We Shouldn't Want Internet Giants Deciding Who To Silence; But They Should Let Users Decide Who To Hear

from the rethinking-moderation dept

A few weeks back I wrote a big piece on internet platforms and their controversial content moderation efforts. As I've pointed out more recently, part of the reason why what they do is so bad is it is literally impossible to do this well at the scale they do things at. That is, even if they can reach 99% accuracy, given the amount of content on these sites, it's still going to take down a ton of legitimate stuff, while leaving up an awful lot of awful stuff. This doesn't mean they shouldn't do anything -- but my own proposal is for them to shift the way they think about this issue entirely, and move the moderation out from the center to the ends. Let third parties create their own filters/rules and allow anyone else to not just use them, but to adjust and modify and reshare them as well. Then allow the users to not just "opt-in" to the kind of experience they want, but allow them to further tweak it to their own liking as well.

I've seen some pushback on this idea, but it seems much more viable than the alternatives of "do nothing at all" (which just leads to platforms overwhelmed with spam, trolls and hatred), and continue to focus on a centralized moderation system. There have been a number of articles recently that have done a nice job highlighting the problems of having Silicon Valley companies decide who shall speak and who shall not. EFF's Jilian York highlights the problems that occur when there's no accountability, even if platforms have every legal right to kick people off their platforms.

This is one major reason why, historically, so many have fought for freedom of expression: The idea that a given authority could ever be neutral or fair in creating or applying rules about speech is one that gives many pause. In Europe’s democracies, we nevertheless accept that there will be some restrictions – acceptable within the framework of the Universal Declaration of Human Rights and intended to prevent real harm. And, most importantly, decided upon by democratically-elected representatives.

When it comes to private censorship, of course, that isn’t the case. Policies are created by executives, sometimes with additional consultations with external experts, but are nonetheless top-down and authoritarian in nature. And so, when Twitter makes a decision about what constitutes ‘healthy public conversation’ or a ‘bad-faith actor,’ we should question those definitions and how those decisions are made, even when we agree with them.

We should push them to be transparent about how their policies are created, how they moderate content using machines or human labor, and we should ensure that users have a path for recourse when decisions are made that contradict a given set of rules (a problem which happens all too often).

Jillian's colleague at EFF, David Greene, also had an excellent piece in the Washington Post about how having just a few giant companies decide these things should worry us:

We should be extremely careful before rushing to embrace an Internet that is moderated by a few private companies by default, one where the platforms that control so much public discourse routinely remove posts and deactivate accounts because of objections to the content. Once systems like content moderation become the norm, those in power inevitably exploit them. Time and time again, platforms have capitulated to censorship demands from authoritarian regimes, and powerful actors have manipulated flagging procedures to effectively censor their political opponents. Given this practical reality, and the sad history of political censorship in the United States, let's not cheer one decision that we might agree with.

Even beyond content moderation's vulnerability to censorship, the moderating process itself, whether undertaken by humans or, increasingly, by software using machine-learning algorithms, is extremely difficult. Awful mistakes are commonplace, and rules are applied unevenly. Company executives regularly reshape their rules in response to governmental and other pressure, and they do so without significant input from the public. Ambiguous "community standards" result in the removal of some content deemed to have violated the rules, while content that seems equally offensive is okay.

Vera Eidelman, of the ACLU similarly warns of the pressures that are increasingly put on tech companies that will inevitably lead to the silencing of marginalized voices:

Given the enormous amount of speech uploaded every day to Facebook’s platform, attempting to filter out “bad” speech is a nearly impossible task. The use of algorithms and other artificial intelligence to try to deal with the volume is only likely to exacerbate the problem.

If Facebook gives itself broader censorship powers, it will inevitably take down important speech and silence already marginalized voices. We’ve seen this before. Last year, when activists of ...anchor markup--p-anchor" rel="nofollow noopener noopener" target="_blank">experiences of police violence, Facebook chose to shut down their livestreams. The ACLU’s own Facebook post about censorship of a public statue was also inappropriately censored by Facebook.

Facebook has shown us that it does a bad job of moderating “hateful” or “offensive” posts, even when its intentions are good. Facebook will do no better at serving as the arbiter of truth versus misinformation, and we should remain wary of its power to deprioritize certain posts or to moderate content in other ways that fall short of censorship.

Finally, over at Rolling Stone, Matt Taibbi makes a similar point. What starts out as kicking out people we generally all agree are awful people, leads to places we probably won't like in the end:

Now that we’ve opened the door for ordinary users, politicians, ex-security-state creeps, foreign governments and companies like Raytheon to influence the removal of content, the future is obvious: an endless merry-go-round of political tattling, in which each tribe will push for bans of political enemies.

In about 10 minutes, someone will start arguing that Alex Jones is not so different from, say, millennial conservative Ben Shapiro, and demand his removal. That will be followed by calls from furious conservatives to wipe out the Torch Network or Anti-Fascist News, with Jacobin on the way.

We’ve already seen Facebook overcompensate when faced with complaints of anti-conservative bias. Assuming this continues, “community standards” will turn into a ceaseless parody of Cold War spy trades: one of ours for one of yours.

This is the nuance people are missing. It’s not that people like Jones shouldn’t be punished; it’s the means of punishment that has changed radically.

This is why I think it's so important that the framework be shifted. People have long pointed out that "just because you have free speech doesn't mean I need to listen," but the way social media networks are constructed, it's not always so easy not to listen. The very limited "block / mute" toolset that Twitter provides is not nearly enough. The more platforms can push the moderation decision making out to the ends of the network, including by allowing third parties to create different "views" into those networks, the better off we are. It's no longer the internet giants making these decisions. In fact, it increases "competition" on the moderation side itself, while also increasing the transparency with which such systems operate.

So, really, it's time we stopped focusing on who the platforms should silence, and give more power to help the end users decide who they wish to hear.

Filed Under: censorship, centralizataion, decentralization, filters, free speech, human rights, intermediary liability, platforms, silence, social media
Companies: facebook, google, twitter

Cities Make Public Their Accomplices In Wrongdoing By Issuing Settlements Tied To Gag Orders

from the buy-us-a-halo dept

It's not enough that lawsuit settlements for police misconduct, brutality or officer-involved-deaths come attached to "no admission of wrongdoing" statements. In far too many cases, they also come with stipulations forbidding recipients from making public statements about the lawsuit or its allegations.

When city officials here settle police-misconduct claims, they usually bar the person who alleged mistreatment from speaking publicly about the case. Violating that restriction can result in a settlement being cut in half.

The city [Baltimore] last year withheld $31,500 of one woman’s $63,000 payment after she posted messages about her case on a newspaper website, for instance.

The Wall Street Journal points out that Baltimore is somewhat unique in its use of confidentiality agreements in settlement agreements. But it is by no means the only government entity to do so. The mother of Darrien Hunt, who was killed by Utah police officers last year, recently rejected a $900,000 settlement from Saratoga Springs over the inclusion of a gag order.

"To me it was a gag order [that said], 'Here's hush money, don't ever say Darrien's name again,'" Susan Hunt, mother of Darrien, told Utah's KSL news about turning down the $900,000 settlement the city offered in response to her wrongful death lawsuit.

"My biggest concern is for the truth to be told," Susan said.

If the locality isn't interested in forcing gag orders on settlement recipients, the law enforcement agencies at the center of these lawsuits will sometimes help out by arresting plaintiffs and using pending charges as leverage against further public disclosure.

Criminal charges are commonly used to thwart the possibility of lawsuits, said attorney Marcus Sidoti, who has handled multiple civil rights lawsuits the city settled in recent years.

For instance, Judi Patrizi was arrested in 2007 and charged with obstructing official business after an officer said she impeded an assault investigation at Bounce Nightclub.

In a report, an officer described Patrizi pointing in his face and swinging her arm away from him -- an account later disproven by video surveillance at the club. The charges were later dropped and Patrizi sued the city saying she was maliciously prosecuted after the unlawful arrest. The city settled her case for $87,000.

In some cases, those who claimed police brutalized or shot them were offered plea deals that would reduce or drop charges against them only if they agreed not to sue the officers or the city -- a move some call unethical.

The Cleveland.com article lists a handful of other cases where lawsuit plaintiffs were arrested and charged with criminal activity after filing brutality complaints. In almost every case, charges were dropped or (in the one felony case) the grand jury did not return an indictment.

This sort of thing isn't strictly limited to lawsuits filed over police misconduct. These tactics also used to protect other government employees and officials from public scrutiny.

The Beecher School District [Flint, MI] paid nearly $250,000 to avoid two lawsuits over alleged sexual misconduct by a former public school administrator.

But no lawsuit was ever filed, so taxpayers did not have easy access to this information because of a state law that allows public bodies to enter into non-disclosure clauses that bar either side from discussing specifics of a case.

Unless allowed to be filed under seal, court documents are, by default, public records. The best way to keep the public from viewing public records is to create as few of them as possible -- hence the quick settlements and the accompanying gag orders. If done swiftly enough, no documents will make their way onto court dockets. Add in a bit more enforced secrecy, and even FOIA requesters won't know what to look for… or if anything even exists.

Court records are created when a lawsuit is filed and eventually dismissed, which could tip a potential FOIA user off to a settlement.

However, a keen eye may be needed to realize when a government agency agreed to a settlement with a non-disclosure clause to prevent a lawsuit from ever being filed, such as the situation in Beecher.

While confidentiality clauses are common to lawsuit settlements between private parties, the use of them in the public sphere is not nearly as acceptable. These agreements force the public to not only pay for the misdeeds of their public servants but shield them from additional scrutiny. Settlements are often offered by government agencies because it guarantees them something they'd can't obtain through a jury trial: the ability to exonerate themselves. It also heads off the chance that they'll have to hand over other incriminating information in response to discovery requests.

Nearly every lawsuit settlement contains clauses indicating that the payout is not an admission of wrongdoing. Government agencies are continually buying "innocence"... using money they've collected from citizens who have little say in how their tax money is spent. Whether it's a confidentiality agreement or a speedy settlement issued in an attempt to head off a lawsuit, the end result is the same: more secrecy, less accountability and the continued diminishing of any deterrent effect these civil suits are supposed to have.

Filed Under: cities, gag orders, police brutality, police misconduct, settlements, silence

Timeshare Company Sundance Vacations Accused Of Forging Court Order To Delete Discussion Of Its Practices

from the getting-into-felony-territory dept

We've seen all sorts of bizarre and questionable attempts by companies to silence criticism or content they just don't like. Bogus copyright, trademark and defamation claims are pretty standard these days. There's also just general "begging" or the random "tortious interference" arguments. But according to Metafilter's Matt Haughey, timeshare company Sundance Vacations may have taken things to insane new levels: forging a bogus court order to try to get Metafilter to remove a years-old thread.

It started nearly two years ago when Sundance Vacations sent Haughey an email, asking him to remove this 2010 thread in which someone asked about Sundance Vacations marketing process -- starting off with "am I about to get screwed by Sundance Vacactions?" leading to a rather tame discussion about the sales tactics of the timeshare industry, and ways to deal with the dreaded "informational meeting" (really a hard sales pitch). The email included an attachment of a "court order" against someone who was supposedly running a "Sundance Vacations Protest site" barring them from speaking negatively about the company online. I don't see how such a court order is actually legal given the whole First Amendment, but we'll leave that aside for the moment. Haughey pointed out that the person named in the court order was clearly not the person who posted the Metafilter thread, and everyone went on with their lives.

Until a few days ago, when Sundance Vacations sent a new email to Haughey, talking about a new court order, apparently against the person running this Boycott Sundance Vacations Facebook group. Bizarrely, the "court order" listed the Metafilter page in the document. Haughey posted the court order as an image:

As Haughey noted, there are some oddities in that court order. For example, the signature page has a "signature" of "Sundance Vacations" (actually signed as such) rather than the name of an individual, as would actually be required. The part at the top listing the plaintiff and defendant has "Sundance Vacations" in a different size and font than everything else -- suggesting a cheap insertion. Even reading through the document, much of it looks to be about a basic restraining order between two individuals rather than a company. The order mentions "Defendant Friedman" despite the fact there is no Defendant listed with the name Friedman.

On top of that, the Order lists out five URLs which it claims are defamatory and says that plaintiffs can use the order to get those articles removed or delisted. While it's not unheard of for local clueless courts to make such rulings, you can't actually make such an order, because it violates Section 230 of the CDA. Thus even if it were true, it wouldn't be enforceable. But it's not actually true. After questioning it, Haughey called the actual court:

Today (Tuesday) I called a clerk in the Hinds County Chancery Court office. They asked me to fax them a copy of the court order so they could verify the document. I did as requested and a few hours later got a call back from the office saying it was not a real document from their court. The case numbers on the first page are from an unrelated case that took place last year. The clerk said they found a case from August 21, 2014 that used similar language but had different plaintiffs and defendants, but the same lawyers on page 3. In their opinion, it seemed someone grabbed a PDF from a different case and copy/pasted new details to it before sending it on to me.

As Haughey notes, forging court documents is a felony. It kind of makes you wonder, if a company will commit felonies to hide some (very mild) online criticism, what kind of crap will it do in its regular business of trying to do high pressure sales to buy into their timeshare?

Ken "Popehat" White points out that Sundance Vacations is now denying having sent the email, though it's pretty difficult to think who else might be sending such an email.

Filed Under: censorship, court order, forgery, matt haughey, silence, time shares, timeshares
Companies: metafilter, sundance vacations

Pressure Mounts Against Telcos To 'Fess Up About Their Involvement In NSA Surveillance

from the open-up dept

Ever since the Snowden leaks began, there's been a clear dichotomy in terms of how different industries have reacted. The various big internet companies, which were named early on as participants in the PRISM program, have been quite vocal (sometimes to profane levels) that they were not willing participants in most of these programs, and are currently involved in an important lawsuit arguing that they have a First Amendment right to reveal how much info they actually share with the government. While those eventual revelations (and they almost certainly will come out, either legally or through leaks) may reveal certain companies were more complicit than others, by all indications, the various internet companies have been very willing to fight the government over this.

On the other side, you've got the telcos -- mainly AT&T and Verizon (but Sprint and some others as well). What do you have there? Total, deafening silence. Seriously. They've said nothing about any of this, despite increasing evidence that they not only are happy and willing participants in the NSA's efforts to spy on everyone, but that they've volunteered to hand over more data than required. Furthermore, it's quite clear now that they've basically let the NSA put taps directly on the internet backbone, by which they can record just about anything, while the internet companies have (from all appearances to date) limited information sharing only to a specific segment of information following a specific court order (which probably doesn't have enough oversight, but that's a different issue).

The contrast here is really striking. From all revealed info to date, the telcos' info sharing with the government is much more massive and has significantly more privacy implications than anything done by the internet companies. And yet it's the internet companies that are both speaking out against this and challenging some of it legally (though some of us still think they should go further -- but, thankfully, we've been hearing significant and credible buzz that much more is on the way). The telcos? Absolutely nothing. Well, except for a Verizon VP mocking the internet companies for pushing for transparency.

Otherwise? Dead silence.

When the internet companies reached out to the telcos about co-signing their letter to the government pushing for more surveillance, the telcos refused. Over the last few months, we've seen pretty much every major internet company release a transparency report, including showing at least some data on government requests for info. The telcos have never released such a thing.

Kevin Bankston, who was instrumental in helping to coordinate that original letter, is now calling out the telcos on their shameful silence. It seems that the telcos are hoping this whole thing blows over, in part because a true transparency report from them would likely reveal just how complicit the telcos were in handing over your private info to governments, often with little to no oversight. In fact, Bankston notes that AT&T has been quietly lobbying against legislative efforts to increase transparency.

What are the telcos afraid of? The answer is pretty obvious.

However, they might not be able to wait this out and hide forever. The ACLU of Northern California has now filed a shareholder proposal with both At&T and Verizon demanding that they file a transparency report concerning their cooperation with government surveillance.

Customer trust is critical for any business, but especially for Internet and telecommunications companies that gather personal data concerning and affecting the lives of hundreds of millions of people in the U.S. and around the world. In an effort to help rebuild consumer trust, major Internet companies including Google, Microsoft, Twitter, LinkedIn, Facebook and Yahoo! have issued transparency reports with information on government requests; AT&T and Verizon have not. Companies, including Google and Microsoft, have also filed in court seeking authorization to disclose further information to the public; AT&T and Verizon have not.

Privacy is fundamental to democracy and free expression -- and transparency is essential if individuals and businesses are to make informed decisions regarding their personal information. AT&T and Verizon must comply with legal obligations imposed by the Patriot Act and other laws. But these companies have no good excuse for staying silent and failing to provide information about how often customer information is being shared with the government. To the contrary, staying silent as other industry leaders release transparency reports and take steps to reinforce a genuine commitment to privacy, makes it appear that these companies have something to hide and presents serious financial and reputational risks. Consumers prefer companies whose information practices they know and can trust. It is already estimated that the risks of surveillance and lack of trust could cost the U.S cloud computing industry $21 billion to $35 billion in foreign business over the next three years. The Chief Privacy Officers at AT&T and Verizon have praised transparency as a goal, but it is time to back up those statements with action by releasing transparency reports.

Of course, thanks to other bad government policies, AT&T and Verizon have market dominant positions, such that there's often no actual customer choice -- which is part of why they can get away with this silence. Hopefully shareholders of both companies will stand up and make those companies reveal some basic transparency, even if it will embarrass the two companies. And, if they're really embarrassed by what the data will show, perhaps that's a sign that they shouldn't be doing it in the first place.

Filed Under: silence, surveillance, telcos, transparency
Companies: aclu, at&t, verizon