Discussion

• Newsmax Niall Carson on x

  Facebook: 6 Percent of Content Users See Is Political

• @kevinroose Kevin Roose on x

  Facebook is absolutely teeming with right-wing misinformation right now. These are all among the 10 most-engaged URLs on the platform over the last 24 hours (per @NewsWhip data) https://twitter.com/...

• @kevinroose Kevin Roose on x

  Facebook says that the pages with the most reach are more mainstream and less hyperpartisan than the rankings on @FacebooksTop10 would suggest. Which might be true! But we can't fact-check or report on it unless they make that data public.

• @grex Greg Hochmuth on x

  @kevinroose notice the use of “weekly active people” in their reach rankings. fb hasn't cared about weekly active numbers in a decade (daily is all that matters internally; monthly is reported for wall st). it makes me think that a daily ranking looks “different”, ie less flatter…

• @kevinroose Kevin Roose on x

  In all seriousness this post is helpful in explaining the various types of data Facebook has on what users engage with (which journalists and researchers can track) vs. what they actually see (which we can't). https://about.fb.com/...

• @kevinroose Kevin Roose on x

  I confess I did not have “get personally subtweeted by the 6th largest company in the world” on my bingo card today but America is a land of possibilities https://twitter.com/...

• @sarafischer Sara Fischer on x

  Got a few questions on this so following up. Facebook isn't saying yet how it defines political content but says it uses its own categorization method to decide. https://twitter.com/...

• @karaswisher Kara Swisher on x

  This continued back and forth of FB and @kevinroose over what is popular on their platforms is my fave thing ever, except that the social media network always manages to show that it's still real bad even if it is not AS bad. https://twitter.com/...

• @rmac18 @rmac18 on x

  “The pages with the most reach aren't hyperpartisan rightwingers, but rather magicians” is an objectively funny argument https://www.axios.com/... https://twitter.com/...

• @ruskin147 Rory Cellan-Jones on x

  Facebook tries to reassure critics it's not what one called an “algorithm-driven cesspool of disinformation and hate speech” https://twitter.com/...

• @sivavaid Siva Vaidhyanathan on x

  Because if we have learned one thing over the past 12 years it's that we can TOTALLY trust whatever Facebook says about itself. https://twitter.com/...

• @jesselehrich Jesse Lehrich on x

  a company whose entire platform is built to maximize engagement now insists engagement is a silly metric — we use reach instead, via data FB won't allow access to. https://about.fb.com/...

• @oliverdarcy Oliver Darcy on x

  Facebook posts the top 10 news publishers on the platform from October 23 to October 29 by reach of their links, which is different than the data @crowdtangle provides which ranks things by interactions. https://about.fb.com/... https://twitter.com/...

• @dlberes Damon Beres on x

  Countdown to Facebook closing public access to CrowdTangle in 5, 4, 3... https://twitter.com/...

• @liz_shepherd Liz Bourgeois on x

  Sharing some more detail on what people saw on Facebook leading up to the US election as there have been questions out there about this. Based on our analysis, political content makes up about 6% of what people see. https://twitter.com/...

• @karaswisher Kara Swisher on x

  In general, I am for this @kevinroose ground war so that it eventually gets FB to release more and more data and be more transparent so we can judge for ourselves things like how big the baseline of hate is.

• @earleyedition Dave Earley on x

  The domination of rightwing content was normal. Everyone noticed when the @FacebooksTop10 list suddenly swung away from conspiracy theory peddlers for a day or two https://twitter.com/...

• @emilybell Emily Bell on x

  As a publisher (or platform) it's not a good sign that you are having to explain your attention data with caveats and extra charts - and tell people effectively ‘you are not looking at what you think you are looking at’ https://about.fb.com/...

• @sarafischer Sara Fischer on x

  Facebook's new CMO Alex Schultz says Halloween saw twice the increase in posting on Facebook that Election Day did. https://www.axios.com/...

• @jason_kint Jason Kint on x

  Hey all, I'm repeating once again. Please do not share links to Facebook's propaganda from comms. You wouldn't do that for any other untrusted source. Link to a report on it with proper context. Link to your competition. Just don't amplify their spin and do their jobs. https://tw…

• @karissabe Karissa Bell on x

  feel like I've heard FB make a version of this argument before and it's .. bizarre? If engagement isn't a good measure of reach what's the point of CrowdTangle? Why not improve it? https://twitter.com/...

• @mattnavarra Matt Navarra on x

  Facebook says its analytics tool Crowdtangle, built to show what's popular on Facebook, is not so great at showing what's popular on Facebook https://about.fb.com/... https://twitter.com/...

• @jason_kint Jason Kint on x

  @emilybell Their definition of reach is no better than the data they provide and @kevinroose had to use. Reach across an aggregation of pages or a domain simply treats every eyeball equally regardless of frequency and duration. https://twitter.com/...

• @craigsilverman Craig Silverman on x

  This post from Facebook aims to offer perspective on page reach vs. engagement. It says “engagement does not predict reach.” https://about.fb.com/... But when we spoke to big liberal page owners who had their reach suppressed by FB, they discussed how connected these metrics are.…

• @deaneckles Dean Eckles on x

  On the disconnect between the “top link page posts” on Facebook and what links most people are seeing in their feed — from @alexschultz & team https://about.fb.com/... https://twitter.com/...

• @rmac18 @rmac18 on x

  I guess it's good Facebook is offering more insight into reach, but that they're doing it now — after getting hammered by people quoting @FacebooksTop10 — only reinforces the idea that the one thing the company consistently responds to is negative PR. https://about.fb.com/...

• @earleyedition Dave Earley on x

  Kevin sees you 👀 Also, 'what you should have taken into account is this data which we don't share with anyone' isn't the best argument https://twitter.com/...

• @oneunderscore__ Ben Collins on x

  This company is so angry about what's popular on their website. The reality is, Facebook ignored the problem until it got completely out of control, and is now indignant as it splashes a squirt gun on a 5-alarm fire it allowed and can't control. https://about.fb.com/...

• @cameronwilson Postmaster-General Cameron Wilson on x

  TLDR: We don't think CrowdTangle engagement data shows the full story. Here's a tiny bit of data showing they're not the same in this specific example. But we're not letting you access the data (unless you're a specific researcher from specific unis who we approve) https://twitte…

• @karissabe Karissa Bell on x

  Looks like Facebook is (once again) pushing back on all those CrowdTangle lists of the top posts on Facebook, says most people don't actually see that many political posts on Facebook https://about.fb.com/...

• @earleyedition Dave Earley on x

  Facebook have been so rattled by the response to @kevinroose's daily @FacebooksTop10 lists showing the huge popularity of far-right conspiracy content that they've published this long explanation, including Reach, which is not publicly available data https://about.fb.com/...

• @natijomartinez Natalie Martinez on x

  I like how Facebook's whole argument is that reach data is more important in determining impact than interaction data and the conclusion of that data is that data is still that Trump and Fox News are the most impactful political pages: https://about.fb.com/... https://twitter.com…