battle over transparency reports heats up,  the actual impact of the reports is unclear.

Public service … or public relations?

In the wake of the Snowden revelations, a crush of companies have rushed to publish transparency reports of their own, often gaining press attention for doing so. This month, for instance, Reddit’s inaugural report merited a write-up in the New York Times even though its contents suggested the government has taken only scant interest in the internet news site.

Screen Shot 2015-02-23 at 2.23.00 AM

From Pinterest’s transparency report

A list on Google’s website shows that at least three dozen other companies, ranging from Verizon to Pinterest, are publishing transparency reports of one type or another (see Pinterest graphic at right). Meanwhile, Google itself continues to publish its own semi-annual report in blog posts under headlines that suggest surveillance is on the rise.

While all of this serves to put a spotlight on government demands, it may also paradoxically serve to to dim it. That’s because the sheer volume of transparency reports, and the regular warning of government spying, could lead to public apathy on the issue.

Public indifference could also be hastened by the complexity of the legal issues (such as the distinctions between search warrants, subpoenas and security letters), and by the rivalries between companies that publish the reports.

Google and Facebook, for instance, have sniped over who deserves credit for standing up to the Justice Department through transparency reports. Meanwhile, the reports themselves don’t provide all the context to determine if surveillance really is becoming more pervasive — or if the rising number of requests is a function of more people being online.

Galperin of the EFF says she would like to see some sort of standardized reporting for transparency documents one day, but disagrees that there is fatigue with the reports themselves.

“I’m less concerned with specific companies, but with ensuring that startups create the reports from the very beginning. If you work with user data, they government will come calling for it.”

How the media should deal with data

An engineer at a major tech company who has helped to produce transparency reports told me he disagrees that the process needs to be more formalized, or that the process needs an oversight mechanism to ensure companies don’t exploit the process for public relations purposes.

“How do you deem from the outset who is the outside audience? I don’t care if someone working with a spreadsheet is a tenured prof or a high school kid good at analytics,” said the engineer, who spoke on condition of anonymity.

He added that certain data sets are more “mediagenic” than others, pointing to less-publicized items like a Google transparency report on unsafe websites, but that publicizing data typically produces change for the better.

If the transparency reports do produce results, this could strengthen Twitter’s court challenge that gag orders directed at them amount to an illegal prior restraint on speech, and that companies have the right to put out so-called “warrant canaries.”

Meanwhile, the media companies’ decision to file briefs in the case appears to validate the engineer’s opinion that the prime duty of those who publish transparent reports is to put out data, not deconstruct it.

“The value of transparency reports is measured by diligent and astute media — those who roll up their sleeves and engage with it and find what it ultimately means, rather than just spitting out facts.”

Tech and media firms join Twitter in key test of FBI gag orders

A bitter fight between the Justice Department and Silicon Valley is expanding as a diverse group of companies have lined up behind Twitter in a case that will help determine the limits of free speech in the age of Edward Snowden.

On Tuesday, groups ranging from BuzzFeed to Wikipedia to the Guardian filed friend-of-the-court briefs (see below) to support a challenge by Twitter to Patriot Act gag orders. Two other large companies, which are only allowed to refer to themselves as “Corporations 1 & 2,” also filed briefs.

The case, which began when Twitter sued the Justice Department in October, turns on how companies may use so-called “transparency reports” to tell users about government requests for their data.

Twitter claims it has a right under the First Amendment to say specifically how often it receives National Security Letters, while the government counters that companies can only do so in broad strokes lest they jeopardize national security.

In recent years, the FBI has made extensive use of National Security Letters to obtain information about subscribers, while also attaching gag orders to the letters that forbid companies from revealing they have even received a letter in the first place. The Justice Department has issued hundreds or thousands of such letters to companies like Google, Facebook and AT&T.

In its lawsuit, Twitter claims it is an illegal prior restraint of free speech for the government to bar companies from even disclosing that they have received a letter. A group of media companies has now voiced support for that argument:

“Twitter’s proposed transparency report is no less entitled to free speech protections than ‘literature’ or ‘movies,'” said the brief filed on behalf of BuzzFeed, NPR, the Washington Post, PEN America, the Guardian and First Look Media.

The brief reflects the media’s newfound legal interest into what has largely been a tech industry fight, but also shows how digital media companies like BuzzFeed are finally taking up the legal fight for free speech, a burden that has long been borne almost entirely by old-line newspaper companies.

“Corporations 1 & 2”

Meanwhile, a separate filing shows that a phone and internet company are also weighing in on the Twitter case, but in the guise of “Corporations 1 & 2.” The companies (which are likely Verizon and Google or Yahoo) are using the pseudonyms at the direction of a judge, and are muzzled in part because they are already before an appeals court in another national security case over the right to disclose government demands.

The right of internet companies to discuss security letters has become more pressing since 2013 , when leaked documents from Edward Snowden revealed massive surveillance operations by the U.S. government. Those operations rely on obtaining information from tech and phone companies, and have been facilitated by the legal process governing Patriot Act letters, as well as a related process for NSA demands.

In response, companies like Twitter have come to claim that free speech and the public interest give them the freedom to disclose how many NSA and FBI letters they receive in the first place. The companies stress they are not arguing for the right to disclose the contents of the letters, since doing so could jeopardize ongoing investigations, but only the existence of the letters.

The docket also shows that a group of other entities  — the Wikimedia Foundation, CloudFlare, Sonic, Wickr, Credo Mobile and Automattic (publisher of — filed a brief in support of Twitter.

Here’s a copy of the media companies’ filing with some of the key parts underlined. Note that a key part of the argument turns on whether the federal judge has authority to hear the case in the first place (as the companies argue) or if the case belongs instead in a controversial secret court (as the Justice Department claims).

Media Amicus in Twitter Case

[protected-iframe id=”c2f560431d071729a0491afe9d08caae-14960843-34118173″ info=”” width=”100%” height=”600″ frameborder=”0″ scrolling=”no”]

This article was updated at 12:35pm ET to note that Automatic is the publisher of; an earlier version said “WordPress” (which refers to the software used by the company, This article was also updated at 1:40pm on Thursday to clarify that it was the Wikimedia Foundation (not Wikipedia) that was on the amicus brief.

A guide to the French national cloud(s)

What does a national cloud look like? In France’s case, it looks like two fairly similar ventures — one based on OpenStack and the other soon to follow in its footsteps. Does this approach make sense?