Hacker News Reader

Nutrition labels treat the symptom, not the cause.

IMO the root cause is a neglected and underfunded education system, that does not adequately prepare people for the levels of psychological manipulation the will experience in the form of marketing.

Sure, a nutrition label might help, but it's kind of passing the buck - "We don't need to fix our education system, people just need to be able to count to five so they know how many truth stars this article has, then they can make up their own mind!"

Think of grandma, man. Make life easy for grandma.

I get what you are saying, but remember that the big, evil, powerful people in the world who are manipulating this system are mostly grandparents or old enough to be grandparents. I'm pretty sure it's not specific to age, but cognitive education.

Yes.. even if you figure out a perpetually unbiased system to do the fact-checking and labeling, uneducated people will still be manipulated into distrusting the labels. Fact checking organizations already exist and people still think Fox is a beacon of truth.

A major target of a lot of these efforts are white men over 40. If your prescription is a better education system, it will take years before it starts making a difference.

It's not at an article level, it's at an institution level. If an institution is known to publish misleading article with half-truths or non-truths, a rating which indicates that (and links to a list with examples) would be beneficial. Institutions that want to be seen as credible will strive to be accurate and fix any mistakes.

There are definitely problems with this, but the idea that we just need to fix our education system seems like just as much of a cop-out to me. The problem is hard, it plays into human cultural and cognitive pitfalls, and affects smart, well educated people too. It's not just a matter of education.

I disagree that education is a cop-out. Telling people what is credible and what is not is part if the mess that we're in now. If I can stand up a fancy looking "news" site, write my opinion as fact, and have people believe me, that is a problem of our education system.

That principle of freedom of information is what the internet is built on. A large number of retweets does not make something factual.

It's not about telling people what's credible and what's not, it's about telling them who has been caught being factually incorrect more than others, linking to additional info, and letting people use that to help them make their own choices. A centralized authority is less likely to be partisan, and whatever shortcomings it has will be widely publicized. Think of it like a credit bureau or a sock/bond ratings company, but without some of the perverse incentives that apply there, and kept from getting too partisan by government oversight.

We investigate and prosecute people that commit fraud. I don't see why we can't at least note where news agencies are failing in a way that's harmful to the populace. It's not perfect, and it may fail, but it might also give us another useful tool to signal the quality of an information source.

Education might be the root cause and it might be neglected but in the US it certainly isn't underfunded.

Source: http://www.politifact.com/florida/statements/2015/apr/21/jeb...

Sometimes I wonder if not preparing students for psychological manipulation is actually a feature and not a bug of the educational system.

`nutrition labels` are already kind of subjective as we keep finding out just what information is actually most important to our health and revising the labels.

Having a "truth vs fiction" rating next to news sites isn't as easy as you think. Sure, some things are easy to qualify - but when it comes to wars, motivations, scandals - many common stories will always be left with two equally disputed sides of the story.

This means the site that supports the governments verdict will when the vote.

This is not an easy task, but it seems like one the Tech Giants would be able to tackle with their brainpower, resource, intelligence, outreach, and general sense of impending doom (read as "morality"). Or would it be best to focus more on emoji.

I assure you that outside, people are still mostly normal.

I have visited this land, and spoken to its inhabitants, most everyone is pretty chill and friendly.

Nutrion labels - would you consider them harmful? I see them as a net positive.

> we keep finding out just what information is actually most important to our health and revising the labels.

As long as we can, and do, revise the labels, that's exactly what we want. Not only would trying to determine the absolute best things to put on the label be time consuming and take forever to accomplish, we would still find more stuff to change later. Perfect is definitely the enemy of good in this case.

Some of the fact checking websites are just blatant partisan punditry disguised as 'fact-checking'. (Verrit is worst offender: labeling something as 'True' if someone said something verbatim, even if it is not a factual assertion).

Even the ones with decent records often display bias, ex. when someone they like says something where the thesis is correct but a supporting fact is incorrect, they'll mark it as "mostly true". But when someone they don't like does the same, they'll mark it as "mostly false".

That's the same as the lion telling the tamer how safe it is in the cage. Needs to be independent firm or NGO or company internal team. Ideas and solutions please!

Verrit verifies quotations, not content. They don't say whether you should trust the source; that's up to you.

I haven’t watched the testimony, but I don’t think the suggestion is for the government to make the judgements for the labels - at least, that sounds like an epically terrible idea, especially (but not only) given the current president’s predilection for calling any news he doesn’t like “fake”. Rather, I think social media platforms would be expected to do that themselves, maybe by contracting with external fact checking sites, like Facebook is already doing in order to label specific articles (as opposed to entire sites) “disputed”.

But yes, creating any centralized pseudo-authority on what counts as truth - even a private one committed to neutrality and limited to a specific social media platform - carries significant risks. I think some existing sites, such as Politifact, currently do a pretty good job of distinguishing “this is false” from “this is political spin I disagree with”. But they’re not perfect - they clearly have a bias, like everyone else - and other fact checkers tend to be somewhat worse, from what I’ve seen, ranging from Snopes to fact-checking articles in WaPo, NYT, etc.

Yes but there IS a legal definition for reporting vs opinion and it is easy enough to put an article in Column Op-Ed instead of Column New Info

Many of the most important issues (e.g. current crisis in Rohingya) are in environments where there is no such thing as absolute truth.

To pick one example, in a conflict, it's very rare that any two independent sources can even agree on the number of casualties (much less on whose side they were, what the cause of death was, etc), and what the western world tends to treat as some of the most reputable sources explicitly ignore any reports not written in English. Which is obviously not useful when many of these conflicts occur in regions where English is not a primary language.

We can define the boundary between fact and opinion all we want, but in the real world it's very rarely so black and white. Sure, pieces that are merely commentary/opinion should be labelled as such, but in all reputable news outlets already do this. I've yet to see a single useful solution to this issue that stands up when there's simply no practical way of accurately establishing what the truth really is.

Well, I don't think it matters which "column" an entire article goes in. Op-eds usually contain a variety of factual statements which are used to support the subjective parts, and those statements still need to be true. Indeed, most 'less professional' websites make no distinction between "straight reporting" and opinion - in effect, all their articles are op-eds - but it's precisely those sites that are most likely to spread false information and thus most need to be fact-checked.

It's true that courts need to distinguish fact from non-actionable opinion in defamation/libel cases (at the level of individual statements, not entire articles) - not to mention that almost all court decisions rely on the judge evaluating competing factual claims, to some extent, even though that's theoretically the job of the jury. The judicial branch does do at least a somewhat better job at staying neutral, serious, fact-based, compared to other branches - but far, far from ideal. I suppose the idea of a government fact-checking agency sounds slightly less terrible if you make it part of that branch (though that would be an odd arrangement, since it would seem to fall under the broad umbrella of 'regulatory agency' that you usually see in the executive branch). But still terrible IMO.

After all, US defamation law grants authors an enormous amount of leeway when it comes to issues of public concern, and that's by design, to prevent it from being used to censor valuable speech even given biased judges. I'll concede that true "fake news" is often a straightforward matter of totally made-up claims, to the point that the people named probably could sue the publisher for defamation and win, under the existing system - albeit at the cost of money, effort, and potentially triggering the Streisand effect. If that's all this hypothetical agency cared about, it could operate relatively safely. But that wouldn't be good enough. There's a broad spectrum of claims that aren't clear-cut enough to be actionable - and are protected by the First Amendment - yet are still misleading enough to merit poor ratings from fact checkers. The agency could try to apply a stricter truth standard in its advisory labels than that used in defamation law, arguing that the First Amendment is less of a concern when speech isn't actually being censored. It'd still face constitutional challenges, but it might survive them.

But "stricter truth standard" is just another way of saying the agency needs to be able to make subjective decisions. Not too subjective - nowhere near the caricature "disagree = false" that I've seen from some critics of fact checkers on HN (never mind the president) - but not 100% objective either. I think it's important for fact checkers to be able to exercise such discretion, because politicians are very, very skilled at muddying the waters. But I also think I'd prefer to keep that power out of the hands of the government, no matter which part of it or in what form.

> Yes but there IS a legal definition for reporting vs opinion

Please cite the applicable law. (There's clearly a legal difference—relevant for, e.g., defamation claims—between fact and opinion claims, but that operates at a much smaller unit than an entire article, which can contain both kinds of claims.)

> creating any centralized pseudo-authority on what counts as truth - even a private one committed to neutrality and limited to a specific social media platform - carries significant risks. I think some existing sites, such as Politifact, currently do a pretty good job of distinguishing “this is false” from “this is political spin I disagree with”. But they’re not perfect - they clearly have a bias, like everyone else - and other fact checkers tend to be somewhat worse, from what I’ve seen, ranging from Snopes to fact-checking articles in WaPo, NYT, etc.

This is why I think the concept is so hopelessly naive. News sources are already supposed to be factual, but there's immense difficulty in finding a single, high quality source, and strong incentives pushing them towards developing sources that serve the beliefs that people want to hold.

Why in earth would an extra layer of misdirection help at all? All you'd end up with is a replica of the news ecosystem we have now.

And who determines what's fact and fiction?

Is it the news agency? So they just have to say "We 100% believe aliens live on the dark side of the moon" and it gets counted as fact?

Or is it the government? And anything the government doesn't like is fake news.. and everything they do like is fact?

I think this is much more difficult than you're assuming. Impossible really.

Independent fact checking firms exist, I don't think it is a stretch to distill their info into an easily comprehended badge icon. Google, according to their Counsel, is already giving anti terror groups flagging software.

Which independent fact-checking firms do you trust?

Currently just Snopes

So you propose Tampa Bay Times Politifact, and The Annenberg Public Policy Center's FactCheck.org become arbiters of truth?

Have you actually read many of their articles? A lot of it depends on judgement calls of the person doing the fact checking. Leeway is given or taken by the fact checker.. and that amount varies by who is doing the checking. (And you can’t eliminate this because human language is so flexible that it will always require a judgement call to be made in all but the most blatant lies.)

There are plenty of examples.. here's one:

Politifact rates this as mostly true: Bernie Sanders says 'real unemployment' rate for African American youth is 51 percent http://www.politifact.com/truth-o-meter/statements/2015/jul/...

Politifact rates this as mostly false: Trump misleadingly puts black youth unemployment rate at 59 percent http://www.politifact.com/virginia/statements/2016/jun/20/do...

The rate is nowhere near that high according to the BLS: https://www.bls.gov/web/empsit/cpsee_e16.htm

If politifact can't call out Bernie's 51%, like they did for Trump.. why would they get this trusted position you seem to think they are qualified for? Politifact is run by human beings. Their judgement shouldn't be substituted for the judgement of others. You can't outsource critical thinking.

+100

One was "accurate but need clarification"; the other "exaggerating and misleading".

> Sanders said that for African-Americans between the ages of 17 and 20, "the real unemployment rate … is 51 percent." His terminology was off, but the numbers he used check out, and his general point was correct -- that in an apples-to-apples comparison, African-American youth have significantly worse prospects in the job market than either Hispanics or whites do. The statement is accurate but needs clarification or additional information, so we rate it Mostly True.

> Trump says the unemployment rate for black youths is 59 percent. The unemployment rate is a widely used term with a specific definition: It refers to the percentage of jobless people in the workforce who are actively seeking employment. In May, the unemployment rate for blacks ages 16 to 24 was 18.7 percent, or less than one-third of Trump’s claim...Clearly, black youths have a harder time finding work than whites. But Trump exaggerates the issue through his misleading use of statistics. We rate his statement Mostly False.

None are free from bias.

The solution to oligarchy is democracy, not monarchy.

The solution to bad speech is more speech, not less speech.

Did you actual read the differences, which are bigger than than simply a number. In one case, the number is backed up by evidence which is provided by the campaign. Politifact can then verify that the number isn't just made up or based on nothing. It's mostly true if you understand the context.

As for Trump, they didn't provide anything to back up the statements. They had to dig to find where he got the information, and nothing they found backs up what he said.

Now, you can argue the sources and the meaning, and that's the problem with statistics. But there is a difference between the two, more than just 8 percentage points.

Traditional media has had a long tradition of informal, ethical standards that goes into "reporting"[1]. It wasn't perfect, but it was something. Now that we've flipped that apple cart with technology it's not easy to replace centuries of culture with an algorithm. What I fear is that the problem is fractal. Solving the easy parts of the problem scales well and is thus economical. The hairy parts of the problem don't scale and are left by the wayside. For an example of this look at YouTube content ID. It works well enough to placate content owners. Making it any better (e.g. fair use) requires human intervention so it's not gonna happen.

[1] A parallel in the political sphere is Trump himself. He's violated a bunch of informal understandings about what a president is/does. Now there's no going back.

just get a bunch of scientists who have no connection to any of the relevant fields to redirect their attentions to finding out if a thing is fact or fiction and i assure you that you will have an answer that is more honest and more accurate than otherwise

PS they'll need funding, and more time than you have to give

This is an old problem. What is new is how fast things spread.

Delaying when the upvotes/retweet/view counts become visible, will reduce the speed with which info spreads, and there will be less chaos.

Speed is highly overrated and it's value falsely tied to "engagement". Easy to run this experiment on HN - only show users their karma a week after their post gets published post quality will increase and engagement will drop. Which is an ideal outcome in a world where fiction trumps fact by a huge and exponentially increasing margin.

I like your style. I am really surprised no one has thought of this. Nominating you for Minister of the Internet Discussion Boards, next time there's an open election :D

>Then, Finally, I got to Clinn Watts' testimony, and this human is a saint: I propose "nutrition labels" for information outlets, a rating icon for news producing outlets displayed next to their news links in social media feeds and search engines. The icon would provide users an assessment of the news outlet's ratio of fact versus fiction and reporting versus opinion. The opt-out rating system would not infringe freedom of speech or freedom of press, but would inform users as to the veracity of content and its disposition.

Then the only challenge is to invent some cross between the biohazard and radiation hazard symbols, and just plaster it on 99% of the net.

Ideally the icon shows a realtime value or reasonably up-to-date value of Ratio of Fact:Fiction, Ratio of Reporting:Opinion. If you saw a piece of news from YCorp and their Fact-to-Fiction ratio was 9:1 you may say, "okay, that sounds likely." If their Fact-to-fiction is 1:18 However, you might just remember to laugh instead of resharing potentially hazardous messages. To me "icon" can also mean values/numbers/measure and not just "this is healthy and green, this is toxic and glowing"

> potentially hazardous messages

THOUGHTCRIME ALERT

Must protect fragile minds from dangerous messages...

This place is not a place of honor.

No highly esteemed deed is commemorated here.

Nothing valued is here.

This place is a message and part of a system of messages.

Pay attention to it!

Sending this message was important to us.

We considered ourselves to be a powerful culture.

The WIPP?! I love it... a desolate plain of windblown sand, gathering around the base of massive, repulsive obsidian and concrete pillars.

Then again, maybe we just go with, "There be dragons."

> The icon would provide users an assessment of the news outlet's ratio of fact versus fiction and reporting versus opinion.

Yes, we should always trust powerful institutions to tell us what the truth is...

I agree, sans-sarcasm. It's better than the alternative, which is allowing Russian disinformation to determine facts.

Absolutely. Such institutions have never abused such power in the past and surely never will.

They would never use fear to manipulate public discourse for example.

Absolutely right!

We should definitely shut down all Department of Education educational minimums as well! Those are just used to control the populace anyway.

And the FDA? That's just a way to control what we put in our bodies!

The SEC and financial regulations? When's the last time they helped anyone?

...

That was sarcasm, in case someone is experiencing a Poe moment.

In all these cases, the rules and regulations help some more than others, and may even be abused to help those with connections. That doesn't mean the institutions do not impart a net benefit to the country, and in many cases many orders of magnitude more than the problems they cause (in my opinion, at least).

A n agency with clear and transparent rules as to how to score institutions and what is considered true and what is considered false could do some real good. It could also be abused, but I don't think the idea should be immediately discounted before we explore how it might work a bit more in depth.

> It's better than the alternative, which is allowing Russian disinformation to determine facts.

No it's not better. "Russian disinformation" is a bullshit term to describe overblown and mostly ineffective tactics used by just about all governments.

Disinformation clogs the web. From WaPo to NYT to Salon to [insert media outlet here]. The very same bozos who sold you the Iraq war.

The dumbness of American discourse is almost too much to bear. Your shining city upon a hill apparently can't handle $100,000 worth of targeted ads. It's embarrassing.

That's the line Trump loves to get people to believe, that random fake blogs are equal to newspapers in their disinformation.

And why investigate it? Everyone is just as bad /s

> And why investigate it? Everyone is just as bad /s

I find the same is true of discussions about tech companies and privacy: "Oh they're all stealing all your data anyway, so why bother choosing on that basis." The irony, of course, is that this subtly undermines the premise of capitalism and especially Libertarianism (I get this from libertarian friends a lot). If you claim that my choice of search provider is meaningless, how can the market self-correct if people start to value privacy?

> "Russian disinformation" is a bullshit term to describe overblown and mostly ineffective tactics used by just about all governments.

No, the particular aggressive global propaganda (approach and scale) by the Putin regime had become a wide concern even before any discussion of it being directly targeted on the US election. See, e.g., https://www.rand.org/pubs/perspectives/PE198.html

"an assessment of the news outlet's ratio of fact versus fiction and reporting versus opinion"

Assessed by who? the Ministry of Information? No thanks.

War is peace. Freedom is slavery.

> Also count "impressions" as the number of times something appeared on screen.

I was under the impression that this was the standard, widely used definition of an impression which you are correct differs from the normal definition of impression.