Facebook suppressed right wing stories
finance.ninemsn.com.au
finance.ninemsn.com.au
Twitter definitely: they removed Milo Yiannopolis' identity verification tick without citing a specific reason. You might not like what he has to say (I frequently don't) but there have been much worse things Twitter doesn't seem to care about at all (someone threatening to murder Donald Trump). I don't want Twitter or any other media to become an echo chamber.
These sites are also started by young programmers who tend to have a libertarian, naive tec-utopian bend. At some point they notice that the wisdom of crowds is usually anything but, or their girlfriend/daughter/advertiser explains to them how the 600th image with a racist, misogynistic attack on your CEO isn't just a shining beacon of free speech.
Or it doesn't actually happen, and at some point you just get a few disgruntled ex-emps running to Gizmodo because they can only think of CENSORSHIP!!1! when you act on the assumption that the NYT is higher quality than Drudge.
They will happily tell you they can scale their system up to a billion people, but what happens when a billion people start interacting with each other they have no idea.
In other words read Jurassic Park.
If I'm an exec that leans libertarian (or left, or right, or whatever) and those are the articles that I see, I assume it's working. Moreso, if I see what's getting posted elsewehre, I just assume the rest of the world is smart like me.
You have to really dig in to see that something is going wrong. This involves setting up unbiased metrics that can't be gamed (hard) and listening to data that disagrees with your preconceptions (even harder).
Facebook (if/when it wants to) could wield tremendous power in swaying politics throughout the world by adjusting which stories/posts people see and which they tend not to. How are people to know that they're not being intentionally manipulated?
(and if that seems like Conspiracy Theory talk, perhaps first consider that Old Media -- TV, newspapers, radio, etc -- throughout the world very often do exactly this.)
When someone at Facebook asked Zuckerberg recently, if they should be doing something about Trump, it shows they are just waking up to what their platform is already being used for.
They don't seem to have any awareness, that there are numerous other ideologues, megalomaniacs, demagogues, they unintentionally support in countries across the world. That's when they run and hide behind free speech, cause facing the truth about what they have created is too complicated.
While I have no problem with 'hit pieces', I do have a problem when candidates are not treated or investigated with the same vigor. The 'MSM' has failed all of us, and the online companies are going down the same 'memory hole', trying to force people to one side or the other.
Sadly, it looks like we will continue to move away from the open Internet.
"In the near future - _corporate_ networks reach out to the stars." (Ghost in the Shell)
I noticed that an increasing amount of people apparently automatically "like" left- and rightwing Facebook pages, depending on other pages they follow and news they like.
This could possibly have legal consequences for the users and I'm looking for a way to legally pursue Facebook's auto-liking bullshit. They knowingly misrepresent people to increase their profits.
I think it would be equally interesting to discuss the biases of algorithmic solutions. It's a bit naive to think people selecting news is subjective while a computer selecting news (based on an algorithm written by people that relies on data created by measuring people) is somehow objective.
https://www.theguardian.com/technology/2016/may/10/facebook-...
Although I don't see any signal in this, it seems like the obvious course of action both if they are and if they aren't supressing conservative articles.
There is no transparency at all. The only way that post was restored was because someone in the Australian government, at avert high level, got in contact with someone else in Facebook who had the decision reversed and apologised to John Dickson.
They get people to control multiple personas on the same computer and post the same story from the point of view of those personas.
If a person encounters the same story a couple of times, they tend to believe it.
http://gizmodo.com/former-facebook-workers-we-routinely-supp...
But not just one person - Facebook had to apologise to John Dickson for deleting his post. So I ask you - what were they doing fooling around with his posts?
"This individual says that workers prevented stories about the right-wing CPAC gathering, Mitt Romney, Rand Paul..."
You're confusing the multiple people who claimed they were asked to inject stories that weren't trending.
In fact, you have multiple sources who contradict the opinion of your source: "Other former curators interviewed by Gizmodo denied consciously suppressing conservative news, and we were unable to determine if left-wing news topics or sources were similarly suppressed."
Finally, underlying contention, that Facebook was at fault, is even denied by that same individual.
"there is no evidence that Facebook management mandated or was even aware of any political bias at work."
And finally, this is Gizmodo, so unless you have something from a trustworthy source, I would treat them as suspect at best.
And then there's the issue that's these kind of algorithms need tuning - they're not intrinsically neutral. It's unlikely that conservative and liberal social networks/stories have the same kind of structures, and details of how the network is connected can influence what's trending. This makes the complaint a little absurd.
Not only are algorithmic results also biased, these manual corrections might conceivably be useful to correct that bias.
There's this naive assumption that there exists the One True Fair and Neutral ranking algorithm that's unbiased. But the whole notion is ridiculous: we want that bias, because we want to discriminate in favor of "trending" content - and "trending" isn't even rigorously defined.
And even if we had a rigorous definition, that doesn't mean it'd be the most useful definition. Elections have strictly defined protocols, yet that doesn't mean an elected US president has the majority of votes. And lest you believe majority voting is the answer, a majority elected president might nevertheless be inferior to a competitor by majority opinion (i.e. it's not stable)... and of course regardless of election result there's no guarantee that current opinions are a good predictor of future opinions let alone achieved results.
In short: not only is this just one person's opinion, the standard they hold Facebook to (no bias) is impossibly high.
Perhaps there's some amount of unwanted (by whom) bias, but dealing with that isn't as black and white as the complainer suggests.