NSA's New Transparency Report Contains Just Enough Info To Be Dangerous, Not Nearly Enough To Be Truly Transparent

from the because-'Slightly-Less-Opacity-Report'-doesn't-have-the-same-ring dept

Before we dive into the latest IC transparency report [PDF] from the Office of the Director of National Intelligence, let's take a moment to recognize the small miracle that it even exists. If NSA contractor Ed Snowden hadn't decided to color outside the official whistleblowing lines, we'd still be expected to put our complete trust in the government with zero evidentiary support.

That being said, the transparency report is still several steps removed from actual transparency, but it will have to do for now. What can we learn from it, even with many of the numbers being seemingly meaningless thanks to purposefully-missing context? Several things, actually. Marcy Wheeler has torn apart the report across four posts, each dealing with the report's fuzzy numbers (or, in the case of the CIA's contribution, a lack thereof).

One of the first misleading numbers in the report is the supposed single search of the NSA's 702 collections by the FBI for non-terrorism-related purposes. According to the report, this happened exactly once. But that's actually not true. The FBI makes far more frequent use of NSA data for non-terrorism investigations. It just does it in a way that won't show up in the IC's transparency report. Parallel construction is the FBI's friend.

FBI’s querying system can be set such that, even if someone has access to 702 data, they can run a query that will flag a hit in 702 data but won’t actually show the data underlying that positive return. This provides one way for 702-cleared people to learn that such information is in such a collection and — if they want the data without having to report it — may be able to obtain it another way. It is distinctly possible that once NSA shares EO 12333 data directly with FBI, for example, the same data will be redundantly available from that in such a way that would not need to be reported to FISC.

So, there's that bit of obfuscation right off the top. And the FBI isn't the only agency using an ostensibly foreign-facing collection to obtain information about US persons. The CIA -- an ostensibly foreign-facing agency -- does this as well. The FBI doesn't count its dips into the NSA haystacks. Neither does the CIA. The report shows 30,000 searches of unminimized US persons' data occurred last year. That number doesn't include the FBI's searches (because the FBI doesn't report its searches) and is quite possibly much, much higher than what's reported. This is only a good faith estimate by the IC, using software, rather than any form of reporting from the CIA.

NSA will rely on an algorithm and/or a business rule to identify queries of communications metadata derived from the FAA 702 [redacted] and telephony collection that start with a United States person identifier. Neither method will identify those queries that start with a United States person identifier with 100 percent accuracy.

As Wheeler points out, it could be 30,000… or 3 million… or 3 billion searches. No one knows. By the time the CIA's required to count its US persons searches, it will likely perform most of its searches under Executive Order 12333 authorities, rather than the more closely-watched Section 702.

Finally, there's a really big number contained in the report. It looks amazingly high, but might be indicative of not much surveillance activity at all, at least not in the entire scheme of things. According to the report, the NSA was able to scoop up 151 million "call detail records (CDRs)" using only 42 selectors.

Read in the (lack of) context in the report, this would look like pure bullshit. There's no way 42 terrorism suspects (and their 3,150 one-hop "friends") are making 130 calls a day. (Or, if they're only talking to each other, 65 calls a day.)

As Wheeler points out, call records are not just records about phone calls. They also pick up records on text messages.

If these were phone calls between just two people, then if our terrorist buddies only spoke to each other, each would be responsible for 24,000 calls a year, or 65 a day, which is certainly doable, but would mean our terrorist suspects and their friends all spent a lot of time calling each other.

The number becomes less surprising when you remember that even with traditional telephony call records can capture calls and texts. All of a sudden 65 becomes a lot more doable, and a lot more likely to have lots of perfectly duplicative records as terrorists and their buddies spend afternoons texting back and forth with each other.

With this, 151 million records looks less like full-blown exploitation of this surveillance authority and something possibly more targeted than the NSA's used to. Then again, it could mean the NSA is sweeping up 65 innocent Americans every day of the year with its CDR demands. There's simply no way to tell.

But CDRs include all "call events," which include a whole lot of related metadata having nothing to do with voice calls.

A CDR is defined as session identifying information (including an originating or terminating telephone number, an International Mobile Subscriber Identity (IMSI) number, or an International Mobile Station Equipment Identity (IMEI) number), a telephone calling card number, or the time or duration of a call.

Further trimming down this seemingly large number are two other aspects of the collection. Records obtained previously by the agency are included in this count, as well as junk metadata related to past selectors that may not be returning any current records.

That means our 3,192 targets and friends might only have had 48 calls or texts a day, without any duplication.

Which is a completely believable number of calls and texts between surveillance targets. The breathtaking 151 million records is suddenly a more manageable number that actually *gasp* looks as though the NSA is engaging in truly targeted collection.

But before we get carried away with the NSA's new "maybe collect a little less than it all" approach to surveillance, we need to remember this only covers a very small part of the NSA's collection activities.

[W]e need to understand the 65 additional texts — or anything else available only in the US from a large number of electronic communications service providers that might be deemed a session identifier — a day from 42 terrorists and their 3150 buddies [is] on top of the vast store of EO 12333 records that form the primary basis here.

Because (particularly as the rest of the report shows continually expanding metadata analysis and collection) this is literally just the tip of an enormous iceberg, 151 million edge cases to a vast sea of data.

That's what we're really dealing with here, unprecedented transparency or no: there is a vast surveillance apparatus operating in near-complete darkness, authorized by a presidential executive order and subject to almost zero oversight. Whatever concessions the NSA makes in relation to Section 702 in the upcoming months, its biggest collections will remain untouched. Unless something changes dramatically, the potential for constitutional violations and agency abuse remains unchanged. And, unless something changes dramatically, it will remain unseen.

Filed Under: nsa, odni, surveillance, transparency

Facebook Reports More Than Half Of Gov't Demands For Content And Data Come With Gag Orders Attached

from the greater-transparency-reveals-more-secrecy dept

US government requests for Facebook data are up, according to the company's latest biennial transparency report. Total requests jumped from 23,000 to 26,000, as compared to the first six months of 2016. Overall, it's an increase of about 12,000 requests over 2015's total.

At this point, Facebook is fielding about 1,000 more requests a month as compared to 2015. While there's not a whole lot of detail in the presented data, the social media platform is now able to report something it hadn't been able to do before the passage of the USA Freedom Act. Both of the 2016 reports now show what percentage of data requests come with a gag order attached.

According to the report, more than half of the orders received in 2016 came with gag orders: 56% for the first half of the year and 50% in the second. Gag orders are standard operating procedure for National Security Letters but those only make up a small percentage of the requests Facebook receives: less than 1,000 total for all of 2016.

A majority of the requests are search warrants, which often come with limited-time gag orders. Or at least they should be limited-time. As we saw in Adobe's recent court win, search warrants appear to be arriving with indefinite gag orders when only delayed notification is justified. Facebook has been pushing back against over-broad requests and government secrecy, but its success has been limited.

Also of interest is the lack of FISA orders. For the first time since Facebook began reporting these numbers, it received no FISA orders for content or non-content data. This would seem to confirm an apparent Section 702 blockage at the FISA court. As was noted here earlier, the FISA court apparently approved zero government requests under Section 702 (internet content/non-content collections) in 2016. The last approved order of 2015 would have carried over into the first couple of months of this year, which may be why Facebook shows FISA requests in the first six months of last year, but nothing past the point of that order's expiration.

So, the government has begrudgingly agreed to a little more transparency, but is ensuring much of what's behind the numbers remains firmly in its grasp. Gag orders accompany at least half the requests the government sends out, and the number of requests continues to increase. At this point, Facebook fields a couple thousand gagged orders a month, with most, presumably, tied to sealed cases and indictments, which kind of makes a mockery of this whole "open court" system. Considering Facebook maintains control of users' accounts, communications, and data on its platform, it makes little sense to withhold notification for extended periods, if not indefinitely.

Filed Under: first amendment, free speech, gag order, government requests, nsl, surveillance, transparency, warrants
Companies: facebook

Bose Lawsuit For Collecting Headphone Data Is Flimsy, But Highlights Continued Lack Of Real Transparency

from the dumb-tech-is-often-smarter dept

Being transparent about what private consumer data is being collected and sold appears to be a hard lesson for hardware vendors to learn. Earlier this month, Bose was hit with a new lawsuit (pdf) accusing it of collecting and selling personal subscriber usage data of the company's $350 QC 35 noise-canceling headphones. More specifically, the lawsuit claims that the Bose Connect smartphone companion app is collecting user preferences when it comes to "music, radio broadcast, Podcast, and lecture choices" -- and then monetizing that data without making it clear to the end user:

Unbeknownst to its customers, however, Defendant designed Bose Connect to (i) collect and record the titles of the music and audio files its customers choose to play through their Bose wireless products and (ii) transmit such data along with other personal identifiers to third-parties—including a data miner—without its customers’ knowledge or consent...Though the data collected from its customers’ smartphones is undoubtedly valuable to the company, Defendant’s conduct demonstrates a wholesale disregard for consumer privacy rights and violates numerous state and federal laws.

To be clear, the complaint, filed last week by Bose customer Kyle Zak in federal court in Chicago, seems more than a little thin. The suit appears to piggyback on growing concern about the wave of internet of things devices (from televisions to smart dildos) that increasingly use internet connectivity to hoover up as much as possible about consumers. Often, this data is collected and transferred unencrypted to the cloud, then disseminated to any number of partner companies without adequate disclosure.

That said, while Bose marketing insists users need the app to "get the most out of your headphones" and get the "latest features" for their headphones, in this instance, users can avoid data collection by simply not using the Bose companion app. And while Bose only appears to be collecting metadata, the suit tries to somehow claim that collecting this type of metadata -- which any and every music service also happily collects -- somehow violates the Wiretap Act:

... customers must download and install Bose Connect to take advantage of the Bose Wireless Products’ features and functions. Yet, Bose fails to notify or warn customers that Bose Connect monitors and collects—in real time—the music and audio tracks played through their Bose Wireless Products. Nor does Bose disclose that it transmits the collected listening data to third parties.

Were Bose, say, using the headphone jack on a headset to monitor actual user communications, the case might have legs. That said, while the suit's central Wiretap Act claims may be weak, the suit once again highlights that consumer data collection policies, if disclosed at all, are often buried in overlong privacy policies few if any consumers actually read -- using language carefully crafted to obfuscate what precisely is happening. Bose doesn't really help its case all that much in a statement on its website that declares the lawsuit "inflammatory" and "misleading," before being a little misleading itself:

We understand the nature of Class Action lawsuits. And we’ll fight the inflammatory, misleading allegations made against us through the legal system. For now, we want to talk directly to you. Nothing is more important to us than your trust. We work tirelessly to earn and keep it, and have for over 50 years. That’s never changed, and never will. In the Bose Connect App, we don’t wiretap your communications, we don’t sell your information, and we don’t use anything we collect to identify you – or anyone else – by name.

While Bose insists it doesn't "sell your information" -- its app privacy policy does note that it "may partner with certain third parties" to "engage in analysis, auditing, research, and reporting" (hey, it's not selling if we call it something else). And while Bose may not personally identify you "by name," we've long noted that "anonymized" data is far from anonymous. Study after study has made it clear that it only takes a shred of additional contextual data to make "anonymous" data easily and personally identifiable. If "trust" were truly Bose's top priority, they'd actually explain precisely what the app is doing, who data is sent to, and why.

Again, many may not care that Bose is collecting this data. Especially in an age where everybody carries around a miniature computer in their pocket, happily oblivious that their every step and click are being monetized by cellular carriers, app vendors, OS makers, advertising networks, and everybody else in the food chain. The problem is that companies continue to believe there's nothing wrong with hoovering up every shred of data they can, then hiding this collection in overlong, carefully-worded privacy policies -- and the false sense of security "anonymization" is supposed to provide.

Filed Under: information, kyle zak, privacy, tracking, transparency, wiretap act
Companies: bose

NYPD Finally Comes Up With A Body Camera Policy, And It's Terrible

from the more-tech,-less-transparency dept

Nearly four years after the NYPD was ordered by a federal judge to implement body cameras, the department is finally getting around to finalizing its rule set for deployment. Part of the delay is due to the NYPD seeking input from the public -- input it has apparently decided to ignore.

As Scott Greenfield notes, the NYPD gets everything wrong about its policies, applying guidelines that directly contradict the responses received from everyone in New York City not wearing a blue uniform.

The first “big” question is when will cops be required to turn their body cams on, since having them doesn’t actually serve much of a purpose if they’re turned off.

Notice anything peculiar? Like the public wants them on a lot, and the cops, not so much? But this belies the problem: if body cams must be on for “use of force,” will cops call a “time out” when a situation develops where they decide to tune up a guy who isn’t sufficiently compliant so he can flip the switch? Sure. Who doesn’t honor the sacred “time out”?

But then, the “it depends” on witness interviews is somewhat disconcerting. After all, why record witness interviews, since they might say something inconsistent with the cop’s recollection or their testimony in court? That could be unpleasant.

The public also wants expansive footage release policies. Unsurprisingly, NYPD officers do not. From the NYPD's body-worn camera report [PDF]:

Officers and the public were asked whether they agreed or disagreed with this statement: “If a person has an interaction with an officer wearing a  body-worn camera, the NYPD should be required to show that person the footage upon request.” Here is how they responded:

Here are the responses for a similar question, but involving third parties like journalists and advocacy groups making requests for footage:

The split is obvious. The public wants access. The NYPD wants sole control. The "compromise" is this:

[T]he NYPD is refusing to take this step. Instead, it will require footage requesters go through the slow and ill-suited Freedom of Information Law public records process — the same one you’d have to use to get, say, the mayor’s travel records.

To date, when people have used the FOIL process to request footage from the tiny number of body cameras already deployed in an experiment, the NYPD has stood in the way, charging the public exorbitant fees and claiming broad exemptions.

When it comes to New York's open records law, approaching the NYPD for documents is an exercise in futility. The department has been called "worse than the FBI, CIA, and NSA" when it comes to responding to records requests. FOIL lawsuits are minimally effective, as the NYPD is as comfortable with slowly bleeding plaintiffs dry as it is ignoring their requests entirely.

The other twist is this: if you're facing criminal charges and want access to footage of your arrest, etc., you're not going to get any preferential treatment. You also will be forbidden from joining the FOIL line and hearing your request will be ignored in the order it's received. Scott Greenfield breaks this down:

But certainly the defendant is entitled to the video, right?

"There is an important exception with respect to release of body-worn camera footage: if a person is arrested and has a pending criminal case, and seeks body camera footage related to his or her arrest, he or she may not come to the NYPD to circumvent the standard discovery process between the prosecution and the defense. Discovery is governed by New York State Criminal Procedure Law. Criminal defendants are entitled to these recordings under the law, but such requests are handled by prosecutors in accordance with existing criminal discovery practices and procedures."

Discovery? That same criminal procedure process that has been the target of reform for decades because it’s nearly useless?

Unbelievably, the NYPD camera policy gets even worse, and even further away from the public's preferences. Here are the responses given to the question of when officers should be given access to body camera footage. Note that the largest split involves the viewing of camera footage before writing reports.

The NYPD policy sides with the 91% of officers who stated they should be given access before writing reports or issuing statements. The explanation of its decision to run contrary to public opinion cites two things:

1. Plenty of other law enforcement agencies have similarly bad camera policies.

The NYPD body-worn-camera working group has reviewed the body-worn camera policies of nearly 30 police departments. All of them allow officers, without restriction, to review body-worn camera video prior to filling reports when there has not been a significant use of force.

2. NYPD internal investigations are probably the most thorough, serious investigations in the history of internal investigations.

The propriety of this approach requires some understanding of how serious use-of-force investigations proceed in New York City… [two pages of Complicated Hypothetical Situation…]

So, the NYPD will join the 21st century, already in progress, with 1,000 cameras and policies that benefit no one but the officers wearing them. The presumption for footage will be nondisclosure and the only people guaranteed to see the footage will be those who can tailor their narratives to recordings after the fact. The NYPD believes it's the best police force in the nation, if not the world. But it's still far behind several smaller agencies, both in terms of tech adoption, as well as transparency and accountability.

Filed Under: accountability, body cameras, new york, nypd, transparency

Trump Administration Kills Open.Gov, Will Not Release White House Visitor Logs

from the beat-it,-serfs dept

It will never be said that the Trump presidency began with a presumption of openness. His pre-election refusal to release his tax returns set a bit of precedent in that regard. The immediate post-election muffling of government agency social media accounts made the administration's opacity goals… um… clearer.

So, in an unsurprising move, the Trump administration will be doing the opposite of the Obama administration. The American public will no longer have the privilege of keeping tabs on White House visitors. (h/t Alex Howard)

The Trump Administration will not disclose logs of those who visit the White House complex, breaking with his predecessor, the White House announced Friday.

The administration is justifying this reversal with the usual: favorable interpretations of FOIA lawsuit rulings and "national security" mumbling.

White House communications director Michael Dubke said the decision to reverse the Obama-era policy was due to “the grave national security risks and privacy concerns of the hundreds of thousands of visitors annually.” Instead, the Trump Administration is relying on a federal court ruling that most of the logs are “presidential records” and are not subject to the Freedom of Information Act.

Yes, it's sadly true. The administration can use this ruling to lock the public out of this small layer of transparency. The rest of it, however, is bullshit.

Whatever "national security risks" may exist during White House visits should be addressed by intelligence agencies and the Secret Service rather than being withheld from the public. The White House hosts top foreign government officials all the time and it is always a "national security risk." Disclosing who's visited the White House AFTER THEY'VE ALREADY LEFT does zero damage to national security.

Additionally, there are likely several visitors to the White House every year that aren't logged for security reasons, and if it's really that much of a concern, the administration could release the logs with redactions, like Obama did.

As for visitors who aren't government officials (domestic or foreign), it's pretty imaginative to assume visits to the most well-known home of public servants in the free world carry with it some form of unbreachable privacy.

On top of everything else, it's extremely hypocritical for the administration to pretend this is about privacy and security when the president has been hosting government official get-togethers at resorts -- a place where logs aren't kept and "national security risks" seem to be less of a concern than how many holes Trump can fit in between government business.

So, to further distance himself from the people he serves (and the people who elected him), Trump and his administration have shut down the transparency portal put in place by the previous Commander-in-Chief:

White House officials said the Administration is ending the contract for Open.gov, the Obama-era site that hosted the visitor records along with staff financial disclosures, salaries, and appointments.

The administration can't even perform this move without meaningless, self-justifying dissembling. It's not about keeping secrets, of course. It's about saving taxpayers money [eyeroll]:

An official said it would save $70,000 through 2020 and that the removed disclosures, salaries and appointments would be integrated into WhiteHouse.gov in the coming months.

Thanks, Trump. I love the phrase "coming months," which means anytime between 2018 and never. The smart money's on not seeing any financial disclosures until nearly a year from now, at the earliest. The only way we'll see anything sooner is if some White House cabinet scandal manages to dislodge it first. Plus, there's this, from Trump himself, who obviously has no idea his past tweets are accessible by everyone:

If you can't see the tweet, it's Trump calling out Obama for doing the same thing Trump is now doing: rolling back a predecessor's openness.

Why does Obama believe he shouldn't comply with record releases that his predecessors did of their own volition? Hiding something?

President Obama was better talking about transparency than engaging in it. President Trump, on the other hand, has expressed zero interest in transparency and appears to be rolling back anything "open" Obama grudgingly put into place. Maybe it's better to have White House animosity towards openness and accountability right there on the surface. But right now, it really doesn't feel like an improvement.

Filed Under: donald trump, foia, open.gov, transparency, visitor logs

Twitter Reports On Government Agencies Using 'Report Tweet' Function To Block Terrorism-Related Content

from the reporting-the-reporters dept

Twitter's latest Transparency Report contains a new section that shows some governments may be trying to use Twitter's own rules to achieve censorious goals. Legislators and misguided lawsuit plaintiffs have been complaining for years social media services don't do enough to curtail terrorists and terrorism-related content. This has been the subject of multiple lawsuits and multiple Congressional hearings.

However, governments can only do so much to pressure social media services into regulating content. If the government steps in to set the rules, then it crosses the line. The US government has, so far, been unwilling to act as a direct censor of content. Other governments have no qualms about censorship, but have found their efforts somewhat blunted by Facebook, Twitter, etc. being US-based companies, where compliance with foreign directives is a nicety, not a legal requirement. Of course, both companies have voluntarily acted as local censors in response to foreign laws and legal threats.

Fortunately for these governments, Twitter has a way to let them achieve their censorship goals without having to resort to legal threats or new legislation. The new way to control content lies in the site's terms of service, as the Twitter blog post points out.

For the latest reporting period (July - December 2016), this new section is limited to data about government reports to remove content in violation of Twitter’s terms of service (TOS) against the promotion of terrorism. This does not include any legal requests, regardless of whether they result in a TOS violation, which will continue to be published in our Removals Requests report.

For the last six months of 2016, Twitter received reports on nearly 6,000 accounts from a total of 716 reports by government agencies. The numbers aren't broken down any further than that, so there's no telling which governments are utilizing this reporting system most. All Twitter is reporting is that less than 2% of account suspensions are the result of government reports and that it's refused to act on 15% of government-reported accounts. Each account is counted only once, even if there are multiple reports or multiple tweets reported by government agencies.

So, does this government reporting qualify as censorship? It only would with Twitter's help. If Twitter is only removing legitimate requests for terrorism-related content, then government agencies are being treated no differently than any private citizen reporting similar content. If it's suspending accounts or removing tweets simply because the reporting government doesn't like what's being said (or who's saying it), then it's acting as a censorious extension of the reporting government.

Filed Under: takedowns, terms of service, transparency
Companies: twitter

How Drones Help Transparency Activists To See Things The Hungarian Government Wants To Hide

from the not-just-about-dealing-out-death dept

It's remarkable how quickly drones have become a familiar part of the modern world. Like most tools, they can be used for good and evil, but it tends to be the latter that is highlighted when it comes to drones. In the last few days, it was widely reported that President Trump has given the CIA power to launch drone strikes against suspected terrorists, in addition to being able to use the technology to locate them. Dealing death from the skies may be the most dramatic application of drones, but there are plenty of other, more benign, uses, even if they receive less attention. For example, activists in Hungary have been deploying them in a variety of innovative ways in order to bolster transparency and openness in a country where these are increasingly under threat. That's because the country's prime minister, Viktor Orbán, is a self-confessed believer in the "illiberal state," which Wikipedia describes as follows:

a governing system in which, although elections take place, citizens are cut off from knowledge about the activities of those who exercise real power because of the lack of civil liberties. It is not an "open society".

The Hungarian organization Atlatszo.hu wants to reconnect citizens with that knowledge about those in power:

Established in 2011, atlatszo.hu -- "atlatszo" means transparent in Hungarian – produces investigative reports, accepts information from whistleblowers, files freedom of information requests, and commences freedom of information lawsuits in cases where its requests are refused.

Atlatszo.hu operates a Tor-based anonymous whistleblowing platform (Magyarleaks), a freedom of information request generator for the general public (Kimittud), a crowdsourced bribe tracker to report everyday corruption anonymously (Fizettem), and an independent blogging platform for other NGOs and independent media.

Atlatszo.hu uses a wide range of modern technologies in its work, and that also includes drones. Here's a post on Open Society Foundations from a few months back explaining why eyes in the sky are a powerful tool for taking a look at things governments would rather keep to themselves:

Through drone footage, we've revealed the hidden assets of government politicians and pro-government oligarchs, including castles acquired by companies tied to the son-in-law of Hungary's prime minister. Such concrete signs of personal enrichment -- which, in many cases, can only be filmed from the air -- give citizens a clear picture of the corruption and inequality that is all around them.

At the same time, drones are useful for throwing into relief the power of civil society. In 2014, we captured aerial footage of the protests against the government's internet tax.

Recording protests from the air is important because it allows more accurate estimates of crowd sizes to be made, which are also harder to challenge given the detailed footage that goes well beyond what is possible to gather on the ground. There's a video showing this and other aspects of Atlatszo.hu's work, mostly in Hungarian, but with English subtitles, that gives a good idea of the huge potential for using drones in this domain -- and of the pushback activists are already receiving from the deeply unhappy authorities as a result. As drones become ever-cheaper and ever-more powerful, that tension seems likely to increase.

Follow me @glynmoody on Twitter or identi.ca, and +glynmoody on Google+

Filed Under: activism, drones, hungary, transparency

Court Says FBI Doesn't Have To Hand Over Its Rules For Surveilling Domestic Journalists

from the petitioners-immediately-placed-under-surveillance dept

A couple of years ago, the Freedom of the Press Foundation sued the DOJ over its refusal to release its secret rules governing spying on the nation's journalists. This was prompted by revelations the FBI had used National Security Letters to obtain information on AP and Fox News journalists. The DOJ then issued new rules on the do's and don'ts of surveilling journalists, but once again (a) redacted them into uselessness and (b) granted the FBI an NSL exception, undercutting the entire point of the recrafted rules.

The OIG report -- in which the Inspector General disputed the DOJ's extensive redactions -- still has yet to be released in a less-redacted form. Sadly, it now appears it will never be any less redacted than the unintelligible mess the DOJ handed over a few years ago. A federal judge has sided with the government, finding its investigative techniques and methods are too sensitive to be handed over to the public, much less journalists it may or may not have surveilled using NSLs. (h/t Trevor Timm, Mike Scarcella)

Underlying everything is the government's barely-contested assertion that these rules contain information of national security interest. It's a handy assertion because it means the DOJ doesn't even have to explain why the redactions it made are relevant to its national security claims. From the opinion [PDF]:

Recognizing the Court’s “limited institutional expertise on intelligence matters,” the Court accords substantial weight to Hardy’s representation that “any greater specificity in the descriptions and justifications . . . could reasonably be expected to jeopardize the national security of the United States.” Hardy Decl. ¶ 46. In the area of national security, “it is conceivable that the mere explanation of why information must be withheld can convey valuable information to a foreign intelligence agency.”

This same argument appears later in the decision when dispensing of the plaintiff's challenge to yet another FOIA exception deployed by the FBI:

[T]he Court finds Defendant has met its burden and properly withheld documents under Exemption 7(E) for all five categories of documents. Defendant described with particularity that the withheld documents all contained non-public information about the FBI’s investigative techniques and procedures. These pages not only identified NSLs as an investigative technique, but also described information such as the circumstances under which the techniques should be used, how to analyze the information gathered through these techniques, and the current focus of the FBI’s investigations. As in Hamdan, the Court concludes that the declarations, which state that further detail would “reveal[] the very information it seeks to protect,” are sufficient to satisfy Defendant’s burden.

The court's also uninterested in double-checking the DOJ's secrecy assertions. As always, the court has the option to review contested documents behind closed doors (and away from plaintiffs) to see if the government's claims of national security/investigative concerns are valid. The court here declined to exercise this privilege.

The Court finds that in camera review is unnecessary here as Defendant already provided sufficiently detailed factual information in support of its exemptions in the Hardy Declarations and Vaughn Index.

The FBI is still free to use NSLs to obtain information -- including call records -- about journalists. The DOJ may have reined in other options, but NSLs have always been the FBI's go-to form when it's stymied by the FISA court, magistrate judges, or internal guidelines. Nothing about that will change. As for what's actually changed at the DOJ as a result of this public outing of journalist surveillance, the public apparently can't be trusted with that information.

Filed Under: doj, fbi, first amendment, freedom of the press, journalists, rules, surveillance, transparency
Companies: freedom of the press foundation

California Law Enforcement Union Sues To Block Police Accountability

from the we'll-take-the-power...-hold-the-responsibility dept

Because there's just not enough opacity shrouding police misconduct and not enough slanting of the criminal justice system against defendants, California police unions have decided to get involved in a judicial dispute over lists of law enforcement officers whose half of "our word against yours" isn't quite as bulletproof as is normally assumed.

A Los Angeles sheriff is trying to do the right thing, but he's running into opposition from his own supposed "representatives."

The Los Angeles County Sheriff’s Department has collected the names of about 300 deputies who have a history of past misconduct — such as domestic violence, theft, bribery and brutality — that could damage their credibility if they testify in court.

Sheriff Jim McDonnell wants to send the names to prosecutors, who can decide whether to add them to an internal database that tracks problem officers in case the information needs to be disclosed to defendants in criminal trials.

I don't imagine prosecutors are exactly thrilled to be the recipient of information that damages the credibility of their favorite witnesses, but it's probably better than having your witness destroyed in open court by a defense attorney. But prosecutors may never see this information, thanks to the police union's belief that officers shouldn't be held accountable for anything.

The union that represents rank-and-file deputies strongly opposes providing the names to prosecutors and has taken the department to court. The Assn. for Los Angeles Deputy Sheriffs argues that the disclosure would violate state laws protecting officer personnel files and draw unfair scrutiny on deputies whose mistakes might have happened long ago.

The union is wrong. Officers' misconduct records are a crucial part of their trustworthiness. Burying these just makes the union look like a willing enabler of bad behavior. There would be no "unfair scrutiny" of deputies. Judges and juries are perfectly capable of determining whether past misconduct is relevant to the case at hand. The union's lawsuit seeks to place the determination of officers' credibility solely in the union's hands. And in its hands, all officers are credible until proven otherwise -- something that will be almost impossible to do with exactly zero information on hand.

The union's move is a preemptive Brady violation. Brady material is exculpatory evidence and information prosecutors are statutorily required to turn over to the defense. That would include misconduct records, which might point to a testifying officer's lack of credibility, or show a pattern of relevant misconduct. These files would not be made public, which undercuts the union's "privacy violation" claims. True, some of the files' contents would be made public during court proceedings, but it's not as though the sheriff is asking the DA's office to post the contents of the list on its website.

The union wants law enforcement officers to have more rights than the people they serve. The body of a person killed by an officer hasn't even begun to cool before department press liaisons are pushing the dead person's criminal background check results into the hands of every reporter covering the incident. No one expresses any privacy concerns when a 20-year-old arrest is used to alter the public's perception of a police shooting victim. But when it comes to cops themselves -- public servants with immense power, layers of immunity, and publicly-funded opacity that separates them from the consequences of their actions -- privacy is of utmost concern.

Filed Under: california, los angeles county sheriff's department, police, transparency

With So Much Public Interest In Our Judicial System, It's Time To Free Up Access To Court Documents

from the pacer-isn't-keeping-pace dept

Like hundreds of thousands of Americans, I am closely following the "airport cases" around the country. In order to keep abreast of the latest developments in one of the fastest-moving cases, Washington v. Trump, I built a Twitter bot that scrapes the public docket mirror hosted by the Ninth Circuit and tweets about new documents and links as soon as they're added.

This case leads a legal push that has attracted incredible amounts of public attention. There have been tens of thousands of protestors, dozens of organizations and companies that submitted amicus briefs (including Techdirt's think-tank arm, the Copia Institute), and over 135,000 people who tuned into the audio-only livestream of the Ninth Circuit oral arguments (which was also broadcast live on multiple news channels).

Those numbers reveal a public demand to be informed and to participate in the law. But they also show the limitations on the kind of transparency that can satisfy that demand. Most notably, any attempts to make court proceedings more accessible to the public has to contend with the expense and overhead of dealing with PACER. My bot is only possible because the Ninth Circuit provides a public docket mirror for individual "cases of interest," essentially duplicating the existing system outside the paywall. Those mirrors are manually updated, which means they are labor-intensive, error-prone, and not always up to date.

By contrast, look at the @big_cases bot run by USA Today reporter Brad Heath. It monitors a set of district court cases, selected by hand, and posts new documents as they get filed. These district court cases don't have public docket mirrors, so @big_cases accesses PACER directly -- and for that, it needs user credentials and ultimately to pay for the documents it downloads. For a journalist whose job is reporting on legal developments, paying these costs makes sense -- and sharing the documents further is a valuable public service. Without institutional backing, though, it's hard to justify the PACER expenses.

The costs go beyond the financial. These bots represent an experiment in meeting members of the public where they are, and those efforts are less likely if they come with a pricetag. Worse, it means these experiments will be limited to cases of widespread general interest. To pick a trivial example: Techdirt readers might be interested in a bot that tweets updates from privacy or copyright dockets. If those public documents were freely accessible, anybody could build a tool like that without worrying about subsidizing the ongoing PACER costs.

At a time when the president and his press secretary are calling into question the legitimacy of factual news reporting, an informed public requires more than ever access to primary sources. Moreover, they need to be confident in the integrity of those sources. Journalists reporting on court proceedings increasingly post the original source documents. Without a free and public government source file, though, most readers can't see the context of the case, and they have to trust that they're getting the full and unmodified documents in question.

The procedural stance of Washington v. Trump is unclear. A Ninth Circuit judge has made a request that both sides brief whether a larger panel should re-hear the question. The White House has issued conflicting reports about whether or not it will appeal Thursday's order to the Supreme Court. And the District Court has indicated that a new briefing schedule might be appropriate. These paths offer various levels of transparency, and it's frustrating to know my bot may not be able to keep up with, say, district court proceedings simply because of the antiquated PACER system.

Meanwhile, the issue continues to attract attention from lawmakers. The House Judiciary Committee will hold a hearing on Judicial Transparency and Ethics on Tuesday, February 14, and is expected to include testimony on PACER. Hopefully, the Committee uses this to recognize that a truly transparent judiciary requires rethinking how PACER functions.

Filed Under: access to information, courts, judicial system, pacer, records, transparency