Hacker News Reader

Comment by entee | Hacker News Reader

I don't think that's true. There seems to be a view that nobody can try to be objective and the objective doesn't exist. I'm a scientist by training and trade, and in my experience the real world is pretty damn objective. Something happened, or it didn't. Something is true or it's not. I used to say as a grad student that nature didn't care whether I graduated.

Granted it's not always easy to get good data, and it's not always obvious what's clearly true. But where possible, we should embrace it, and where it's hard, then we should have a clear discussion about what we can and cannot believe.

Fact checkers in my experience do a good job of that, they present a clear picture of what is known and what isn't, and they take that responsibility very seriously. If someone tends to make more false statements than someone else, that doesn't mean pointing that out is biased, it's just reporting the facts. Sometimes reality is biased.

View on HN

After looking at many fact checkers and the job they do, I could not disagree more.

People do not speak in simple statements that can be either scientifically "verified" or "not verified". When talking, people often use hyperbole. They speak in approximate numbers (e.g. 20% instead of 17%, simply because it sounds better). Not to mention people often talk about socially complex subjects that cannot be, in ANY way, verified as simply "true" or "false" without interjecting the fact checkers own personal opinion into the mix.

Fact checkers have tried to solve these issues by having "truthiness" ratings. For example, Politifact has the ratings "True, Mostly True, Mostly False, False, Pants on Fire". However, this is obviously a scale that has no objective meaning. As a result, these ratings are applied inconsistently and with a heavy bias towards the fact-checkers own agenda (which, for Politifact, is heavily left-leaning).

How do you objectively decide whether a hyperbole is true or not? How do you objectively decide that numbers are too far off the actual scientific numbers? What rounding is too much? How do you objectively decide the truth of complex social issues and statements, and what data do you trust for determining these "objectively true answers"?

The answer is you can't, and you don't.

It might be hard to determine a false statement from a pants on fire statement, but just because it's not purely objective doesn't mean its not useful.

What would are some Politifacts that you think were wildly off base? Like statements that you believe to be true that are labeled "Pants on Fire".

That isn't what he was saying.

He is saying that the 'fact checking sites' are biased.

This is true. It's the old problem of "who guards the guards".

Snopes.com is supposed to be a fact checking site.

Snopes.com recently rated as 'mostly false' a notorious video of a Trump supporter being beaten up and suggested it was primarily about a traffic incident.

The guy who got beaten up actually offered his own testimony of what happened on camera later on. This is not mentioned in the 'fact check'. There is also the fact that his attackers are screaming at him that he is a Trump supporter and shouldn't be. They repeatedly shout "Don't vote Trump".

Then, just today, I see they have updated their rating to 'mixture' and added more text suggesting that the spread of the video is about racism. It's a pity I didn't take a screenshot from earlier.

This is the url: http://www.snopes.com/black-mob-beats-white-man-for-voting-t...

So no. I don't take the 'fact checking' websites seriously.

Imagine for instance a white nationalist walked into a store with a Mexican shopkeeper, got into an altercation because the Mexican "didn't give the right change" when purchasing an item there, and then followed that up by bringing his friends to trash the store.

On the store's CCTV there is audio recording showing racial slurs are used. Furthermore the shopkeeper says to reporter they attacked his store because he was Mexican and denies that he gave back the incorrect amount of change. He claims they were looking for a fight.

Then along comes snopes.com and they make an article claiming this was really about a shoplifting incident. Police sources say the white gang stole some items when they left.

That is what this is like. The fact checker is obviously biased.

That Snopes entry is a perfect example of promoting a bias without specifically lying:

"The clip has since been shared by white supremacist publication The Daily Stormer..."

Is that relevant to the beating? No. Is that relevant to the truth or falsity of what preceded the beating? No. It is simply there to inflame one side using "guilt by association". A similar example was that an endorsement of Trump by some racist organization was widely reported, even though nothing else they had ever said or (even allegedly) done had been considered newsworthy to that point. Had Trump sought such an endorsement, that would be relevant.

Ultimately, selective reporting can be just as powerful as false reporting... and there is far, far more false reporting than most realize (especially misreporting on legislation). People need to remember that media have no legal obligation to report truth (let alone the whole truth) and, unless they libel or slander a specific entity, suffer no consequences for stating flat lies. For-profit media's primary motivation is maximizing profit, and accuracy often does not maximize return or profit, especially in the short term.

Reply on news.ycombinator.com