Tuesday, September 04, 2018

WikiTribune is hiring fact-checkers


Kavanaugh’s Kozinski Test: The White House has painted Judge Kavanaugh into a very tight corner when it comes to his former mentor; But we deserve to hear how the SCOTUS nominee thinks about issues of harassment and the workplace.”Dahlia Lithwick has this jurisprudence essay online at Slate.



Almanac: Evelyn Waugh on writing
“He delighted in writing, in the joinery and embellishment of his sentences, in the consciousness of high rare virtue when every word had been used in its purest and most precise sense, in the kitten ... read more


Trump tweets a fake Google video

U.S. President Donald Trump directed his tweeting ire to Google this week.
On Tuesday, he accused the search engine of hiding good news and overwhelmingly displaying “National Left-Wing Media.” His accusation relied on a highly questionable graphic, but plays on genuine concern over the transparency of the platforms’ algorithms. (Disclosure: Google funds some IFCN training, see our transparency statement.)
“There’s a reason that misleading claims of bias in search and social media enjoy such traction,” argues Tarleton Gillespie, who’s written a smart new book on online moderation.
The attack, which (Poynter-owned) PolitiFact rated false, also came as a few dozen Facebook employees organized an internal group to complain about how the company is allegedly stifling political diversity. What they object to, precisely, is not clear.
As the always excellent Kara Swisher writes: “most tech leaders are more often lightweight versions of libertarians and largely apolitical except for backing gay and transgender rights and wanting to allow more qualified immigrants into the country to make more tech. Otherwise, in my reporting, I have found that they love those tax cuts and adore the repatriated income and can’t get enough of deregulation.”
Wednesday’s tweet was even more baffling. Accompanied only by the hashtag #StopTheBias, @realDonaldTrump shared a video composite showing Google’s homepage highlighting Barack Obama’s State of the Union over the years — and ignoring President Trump’s.
The company denied the accusation to BuzzFeed News. More convincing still, BuzzFeed found that both the Internet Archive and Trump-universe Reddit channels have the receipts: Trump’s 2018 SOTU was indeed highlighted under the search bar.
“If you're going to level big angry presidential complaints at Google...at least...don't flagrantly lie about something that's easily provable with archive.org and a screenshot!” BuzzFeed’s Charlie Warzel tweeted.
(AP Photo/Natacha Pisarenko)

This is how we do it

  • Here’s how Full Fact has helped fact-checkers live fact-check political events. The project also appeared on a BBC radio show to talk more about its automated fact-checking technology.
  • PolitiFact’s Katie Sanders has been promoted to managing editor from deputy editor.
  • BuzzFeed News’ Jane Lytvynenko got far-right commentator Mike Cernovich to admit that he should be publicly shamed.

This is bad

  • Move over recipes and wedding inspiration: Pinterest has become a hotbed of conspiracy theories.
  • There’s a new social media platform for American liberals — and it’s filled with false rumors about Donald Trump.
  • After a mass shooting in Jacksonville, Florida, several right-wing websites and commentators in the U.S. falsely identified the shooter as a 23-year-old Reddit user. Only a few have issued corrections.
(AP Photo/Nati Harnik, File)

This is fun

  • A fake Warren Buffett account that posted inspirational tweets racked up hundreds of thousands of followers before being taken down.
  • BuzzFeed News’ Charlie Warzel tricked his mom using a digital recreation of his voice powered by artificial intelligence.
  • No, this teenager did not hack a state election, ProPublica reported.

Coming up

  • Are you a reporter in Florida, Georgia or North Carolina wondering how to cover disinformation in odd corners of the internet? First Draft and the IFCN have just the right workshop for you.
  • WikiTribune is hiring fact-checkers located anywhere in the world.
  • MediaWise is assembling a team of part-time teenage fact-checkers to debunk misinformation on Instagram. Fill out this form to express interest.
(AP Photo/Don Ryan)

A closer look

  • A doughnut shop in Portland, Oregon, has recently had more phone calls from conspiracy theorists than customers. Here’s why.
  • NewsGuard has launched its first product: Browser extensions that show users a quality rating for different websites. Nieman Lab wrote about how that system works in practice.
  • Daniel found a network of at least 30 fake Instagram profiles that impersonated news organizations to promote a company that promises to boost people’s followers. Instagram removed 22 of them after he reached out.

If you read one more thing

Motherboard published an in-depth look at how Facebook’s content moderation strategy has developed over the past several years. It comes on the heels of a Radiolab episode about the same topic.

10 quick fact-checking links

  1. New research suggests that, when reporters include the term “fake news” in their headlines and tweets, people become more confused about what actually constitutes misinformation.
  2. In Latvia, police arrested a 21-year-old man for allegedly spreading fake news.
  3. http://go.pardot.com/e/273262/trump-misinformation-site-true/cl9n6/240623831
  4. The Memphis, Tennessee, police department used a fake Facebook account to monitor Black Lives Matter — and it may have been against the law.
  5. Fact-Nameh, one of the only Iranian fact-checking organizations, has been blocked within the country.
  6. The Washington Post Fact Checker used the word “lie” for the first time.
  7. Two researchers detailed in CJR a possible way that WhatsApp could regulate misinformation. But the company won’t be executing it anytime soon.
  8. Correctiv fact-checked that study about Facebook inciting violence against refugees in Germany.
  9. Political scientist Brendan Nyhan talks about the backfire effect of the backfire effect in a podcast.
  10. PicPedant, still doing his thing. Tommaso De Benedetti — also apparently still doing his thing
via Daniel and Alexios



Mantelero, Alessandro, AI and Big Data: A Blueprint for a Human Rights, Social and Ethical Impact Assessment (May 12, 2018). (2018) 34(4) Computer Law & Security Review 754-772 . Available at SSRN: https://ssrn.com/abstract=3225749