Mark Zuckerberg is doing an interesting dance these days. He is facing criticism that Facebook’s promotion of “fake news” may have influenced an election.
Zuckerberg thinks this is all hogwash. He posits that Facebook users cannot be influenced by “fake news” on Facebook in their election decisions and further contends that fake news is not very prevalent on Facebook, despite evidence that seems to say otherwise. David Crotty did a nice overview of this earlier in the week.
As someone who uses Facebook to access news by following many different mainstream news outlets, I can say that almost any click on a Washington Post or New York Times article will result in the delivery of “recommended for you” articles from fake news sites and conspiracy theory pages.
Facebook’s fake news problems started before election day when it replaced human curation of their “trending” news list with an algorithm. People had complained that the trending story selections were biased. Personally, I find that most of the time the trending stories are about what happened on Keeping Up with the Kardashians or include “news” about celebrities I have never heard of. So Facebook did the easy thing and removed humans from the process. Enter fake news.
The first logical question for Zuckerberg would seem to be how he can claim that fake news is not “trending” when fake news was doing exactly that based on their algorithm for usage. Fake news became such an internet issue that Google added a fact-checker function in the Google News results. It doesn’t always work. As of 11 a.m. on Monday, November 14, 2016, Googling “Final Election Count” gets you this:
The top result is complete fake. The third result is about the first being completely fake.
I’ve written before about the blurring of the lines on online platforms. When you hear someone make a dubious claim, about anything, and you ask where they heard that, they will often say “It was on Facebook” or “on Twitter”. The original source is lost. Facebook is not creating content, Facebook is distributing content. No one knows (or even cares) who created the content anymore.
Sue Halpern wrote about this over the weekend:
“While it is true that this can be confusing to some readers, who are led to believe that the sites they rely on for information are honest and objective when, instead, they are designed to throw poisonous content into the news cycle, the actual effect is even more insidious: it has created an equivalence between those ideological sites and traditional journalism. In the Internet world, there is no difference between The New York Times and Breitbart.”
Fake news is becoming (or even already is) indistinguishable from real news for many people. Real news is competing with fake news and it’s not going well for them. Trust in journalism has taken a major hit. Further, journalists being able to share minute facts or hunches on Twitter circumvents the role of the editor. It also turns “journalists” into opinion sharers, blurring the lines for readers again. It is difficult to distinguish between fact and opinion among the press corps today.
The dumping of unfiltered and likely altered information from Wikileaks has also corrupted the news, real and fake. News organizations of all shades trolled through emails hacked by Russia and delivered by Wikileaks to find wrong-doing on the part of Hillary Clinton. Much of the “news” that came out of this was completely out of context and old. Some of the information also seems to have been deliberately altered, either by the hackers or Wikileaks.
It seems pretty clear that anyone who thinks “fake” news is not harming “real” news is sadly misinformed. Part of the problem is that newspapers gave away content in the digital revolution. Gave it all away. Now, no one wants to pay for it and an entire generation of “digital borns” never had to. The result is the shutting of papers and dwindling news staff.
Indulge me while I start to weave in some thoughts on how this related to scholarly publishing.
Predatory journals are not exactly the same as fake news organizations but they certainly are publishing papers of dubious quality. Some real journals are being hijacked and turned into fake journals. Finding these papers is not difficult.
Much in the same way that fake news is easily discoverable on Twitter and Facebook, fake journals and their contents are delivered via Google Scholar search. The question that often comes up is, what harm are predatory or fake open access (OA) journals having on real OA journals?
In every discussion I have seen on predatory journals, someone comes forth and declares that these are not a huge problem because real researchers are not being duped and no one would bother to read or use these papers.
Let me explain why this is short-sighted. In a meeting last weekend with 29 engineer editors, three told me in separate conversations that they have had a paper they published stolen by others who stripped the original author names off the paper, put their own on, and paid to publish them in an OA journal. Anyone within earshot of these stories was either shocked, or knew someone who had suffered a similar theft.
Over the course of a year, I get the chance to meet with about 100 editors and associate editors across our journals. The complaints about OA journals is one constant from these meetings.
A familiar complaint is about mass emails sent to presenters from one of ASCE’s conferences seeking solicitations for a Journal of Hydraulic Engineering, which is not the ASCE journal of the same title. Each year I get confused recipients contacting me. Some are concerned with the overlap, others are angry that we would ask proceedings authors to pay to publish the paper in our journal, which we, of course, are not doing.
The editors’ always complain about the massive amount of emails they get from OA journals. Some have been added to editorial boards of these journals without their permission and they can’t find anyone to take them off.
These experiences shape the discussions around openness and access. The editors don’t trust it. They feel a pay-to-publish model is dangerous.
I tell them that not all OA journals are like this. I tell them that some of the things they find in OA journals could happen at non-OA journals; but what they believe is what they see in their inbox or their Google Scholar results.
So is all hope lost when it comes to the reputation of journals in general? Are we destined to suffer some of the same issues facing news organizations today? Yes and no.
For one, scholarly publishers did not choose to give digital content away when journals started going online. It was heavily discounted from print, but there was a cost. Even OA journals, many of which charge an article processing charge (APC), are functioning on an income model. Payment for content supports the idea that “that which has value, costs money” and offers an alternative to the “you get what you pay for” attitude that comes with free content.
Another advantage for scholarly publishing is that the core users of content (researchers) do make value judgments based on the source. The first hallmark of quality will be the author names and if unfamiliar to a reader, the second will be the journal title. The third would be the research institutions affiliated with the authors. If all of those are unknown, a reader is wise to be skeptical.
There are many in the sciences who believe that the great take-away from OA and federal public access programs is providing access to the masses. This is not an uncontroversial idea. I have seen some pretty nasty arguments back and forth about whether the “general public” has a need or use for scholarly papers. Regardless, they do have access to OA content and have access to federally funded works and preprints now and even more in the future.
As the general public is being introduced to more and more scholarship, it seems even more important to be concerned about the bad actors. We have already seen skepticism of the entire scholarly publishing process in mainstream media. How do we also expect the public to differentiate a “fake” or “predatory” journal from the ones that perform due diligence? If you are expecting the dad researching a child’s condition or a daughter looking for new treatment outcomes for her mother’s illness to sort through the DOAJ list to see if what they found is from a reputable source, then your expectations are unrealistic.
The general public is also learning about scientific discovery through the mainstream media, which have been cutting their science and health reporting staff for years. Will the general assignments reporter know the difference between a reputable journal and one that offers no validation of any kind?
The cable news networks and national news organizations are doing some major soul searching this week. They are in the position of having to convince users that they should pay for content. This is a tough sell when the person who won the election campaigned partly on the deficiencies of the media.
Further, Facebook is having to defend itself again on whether it’s a social network or media platform. Scholarly publishing and the networks that feed into it and off of it would be wise to take a few steps back and think about any lessons learned that may apply to us. Martin Baron, Executive Editor of the Washington Post, had this thought on the topic:
“People will ultimately gravitate toward sources of information that are truly reliable, and have an allegiance to telling the truth. People will pay for that because they’ll realize they’ll need to have that in our society.”
I guess time will tell if he is correct about that.