Slashdot Mirror
Facebook's Fight Against Fake News Was Undercut by Fear of Conservative Backlash (gizmodo.com)
Facebook has been concerned about fake news stories that circulate on its social platform and how often such incidents occur. The company has had high-level internal debates over the matter since May, discussing different options to curb movements of hoax and false stories. Gizmodo reports Monday that Facebook executives conducted a wide-ranging review of products and policies earlier this year with "the goal of eliminating any appearance of political bias." The company even had a major update for the News Feed planned which could have supposedly filtered fake stories, but the update never saw the light of the day because it was afraid to use it. From the report:One source said high-ranking officials were briefed on a planned News Feed update that would have identified fake or hoax news stories, but disproportionately impacted right-wing news sites by downgrading or removing that content from people's feeds. According to the source, the update was shelved and never released to the public. It's unclear if the update had other deficiencies that caused it to be scrubbed. "They absolutely have the tools to shut down fake news," said the source, who asked to remain anonymous citing fear of retribution from the company. The source added, "there was a lot of fear about upsetting conservatives after Trending Topics," and that "a lot of product decisions got caught up in that." In an emailed statement, Facebook did not answer Gizmodo's direct questions about whether the company built a News Feed update that was capable of identifying fake or hoax news stories, nor whether such an update would disproportionately impact right-wing or conservative-leaning sites. Instead, Facebook said it "did not build and withhold any News Feed changes based on their potential impact on any one political party."
If they hadn't been rigging the news feeds and injecting their own bias, they wouldn't have gotten into this mess.
The most recent fake one that I've seen, with its supports absolutely adamant that it's real, is the "Clinton didn't really win the popular vote, Trump did!" thing. They defend it to the day they die, despite the fact that it's flatly contradicted by all official sources, can be traced back to the guy who made it up, and is based around factually incorrect statements about how votes are tabulated.
Not that the left is innocent in all of this. I still keep seeing that fake quote about Trump saying that Republican voters are idiots who will believe anything. How many times do you have to point out that it's fake for people to stop circulating it?
We need more fact checks, period. It bugs me to no end that news stations just broadcast politicians giving speeches and pundits making claims, wherein they may reiterate a dozen different things that have literally zero basis in reality... and just let it go uncorrected. That's journalistic malpractice, plain and simple. I know they want to jeep the pace of coverage up, but they're willfully letting their viewers get misinformed in order to do so.
It's times like this I wish I had a friend named 'The Professor'.
The auto-filtering criteria apparently caused far more right-leaning stories to be filtered than left-leaning stories... So basically, it was bad code...
Was the decision made purely based on the fact that it was filtering more right-leaning stories? Did someone evaluate whether it was because there were more false right-leaning stories being posted?
The pollsters used the same techniques they did as before with reasonable success. McCain and Mitt's results pretty much matched them. The problem is that Trump is not a normal candidate and that surveyee's didn't react to him like they did a normal candidate. He's thrown monkey wrenches into a lot things (for good or bad).
There was no reason for DNC to manipulate the polls. A close election produces more turn-out, which is what they wanted. If anything, the bad polls hurt Hillary rather than Trump.
Table-ized A.I.
Plus the fact that Hillary won the popular vote (by +650K votes and still counting)
I really wish people would stop talking about this. I'm NOT a Trump supporter, but talking about the popular vote is emphasizing an irrelevant aspect of the data given how our system is set up.
Trump and Clinton did NOT campaign to win the popular vote. If they were doing so, they likely would have skipped rallies in many "swing states" and instead held them in places more likely to get out the maximum votes for their side. That could have led to a very different popular vote split.
It's kinda like playing a game of Monopoly and losing but saying, "But, but I had more properties! I should win! I had more properties!" Except Monopoly isn't about accumulating the most property, it's about accumulating more money and bankrupting the opponent. Those are the rules of the game. If you want to play by different rules, fine... but that's a different game. The US election is set up one way, and the candidates "played" to win by those rules (i.e., Electoral College).
By the way, I'm not defending the Electoral College either, and there are legitimate reasons to get rid of it. But the mismatch here isn't really a strong argument -- if you believe that campaigns and rallies and advertisements have ANY effect on voter turnout, then there's absolutely no guarantee that the numbers would have been the same if the candidates were trying to win the popular vote and made campaign choices based on that.
It has been official republican strategy to block his nomination until the election so that there would be a chance that the next president might be a Republican and they could get a more conservative court instead.
That the Republicans then talked of further delaying for the next 4 years should Clinton win sends an even more ominous message that they care more about politics than the Constitution, the Country and *all* of its people.
It must have been something you assimilated. . . .
The alleged email quote: "We are going to try to do an oversample of seniors on the poll. Sample too small otherwise..."
That can be interpreted at least in two ways. It may mean they simply don't have enough data for a given factor and so use a subset with more factors to extrapolate that factor to a general set. It's a statistical "trick" to tease more info out of a limited data set.
Perhaps one can argue that they are "over-guessing" which makes their poll bad, but that's not the same as introducing intentional bias. It could be being a cheap-skate rather than propagandist. I don't know enough about their data to say for sure.
Further, I cannot tell from that alone that they are talking about an internal poll or a public poll. If it's an internal poll for internal usage, then it's not "public manipulation". It's then for internal reports.
Without more evidence about the context, I see no reason to make a default assumption of malice. Context matters. Don't jump to conclusions.
Table-ized A.I.
That's one of the things that blows my mind. There are a lot of people in the U.S. who basically said, "I'm sick of all you billionaire New York atheists controlling my life! You're out of touch with the working man, and don't care about me!" and then proceeded to elect Trump. It's like an Onion article.