Michael Shermer: Don’t denigrate open source academic journalism.
Rebecca Watson (who is indisputably awesome) beat me to it. Sort of. In a video she posted a few hours ago (although probably a few days ago by the time I publish this) she explains how Michael Shermer made a handful of logical errors when he wrote this in Scientific American suggesting that the new hominid species recently discovered, Homo naledi, probably were all murdered.
So I’m not going to write about that. A much better explanation is also written by John Hawks, where he goes into detail about how Michael Shermer probably didn’t even read the original research he criticises.
What I’m going to write about instead is based off of one quote from this article (emphasis mine):
Instead of publishing in Science or Nature, the prestigious journals in which major new fossil human finds are typically announced, the authors unveiled their discovery in eLIFE (elifesciences.org/content/4/e09561), an open-access online journal that fast-tracks the peer-review process. And instead of meticulously sorting through the 1,550 fossils (belonging to at least 15 individuals) for many years, as is common in paleoanthropology, the analysis was published a mere year and a half after their discovery in November 2013 and March 2014.
There’s a growing, let’s call it stigma, in academic circles regarding open source journals. In order to explain this I need to explain how Academic Journalism works.
Let’s say you’re a researcher. You’ve run your experiment, or discovered something, and have done your data analysis. You write up your results in a paper (which itself will take about a year) and you want to share your findings with the world. You have a few options available.
The standard route will involve you submitting your paper to the most prestigious journal in your field, something with a high impact factor. The editor will read your paper and determine if s/he wants to spend the time putting it through peer review. Assuming s/he does, your editor will send your paper to three (or more) reviewers who are experts in the subject of your paper. They will read it and make recommendations. Your editor will forward this to you and it is up to you to either make the suggested changes or take your paper elsewhere. This is even if the reviewers thought your paper was publishable. This process will be repeated at least once more and possibly as often as 5 times before a final decision is made. This entire process takes at a minimum one year and more likely three or four.
Of course you could just publish all of your data on your website, either raw or written up in your paper. This is generally frowned upon for some pretty obvious reasons.
Option one is how the mainstream scholastic journals operate. These run on a subscription service: professionals subscribe to the journal or pay for each article they wish to read piecemeal usually at about $30 a piece. Most institutions and universities subscribe to these journals so the students and staff get access for free. This is how the journal pays for itself.
Your other option is to publish in an “open access” journal. This is where things get tricky.
The more prestigious open access journals do real peer review, just like the standard journals. Often this occurs more quickly than a standard journal, but not always. Open access journals make their articles available freely online to anyone and rarely print copies on paper. Since they don’t charge for people to read the journal they make their money by charging the author.
That’s right, most open access journals are similar to a vanity press.
Some of these open access/pay to publish (rather than pay to read) journals are sub par or simply dishonest, as shown in the infamous “Get me off your Fucking Mailing List‘ incident shows.
Yet not all are. Some are becoming very prestigious such as Scientific Research. And even standard model journal publishers will publish non-scientific nonsense. This issue isn’t with the process (open source vs standard model) its with the individual journal. Certain journals do peer review better than others.
The fact that a paper is published in an open source Journal does not necessarily mean that it is sub-par or has been “fast tracked.”
In fact, Hawks responds to Shermer’s claim in his article:
He expresses disdain about our team’s decision to publish in open access journal instead of “prestigious journals” like Nature or Science. He assumes that our team has not “meticulously sorted through the 1550 fossils” because we haven’t spent “many years” describing them.
That last one especially irks me, since “meticulously sorting through the 1550 fossils” has basically been my job description for the past two years.
I fact, this article made so much press that had the peer review process been “rushed” as Shermer states others would have commented upon it. Shermer himself would have said so in his article (although John Hawks posits that Shermer didn’t even read the original research).
The problem isn’t with open source. In fact open source academic publishing is an answer to a big problem with academic publishing: the paywall. Much of the scientific literature is hidden behind a paywall that, unless you’re a university staff or student, means shelling out about $30 per journal article you want to read. The issue is with the peer review process. Some journals do it well, others (as I noted above) fail at it.
So when evaluating a paper, especially one that has made such headlines in the past months, you shouldn’t dismiss the findings because the publishing method is seen as less prestigious. Instead, Shermer should have read the actual article and judged the findings on their merits.
The point of the long ramble is this: don’t denigrate open source publishing. Open source seeks to make science more publicly accessible to readers who are interested in science but don’t happen to have access to a journal subscription. Instead, judge the journal article on its merits rather than its source.