Facebook: Why a Safety Center, not a 'panic button'

The Facebook news in the US today was its new expanded Safety Center. The news in Britain was that Facebook "STILL refuses to install [a] 'panic button'" on its pages, as the UK's Daily Mail put it. However, Facebook also announced today that its UK users will "now be able to report unwanted or suspicious contact directly to CEOP [the UK's Child Exploitation & Online Protection Center] and other leading safety and child protection organizations via its own reporting system," as CNN reported, so CEOP has come very close to getting its wish.

But this "panic button" concept is really problematic – and not just because of the word "panic," which suggests brains in crisis mode, with all rational thought switched off. Here's why it's problematic:

A single reporting mechanism doesn't cut it. In the offline world, we call 911 (or in the UK, 999) about crimes and medical emergencies. But the social Web – especially a fairly basic social utility like Facebook – is a mirror of its users' social lives and networks, of a full spectrum of behaviors, mostly good and, when bad, definitely not just criminal bad behavior. So if you just consider the really negative behavior that might lead to an abuse report, research shows that it's bullying, not predation, that would get reported far more often. Is law enforcement designed to deal with noncriminal but bad adolescent behavior? Fortunately, the new system Facebook put in place sends only reports of criminal behavior to CEOP.

Would a "panic button" have helped Ashleigh Hall? CEOP reportedly has said that the British teen whose murderer was convicted last month (see The Guardian) may have lived if such a button had been in place in Facebook. Ashleigh was reportedly communicating with someone who she thought was a boy, and fear didn't seem to be involved at the time of that FB communication. It isn't a factor when a child is being "groomed" online (see this).

A crime not involving panic. Ashleigh's case was far from typical of Net-related sex crimes. Presenting research from law enforcement files on Net-related child sex crime cases, David Finkelor, director of the University of New Hampshire's Crimes Against Children Research Center (CACRC), said in Washington in 2007, "These are not mostly violent sex crimes ... they are criminal seductions that take advantage of common teenage vulnerabilities" and are characterized as statutory rape. "In 73% of the crimes," he continued, "the youth go to meet the offender on multiple occasions for multiple sexual encounters. The law enforcement investigators described the victims as being in love with or feeling a close friendship for the offenders in half the cases that they investigated" (see this for his "Jenna" profile). Panic buttons in social sites do nothing to mitigate this problem.

Largely the wrong location. The Internet, that is. It's important to remember that the vast majority of sexual abusers of children are people they know in real life, not strangers they meet online, much less predators trolling the social Web. In a much-anticipated 2009 update of its research on Internet predators, the CACRC reported, "There was no evidence that online predators were stalking or abducting unsuspecting victims based on information they posted at social networking sites," and we've seen no reports in this country of convicted sex offenders being arrested for violating parole agreements by contacting minors in social sites. [As for non-Internet-related sex crimes, University of California, Berkeley, law professor Franklin Zimring was recently quoted as saying people are "more likely to get struck by lightning than to get raped and murdered by a stranger," The Press-Enterprise in southern California reported last month.]

Facebook actually tested a similar proposal made by New Jersey's then-attorney general, Anne Milgram, a couple of years ago: the test of a "Report Abuse!" icon involving "at least 1.5 million randomly-selected page impressions" for nearly a year (see MSNBC). What FB found (after running the test longer than MSNBC reported up front, a spokesperson told me today) was that the number of abuse reports was "significantly lower" when there was a special icon in a different location from the rest of the reporting links on a page. Third-party buttons and graphics "intimidate and confuse people," Facebook's European policy director Richard Allen told RedOrbit.com. "We think our simple text link, which gives people the option to report abuse to CEOP as well as to the Facebook team, is a far more effective solution."

A button is not enough. Even the host of Britain's "To Catch a Paedophile," Mark Williams-Thomas, a child-protection expert and former detective, said that "the much called-for report-it button alone does not make using social networking sites any safer, but a coordinated approach providing the additional reporting to CEOP is clearly worthwhile, as is a dedicated phone line for law enforcement." The dedicated line he's referring to is similar to one Facebook has for US law enforcement and part of the safety package it announced this week, including the Safety Center mentioned above and 1 billion public-service ad impressions in the site (which CEOP called "a 5 million-pound [or $7.7 million] investment in education and awareness" in its press release, which was not yet online as of this writing).

Having said all that, everybody can thank all parties to this agreement for an important pilot test we all need to watch. Not before in history has there been a service playing host to the visual socializing of 400 million users in multiple countries, much less developing some sort of reporting system for when something in all that socializing goes wrong – the online version of dial-911 or -999 (UK) but for many more kinds of "wrong" (not just the criminal kind). I don't know about CEOP, but our NCMEC has a CyberTipline.com, a sort of online 911 service, and it still tells people to call their local 911 service in emergencies. Physical proximity is still and always will be a factor when people need help – so just what is the role of a global online service, here? We all – social-Web companies, their users of all ages, parents, educators, law enforcement, risk prevention practitioners, psychologists, etc. – need to figure this out together. It just won't work if the onus is placed only on companies', or law enforcement's, or policymakers' shoulders – not in a highly participatory, grassroots-driven media environment.

But for heaven's sake – or even better, for youth's sake – let's please take the "panic" out of this whole important test. It simply doesn't lend itself to the calm, mutually respectful conversations that help youth develop the critical thinking that protects on the social Web. We had our predator panic on this side of the pond starting in 2006. At the Family Online Safety Institute's annual conference in Washington last fall, the Net-safety field declared it over with a strong consensus that scary messaging is not productive. Why? Because it makes young people less inclined to want to come to us for help. They tend to get as far away as possible from scared, overreacting adults; find workarounds that are readily available to them; and then leave us out of the equation right when loving, steady parent-child communication is most needed. The other reason is, even the research shows fear tactics don't work (see "Let's not create a cyberbullying panic" at CNET).

[Disclosure: Facebook is a supporter of a nonprofit project I help run, ConnectSafely.org, but I so hope you've seen in the above that that's not why I've blogged about this issue.]

Related links

"Why technopanics are bad"

More on why fear tactics don't work

The US's perfect storm of parental concern in 2006: created by MySpace's exponential growth, adults not understanding social networking, news media hyperbole, Dateline's "To Catch a Predator," and a mid-term election (hinted at but not fully described in this Business Week article about MySpace's safety efforts of that time). Now – even after the sanity of the Byron Review and ensuing government-industry-NGO cooperation – the UK is experiencing its own perfect storm, with an election, a tragic crime story, a "To Catch a Paedophile" show, Facebook's rapid growth, and continuing cognitive dissonance over social media. Storms are destructive; these national-level storms in a new-media climate distract us from calmly sorting through complex problems and finding real solutions.

The Congressional Internet Caucus Advisory Committee, which put on the Capitol Hill event where Dr. Finkelhor spoke in 2007