Facebook Giving Users a Reputation Score
It sounds like a Black Mirror episode, but so far it appears to be internal. What it does, though — like the entire effort against so-called “Fake News” — is further ensconce fact-checkers in the role of arbiters of what’s true:
“One of the signals we use is how people interact with articles,” Lyons said in a follow-up email. “For example, if someone previously gave us feedback that an article was false and the article was confirmed false by a fact-checker, then we might weight that person’s future false-news feedback more than someone who indiscriminately provides false-news feedback on lots of articles, including ones that end up being rated as true.”
The score is one signal among many that the company feeds into more algorithms to help it decide which stories should be reviewed.
“I like to make the joke that, if people only reported things that were [actually] false, this job would be so easy!” said Lyons in the interview. “People often report things that they just disagree with.”
Duh. The problem is, fact checkers also rate as “false” things they disagree with. Whether it’s Carly Fiorina correctly claiming she went from secretary to CEO, or an ad correctly saying Mary Landrieu was the deciding vote on ObamaCare, or the true statement that “Zero Planned Parenthood facilities are licensed to do mammograms” — “fact checkers” love to label true statements as less than true, if they don’t like their political content. I could go on and on and on and on. Fact-checkers get things demonstrably wrong, like saying Obama never ruled out running for President (he did).
And Facebook is trying to put them in charge of your online reputation.
If these scores become public and part of your profile — if this truly becomes a Black Mirror episode — someone is going to sue Facebook for calling them untrustworthy.
I might help fund a lawsuit like that.
[Cross-posted at The Jury Talks Back.]
I can care less about Facebook but government is poised to do the same.
AZ Bob (885937) — 8/21/2018 @ 8:43 amKeeping track of how often people abuse the system by reporting things that shouldn’t be reported seems entirely legitimate, and smart.
Dave (445e97) — 8/21/2018 @ 8:48 amcrime think dontcha know,
narciso (d1f714) — 8/21/2018 @ 8:55 amBloggers too. Just take a gander at some of the nutty over-the-top accusations made here aganist Donald Trump by commenters smart and sufficiently well enough educated to know better, but still much too shocked, butt-hurt, disillusiond, and white-hot angry to even try to control their morbid tantrums.
ropelight (a560e8) — 8/21/2018 @ 9:03 amIf only people could be trusted. But they can’t.
Ingot9455 (68bf96) — 8/21/2018 @ 9:45 amSolution:
Don’t.
DCSCA (797bc0) — 8/21/2018 @ 9:48 amUse.
Facebook.
The point to the post just sailed right over your head, didn’t it?
Chuck Bartowski (bc1c71) — 8/21/2018 @ 10:20 amNo.
Dave (445e97) — 8/21/2018 @ 10:22 amTIL, Facebook was founded in 2004. How did the human race survive without it before that?
And that’s what this is all about. Zuckerberg trying to puff it up as something important and relevant, and not just another way for people with too much leisure to pass their time which is what it really is.
nk (dbc370) — 8/21/2018 @ 10:31 am6. HotAir has a post about FaceBook restoring PragerU after “accidentally” labeling them hate speech and ghosting their entire body of work. No one believes FaceBook accidentally did this of course. In order to comment though I need to sign in with FaceBook as that is the engine they use for commenting.
With so many other options why do conservatives continue to subject themselves to these companies?
Nate Ogden (223c65) — 8/21/2018 @ 10:35 amActually, the author at HotAir believes it, as do I, and probably most other sane people.
This type of de-platforming by mob reporting is precisely why it’s a good idea to keep track of who submits legitimate reports and who submits bogus ones.
Dave (445e97) — 8/21/2018 @ 10:40 amIf Twitter is the “Devil’s Dandruff”, FB is teh “Devil’s Suppository”…
Colonel Haiku (e37049) — 8/21/2018 @ 10:43 amSounds like they learned from China’s Social Credit Score. Authoritarians of a feather…
NJRob (b00189) — 8/21/2018 @ 10:59 amhttps://pjmedia.com/trending/latin-american-conservative-blogger-facebook-suspended-and-doxed-me/
Colonel Haiku (e37049) — 8/21/2018 @ 11:00 amThen you are being disingenuous. The point to Patterico’s post was that the people doing the fact-checking are introducing their own biases, and are often wrong.
Chuck Bartowski (bc1c71) — 8/21/2018 @ 11:05 amMinistry of Truth, here we come.
felipe (5b25e2) — 8/21/2018 @ 11:34 amPeople need to get off Facebook. Simple as that.
Communicate through another medium. Collect your news piecemeal. Facebook is a dangerous institution, and will only become more dangerous with time.
Leviticus (efada1) — 8/21/2018 @ 11:45 amIt needs to fail before it becomes “too big to fail.”
Leviticus (efada1) — 8/21/2018 @ 11:46 amThis is an example of why I like Wikipedia. Yes, any bigoted or partisan —- can insert lies or suppress truths at will. But anyone who is wants the truth can undo the lies. But such liars can also infest “authoritative” sources. Those lies are protected and entrenched, which is much more dangerous.
Rich Rostrom (b9a5b2) — 8/21/2018 @ 11:49 amPatterico,
Unfortunately for any such suit a “trustworthiness” rating would be entirely opinion and thus fully protected by the first amendment regardless of how atrocious the basis for the scoring. Even claims about particular “fact” items would likely be subjected to the Sullivan actual malice standard if the plaintiff were any kind of public figure (which seems likely for anyone verbose enough to get a sufficiently low trustworthiness label).
Soronel Haetir (86a46e) — 8/21/2018 @ 11:54 amAntifa, the gay mafia, metoo and many other left movements have already showed a purpensity for doxing average people and businesses. They would crash the score of any person they could if they thought it would hurt them.
Nate Ogden (223c65) — 8/21/2018 @ 12:02 pm19 on numerous occasions the liberal overlords of Wikipedia have locked edits to hide the truth. The racist NYT editoral hire being one recent example.
Nate Ogden (223c65) — 8/21/2018 @ 12:05 pmI wouldn’t call that “the point”. It’s a valid observation he makes, but the point of the post is opposition to a reputation system.
Patrick subjectively (and summarily – there is no appeal that bypasses him) reviews and sometimes moderates content here, to improve the quality of his own corner of the internet.
Thus, while I don’t want to put words in his mouth, I suspect he would acknowledge that other internet fora are within their rights to do as he does. If this blog had 1000 times as much traffic, it would have to rely on some kind of automated or semi-automated system to do what Patrick does by hand, likely driven by a reporting system (which is hardly an invention of Facebook), and Patrick would probably have to rely on the judgment of hireling or volunteer moderators whose take on any given question might be different from his own.
As simple matters of common sense and efficiency:
1) Moderation of content on high-traffic internet forums is inevitable
2) Such moderation has to deal with reports of inappropriate content from users
3) It therefore makes sense to keep track of which users report stuff that “should” be removed, and which report stuff that “shouldn’t”
The policy determining what constitutes “should” vs. “shouldn’t” is a completely separate issue. If you want to run a pro-Trump (or previously pro-GOP) site, like RedState, you can enforce that (as they have, in one form or another, for a long time). Or if you want to run a left-wing bubble site like Daily Kos, you can do that.
Filtering content based on peoples’ opinions is relatively easy. Filtering it based on truthfulness is trickier, but I don’t think Facebook’s goal is to eliminate hyperbole or mistaken assertions, but rather intentional disinformation and fraud. I may be wrong, but I don’t think any of Patrick’s examples are the types of things Facebook intends to stop.
Ultimately, just as Patrick has to decide what he will and won’t allow here, Facebook has to do the same for their site. People who disagree with the choices made at either place can complain, and/or go elsewhere. It’s really not so complicated.
Dave (445e97) — 8/21/2018 @ 12:08 pmAs described, there is no way for anyone else to affect your reputation score without gaining login access to your Facebook account.
Dave (445e97) — 8/21/2018 @ 12:12 pm11 if you allow random people to falsely label something hate speech and then suppress all their work for weeks you consider that an accident? They then protect those who made the false reports. I don’t see anything accidental in that entire process.
Further consider the leaked media matters soros notes where they game plan doing exactly that while working with the socials, to see it as accidental is just naive.
Nate Ogden (223c65) — 8/21/2018 @ 12:12 pmI think FB is also trying to persuade. They are always sending me “suggested” reading and it’s all extremely left wing. No way could I be mistaken for a leftist, so it’s advocacy!
Patricia (3363ec) — 8/21/2018 @ 12:27 pmAnd if you report something, you will note the choices never really allow you to state your objection. You just pick one, and FB decides where you fall on the political scale, and they approach you accordingly.
Patricia (3363ec) — 8/21/2018 @ 12:29 pmAll true, except that I can imagine (in rare cases) a court finding that a statement that a person is “not trustworthy” to be an implicit assertion of fact that could be defamatory. Imagine that a former employer said that about you to your prospective employer. In the Internet context, it’s true that it’s almost impossible to imagine a situation where a Facebook characterization would be actionable. But totally impossible? Not necessarily.
For example, I know Ken White, a consistent warrior against libel claims, had misgivings about Maajid Nawaz’s lawsuit against SPLC. But I supported it. Sure, it was SPLC’s opinion. But it was so over the top wrong and baseless, and their refusal to retract so irresponsible, that I applauded when they had to reach into their pockets and pay Nawaz. And I can imagine rare situations where I would applaud and support Facebook having to pony up against someone they had dragged through the mud based on clearly flawed arguments.
I agree that in the real world, this is all very close to fantasy.
Patterico (115b1f) — 8/21/2018 @ 12:29 pmIt depends on how you define abuse, and how the “keeping track” happens. My point is that this system is fraught with difficulties.
And this runs the risk of deterring reporting of (or criticism of) actually flawed stories if the Conventional Wisdom is that the stories are true. Look at all the people who said Obama was telling the truth about how you can keep your doctor, up until it was clear he had lied. But by then it was too late. Back when it mattered, taking issue with that claim might have harmed your reputation and made you “untrustworthy” in the eyes of Facebook.
Patterico (115b1f) — 8/21/2018 @ 12:32 pm“Accidental” is your word, not Facebook’s, but if a piece of software does something undesired in response to unanticipated or malicious input, I think it can be considered an accident, yes.
The opposite of “accidental,” I suppose, is “purposeful.” If they purposefully removed those specific videos, why immediately restore them after a couple hours? It makes no sense at all.
What Facebook actually said was (emphasis added):
“Mistakenly” is not the same as “accidentally”. It’s a frank admission that someone or something erred.
The HotAir author then went on to twist the statement “We’re … continuing to look into what happened…” into a claim that they didn’t know what happened, which is also not what Facebook said.
To me, “continuing to look into it” means they are working to fix whatever caused the problem. They may know perfectly well what happened, but that doesn’t mean finding and implementing a fix is trivial. It may also mean taking action against the perpetrators if it was a coordinated attempt to abuse their reporting system.
Dave (445e97) — 8/21/2018 @ 12:33 pmMy favorite quote from the linked article:
No way! They instituted a reporting system for fake news, but then it turned out that people abused it to report things they disagreed with!!! (At this point you should visualize me whacking my forehead with the base of my palm.) Who on Earth could have seen that coming?!?!?!!
Patterico (115b1f) — 8/21/2018 @ 12:46 pmRight to privacy needed to be updated in constitution. But no! Right to lifers said no privacy rights and for cheap political gains conservatives agreed. Now we have no privacy rights. Isn’t it about time we give the right to lifers a post natal abortion (at least politically speaking)
wendell (0d5882) — 8/21/2018 @ 1:20 pmAs Patterico discusses in comment 28, I think a defamation case can be won even if it is based on opinion if the opinion arises from defamatory facts that the speaker/author knows or should know are false. I agree it is not the usual defamation case but IMO it might be easier to prove in an online context, perhaps especially the Facebook scenario, since there should be a clear and discoverable history of the alleged defamation.
My thought is that there would have to be specific written reasons that Facebook’s algorithm produces and that causes it to label someone untrustworthy. If so, that would be discoverable. Also, I don’t think Facebook should be given much latitude for mistakes since its business isn’t passing judgment on peoples’ trustworthiness.
DRJ (15874d) — 8/21/2018 @ 3:01 pmThere is no reason for Facebook to make this reputation score visible, and countless reasons for them not to, so …
Worst case scenario, if you have a very low score, Facebook ignores it when you report things.
To my non-lawyerly mind, that does not sound like much of a basis for a defamation claim…
Dave (445e97) — 8/21/2018 @ 4:55 pmSure, most people won’t care but all it takes is one claim to start discovery. Does Facebook really want that?
DRJ (15874d) — 8/21/2018 @ 5:08 pmI’m not that familiar with Facebook but I think some people have Facebook pages for their businesses or jobs. A low ranking of that could impact their business, profession or livelihood. That could be treated more seriously in defamation law.
DRJ (15874d) — 8/21/2018 @ 5:24 pmFake fact check.
Kevin M (5d3e49) — 8/21/2018 @ 8:40 pmLyons said she soon realized that many people were reporting posts as false simply because they did not agree with the content.
Wait. You mean the honor system doesn’t work well on the internet?
Kevin M (5d3e49) — 8/21/2018 @ 8:41 pm30 couple hours? http://dailycaller.com/2018/08/17/prageru-facebook-google-youtube/
They were taken down on 16th, that is days. History shows they play this game with conservatives. Those with a big enough voice can get Facebook to back off, smaller voices stay banned.
This isn’t a new method of attack from the left, they have used it for years.
Nate Ogden (223c65) — 8/22/2018 @ 6:08 am36 I haven’t checked lately but Facebook might be the number one advertising channle for small business. Low ranking or banning would bankrupt them if they lost that distribution channle.
Nate Ogden (223c65) — 8/22/2018 @ 6:10 am“About 80 percent of small business owners use Facebook for marketing, according to a recent study. That makes the iconic social media platform the most popular tool for small business marketers in the digital world and beyond. It’s more popular than other social channels like Twitter and LinkedIn, online advertising networks like AdWords, and traditional advertising methods like newspapers and radio.”
Nate Ogden (223c65) — 8/22/2018 @ 6:12 amThis isn’t a new problem with Facebook, here is a story from 2015 and it goes back older than this;
https://growth.org/blog/a-competitor-bought-200-1-star-reviews-for-our-facebook-page-here-s-our-story-1
Nate Ogden (223c65) — 8/22/2018 @ 6:16 am““You know, I’d rather have fake news like CNN,” he said, waving toward the bank of reporters in front of him. “I would rather have fake news — it’s true — than have anybody, including liberals, socialists, anything, than have anybody stopped and censored.”
Trump’s mention of CNN, a favorite foil, drew some groans. But Trump brushed them aside.
“We gotta live with it. We gotta get used to it,” he said. “We gotta live with fake news. There’s too many sources. Every one of us is sort of like a newspaper. You have Twitter. You have whatever you have. Facebook.”
Trump continued.
“You can’t have censorship. You can’t pick one person and say, ‘Well, we don’t like what he’s been saying. He’s out.’ So, we’ll live with fake news,” he said. “I mean, I hate to say it. But we have no choice. Because that’s by far, the better alternative. We can’t have people saying, ‘censorship.’ Because you know what? It could turn around. It could be them next … We believe in the right of Americans to speak their minds.”
Trump the First Amendment champion seems like an oxymoron to his critics who constantly insinuate that America is just a step away from jailing journalists, shutting down opposition newspapers and yanking television licenses.
https://www.lifezette.com/2018/08/trump-a-first-amendment-champion-its-true/
Tell me again who’s a bigger threat to freedom of speech…..Trump, the media, academia or social media. I’d say he’s a distant fourth.
harkin (fabd28) — 8/22/2018 @ 9:10 amQ. In 2003, after ten years practicing law in LA, you started Patterico’s Pontifications, your blog, as a way to give voice to your conservative opinions in a way that didn’t involve the mass sending of emails. You‘ve spent your whole adult life in extremely liberal towns, something that does raise the question of where those views came from. Are they the product of where you grew up? How you were raised? Is it just that you’re a freethinker?
Pharmacycenter (3b05ab) — 8/31/2018 @ 12:26 am