鶹ýӰ

The Biggest Scandal in Science

— Why should the public pay twice, even three times, to see the research it funded?

MedpageToday

Following is a transcript of this video; note that errors are possible:

Rohin Francis, MBBS: Welcome back to yet another night shift. It's Shift #4 and in a fit of narcissistic delirium I decided to search for one of my own publications, but I can't actually see it without paying. The per-article price is about standard, £30 or $40.

If you're writing a paper, you might reference 50 studies, which would be a huge expense. Luckily I get access through my university, who are in turn funded by me and other students through our fees and by the government. An establishment like UCL will pay something £10 to £12 million pounds a year for subscriptions to scientific journals, most of which belong to just five publishing companies.

OK, we're all used to paywalls these days.

The Times or The Economist pays their journalists either per article or with a salary. But scientific journals don't produce their own material, scientists do. They're not paid by the journal, they're normally paid by us, the taxpayer in most cases.

Then journals have people vetting submissions, the peer-review process. But peer reviewers aren't paid either. They are other scientists giving out their time up for free, any working scientist who has been involved in peer review on the expectation that others will do the same for you when you submit something.

Now, that in itself has problems. A peer reviewer could be a clueless YouTubing doctor who is half-arsing his review -- I'm joking, I do take it seriously when I do it -- or an experienced professor who spends a great deal of time analyzing the data and checking the statistics. To the author, we just go down as "Reviewer 1" and "Reviewer 2."

I have had this experience, one reviewer tore me to shreds, said the article was rubbish, asked if I was even a native English speaker and the other reviewer's entire comment was "seems fine." But the multiple problems with science publishing -- like peer review and bias towards novel and positive research -- is a tangent for another day.

Science journals don't pay their authors, they don't pay their reviewers, the public pays for the content -- the science itself, in the form of tax -- and the public then pays millions to access that science. You, as a taxpayer, are paying twice to benefit from any of this science.

Is it just me or is that the biggest scam since Charles Ponzi said "it's not a pyramid scheme, it's a trapezoid scheme"? Incredibly, the scientific publishing industry is valued at around $20 billion dollars, somewhere around the music and movie industries, despite being far smaller as it's so much more profitable.

The biggest publisher, Elsevier, had a 2018 profit margin of 38%, placing it way above all the big tech companies, and has been constant for years. Poor, impoverished companies like Apple, Amazon and Google can only dream of figures like that.

Let me ask you a question. Should knowledge be free? Should we have to pay to be educated? Now, don't think about an academic in an ivory tower. Think about a regular person, unsupported by a university, who wants to learn more about a certain field. Maybe their child was born with a rare disease. Maybe they want to really understand the science of climate change instead of reading the media. It's all well and good internet smart arses, like me, who say please do your own research. But if the average citizen can't afford to, they will instead find misinformation, which, as we all know, is freely available.

I want to highlight two people in this story, two people that have shaped this whole landscape more than anyone else. People outside Britain might not have heard of Robert Maxwell, but you've certainly heard of his daughter, the widow of convicted sex offender, and dubiously-suicided Jeffrey Epstein. Ghislaine Maxwell is the daughter of Britain's most notorious media tycoon, Robert Maxwell, fraudster, alleged spy, and one of the inspirations for Logan Roy's character in "Succession."

If you go back a few decades, the idea of making money out of scientific work was absurd. Of course, businessmen used scientific ideas throughout the Industrial Revolution and people could patent use of their ideas, but knowledge itself was shared freely particularly among scientists until Maxwell realized he could turn science into profit and created Pergamon Press. Stephen Buranyi colorfully illustrates Maxwell's rise to power and influence in his excellent article that I'm linking below.

But essentially Maxwell wowed scientists with flash hotels, glamorous parties, and cold hard cash, then signed them up to exclusive deals with his journals. We would get dinner and fine wine, and at the end, he would present us a check, a few thousand pounds for the society. It was more money than us poor scientists had ever seen.

I didn't realize until researching this video how enormous his influence on modern science has been. The whole system has been shaped by his model: the paid subscriptions, the journalistic way controversy and novelty are prioritized, the dominance of a handful of journals, and the way scientists dream of being published in those high-profile periodicals.

Long, meandering pursuits of the big questions in science or attempts to recreate the work of others, replication studies, have been replaced by quickly turned around experiments in whatever flavor is hot with journal editors at the time. This year, of course, if you're not publishing COVID research, who the hell even are you? Oh what's that? Your unblinded observational study of vitamin D and COVID is going to change the world? Oh, and you don't have a control group? Even better! Call the Nobel Prize committee. We got a gamechanger right here.

While I made a comparison to print journalism earlier on, there is a big difference. You don't need a subscription to The Atlantic or Time magazine to live your life, but scientists working in a field have got no choice. They can't decide they're not going to read research published in their area or they won't be able to work. This is a true monopoly.

Maxwell's eventful life prematurely came to an end some years later, but prior to that he sold Pergamon Press to Elsevier, making it the biggest scientific publisher by a long way. Soon, Elsevier started sharply hiking up the prices to their thousand titles. University libraries were, for the first time, forced to stop subscribing to the less popular titles. If you worked in a niche field, you might find yourself unable to access the very science you're pursuing.

By the mid-'90s, Forbes, the magazine, asked if the immensely profitable scientific publishing industry would be killed off by the new hotness in town, the Internet, which offered to spread knowledge around the world for free. But Elsevier, now in control of about a quarter of all scientific literature, told universities to pay a lump sum for their entire catalog, or get nothing.

That figure has been steadily climbing ever since. I don't want to single out Elsevier here -- there are a few other companies that collectively exert a stranglehold on scientific publishing -- but Elsevier do help us segue into the second person I want to highlight because they are being sued by Elsevier for $15 million dollars.

I have described Alexandra Elbakyan as having done more for science than any other living person and I stand by that. If you don't recognize the name, you might know her website, Sci-Hub, a server dedicated to offering almost all scientific articles for free, which is of course, illegal. But illegal isn't always wrong. It is now home to almost 80 million scientific papers.

I caught up with Alexandra via Zoom, all the way from her home in Kazakhstan.

Francis: Why did you create Sci-Hub? What were you trying to do?

Alexandra Elbakyan: I made this website so anybody could go there and have access to any research article for free. It took only 3 days to create and launch the very first working version of Sci-Hub. This was the very version.

Francis: Does it occupy a lot of your time now or is kind of not such a big commitment at the moment?

Elbakyan: Here, you must consider the fact that I do not only the technical part, but also the promotion. For example, even the interviews, it also takes a lot of time. So not only the technical part, programming, but also promotion. I don't keep track of time to be honest. I don't only do programming, but also kind of political promotion of this website, which also takes up a lot of time.

Francis: How is it funded?

Elbakyan: It is funded entirely by user's donations.

Francis: The science publishing companies are not very happy with it, particularly Elsevier has been very unhappy and attempted to sue you. Have you had to pay for lawyers and things? What's happening from that side of things?

Elbakyan: Nothing is happening there because I have been completely ignoring this issue for the time being. I know that I owe $15 million, but I haven't paid anything yet.

Francis: But if you were to talk to them, what would you say to these publishing companies? Not about the legal side, but what about what their view is that you should pay for this knowledge? What is your opinion on that? What would you say to them?

Elbakyan: I wouldn't say anything to them. I would say "Just give up."

Francis: There is an alternative approach to academic publishing where the authors will pay, so open-access. Sometimes I have been asked to pay for articles that I have submitted to journals. Do you think scientists should be paying to have their work published?

Elbakyan: In principle, I think no one should be paying, neither for publishing nor reading research articles. If we consider this model where the author pays -- of course, not from his wallet, but from the grant they receive from the government -- this is a better model than the one in use right now. I don't know about which model will be in the science world in the future, but we'll see that ourselves. It could be something completely different or maybe research articles will cease to exist and we'll have something like 'research social media.'

Francis: What about yourself for the next few years? Where do you see your own work going? I know you've had trouble having to change servers and things for where Sci-Hub is hosted. Do you think it's going to continue to exist? Is it going to be safe?

Elbakyan: I hope that Sci-Hub will be fully legalized, that is they would stop coming after this project. Now, I have entered graduate school and I am working on the topic "Philosophy of Science". I haven't abandoned neuroscience. Overall, I'm thinking that it would be great to open a research institute and work on projects like this in the future.

Francis: Thank you so much, Alexandra. Thanks, Rachel.

Rachel: OK. Bye.

Francis: Bye.

People have called Sci-Hub the "Napster of science" -- it's a pretty Gen X reference -- but I don't actually think it's the same. Yes, record companies cream huge amounts off revenue from buying music, so they are kind of analogous to science publishing houses, but piracy costs the artists. They have created that music. They have intellectual property rights. Scientists are not paid per journal sale. Scientists don't get royalties from journals, so downloading academic articles illegally makes little financial difference to the scientist.

Now, just so I don't end up in a LegalEagle video again, I better say that I would not advise anyone to break the law, but I do believe it is unjust that science is not publicly available. In fact, when researching this video, I almost suffered a lethal irony overdose when an article about Sci-Hub in the journal Science was hidden behind a paywall. Luckily, I was able to access it.

Aaron Swartz was another brilliant young hacker, who was a co-founder of Reddit and involved in the creation of RSS, Markdown, and Creative Commons. We, internet users, owe him a huge debt. He also downloaded thousands of academic papers illegally from MIT's restricted archive. These were not classified documents. They were not state secrets like WikiLeaks. They were thousands of pages of knowledge, which should have been in the public domain. He was caught. He was aggressively prosecuted by the federal government and charged with wire fraud, facing 35 years in prison and $1 million in fines. He hanged himself at the age of 26.

In the interest of avoiding an unconscious bias, I deliberately sought out the views of the publishing industry and their arguments against Sci-Hub. An Elsevier exec said Sci-Hub has mysterious backers with unclear motives, trying to sound very conspiratorial. "Yes, those evil people paying to make science publicly available. Quick, get QAnon on the case! These people must be stopped!"

A more reasonable complaint is that illegal downloads mean scientists cannot track data regarding their work, how many times it's accessed and so forth. But, surely, you can see that that is a problem created by the current model. If the papers were freely available on journal websites and didn't have to be accessed via a backdoor, you could see all the same analytics that I can see for this video, for example. The most compelling reason is that journals offer extra things in exchange for the money we pay. They have in-house fact checkers, statisticians, editors, and illustrators.

To point #1, let's take a look at Retraction Watch, a website which monitors retracted scientific papers. Just for COVID-19 alone, they have an ever-growing list of papers that were retracted after being exposed to public review. It would be easy for me to cherry-pick the worst journals and hold them up as examples. That would be like saying Hershey's demonstrates that all chocolate tastes like cardboard soaked in puddle water, so instead let's choose the best chocolates there are, the Lindt Lindors of publishing, the New England Journal of Medicine and The Lancet. They are the two biggest medical journals in the world.

In just the last few months, both of these journals, with their crack teams of fact-checkers that cost so much money, have had to withdraw papers that were filled with glaring errors that many people on Twitter were able to spot easily. The Lancet, of course, ran the fraudulent and fabricated MMR and autism paper, the aftermath of which we are still dealing with two decades later and which has directly harmed children through the anti-vax movement. By the way, that was only retracted 12 years after it was published.

Back on Retraction Watch, there is a top 10 most-cited retracted article leaderboard. # 1 is a large dietary study in the New England Journal of Medicine and The Lancet occupies no less than 3 shameful spots on the list.

People like Professor Darrel Francis and Dr. Elisabeth Bik routinely expose scientific fraud in published research, all of which is in mainstream, reputable journals. They're not combing the Azerbaijani Journal of Incontinence, no offense to any Azerbaijani urogynecologists.

But much more common than fraudulent research is research that's just crap. You only have to take a look at the video I made about COVID science to see a few examples. I wanted to concentrate in that video on challenging some of the trials that made worldwide headlines. But if I had cast the net a bit wider, that video could have been hours long. Yet, all of this hot trash got published in the journals that are apparently offering so much value to the process.

I'm not sure quality control, fact-checking, and statistical analysis is really a credible defense. As for illustrations, all the team publications I have been involved with have prepared our own figures. While a journal might adjust it to a house style, the onus of creating data visualization still falls to the authors.

Really, you're left with editors, who do provide an important role in terms of summarizing work and commissioning editorials. But then, again, now we have a wealth of places to obtain all of that. I can find the views of leading scientists on new research on Twitter, blogs, podcasts, or, hey, even YouTube channels.

These are the modern science editorials. The world has moved on from the old model. Why choose two or three anonymous peer reviewers when you can have dozens by opening the paper to public scrutiny? It would be nice if the press would keep their noses out until the revisions have been made, but, yes, I realized I have just strayed into pie-in-the-sky dreamland.

What about some possible ways forward? An alternative model already exists called open-access. This is where the article is available to anyone. Great! But it simply shifts the cost to the authors.

When I first saw the option asking if I wanted the study I was submitting to be open-access, of course I clicked yes, but then I was asked for £1,500. Now, you might say that's not a huge expense for a research team with money in the bank. But just think what a barrier that erects for researchers doing important work in the developing world or smaller independent groups.

In fact, universities say that what they're now doing is paying for the open-access fees and paying for the subscriptions because we're operating both systems at once. Yet, more options exist, namely exclusively open-access journals, which operate with far lower profit margins or no profit margins. But scientists need to be willing to say "No, I'm not going to submit to the 'Super-Flash International Journal of Aren't I Important?' and publish it in this online-only, open-access journal."

For scientists starting their career and trying to build a name, that's not always an easy decision to take. Springer and other publishing houses have said open-access threatens to undermine the whole publishing model. Good job, guys! I think you're getting it now.

Elsevier has been dismissive of alternative options, saying "if you think information shouldn't cost anything, go to Wikipedia." An interesting counter-argument, when you consider how paid encyclopedias have gone the way of Robert Maxwell and his son-in-law.

Now, I have talked about preprints on here before. In my video about the poor quality of science research related to COVID, I was fairly neutral on preprint servers and I still am. Criticizing Facebook for hosting misinformation has a basis, as they are a multi-billion dollar company with immense resources, but preprint servers are small operations who simply can't provide the same peer-review system as journals. It's not really their fault that scientifically-illiterate journalists and armchair epidemiologists uncritically believe everything they read.

Are preprint servers potentially a way forward? Just one of the several dozen I have found, medRxiv, is a preprint server that's seen a 400% fold increase in submissions this year for obvious reasons.

Now, during COVID, this has clearly had enormous and unforeseen consequences, but they represent an option that will free us of the tyranny of publishers. We don't even have to look that far for an even more basic option. Look at this very website. You are watching a video without having to pay for it due to advertisements. If you have any respect for the scientific method, you will give it a thumbs up because views, likes, retweets, and things like that are crude metrics, but they can be built on.

The internet is the ultimate review process. Twitter peer review has revealed so much more about new papers than conventional peer review. I'm obviously aware it's more complicated than that and we would need to have some structure. But if we can figure out ways to rate electronics and cars online, then why not science? Indeed, some of the preprint servers already offer the ability to leave critiques.

The changes required to make some of these things happen are immense. If scientists weren't fearful for their livelihood, with their very employment hinging upon how many papers they publish, they would be less inclined to rush out half-arsed crap. Scientists don't go into science to get rich. They love science. They want to contribute to their fields, not have the dreaded 'publish or perish' sword of Damocles looming over their heads. Publish or perish, as a paradigm, must perish.

Maybe I can speak a little bit more freely than some because I have decided that a career in academia is not for me. I'm not going to be as reliant on the journals for my livelihood.

Now, that's not a knock on academia, it's a comment on me. I am just not cut out for it, I don't think I'm smart enough. I look at my friends in academia and they are the brightest in our ranks. But are they really fulfilling their potential by spending their lives fending off impending doom, by constantly writing grant applications to stay afloat, and jumping through hoops for journals, rather than really pursuing some blue-sky thinking?

It's no wonder people idolize Elon Musk because his billions have funded some scientists to do things they want, bold things like SpaceX, Tesla, or Neuralink. Meanwhile, Bill Gates is also putting his billions behind scientists, but they're tackling problems that disproportionately affect the developing world and hence he is vilified. But, hey, it's the end of the video. What the hell am I doing introducing a tangent at this point?

The great scientists in history were able to have long careers following their curiosity. Fred Sanger achieved the rare feat of winning two Nobel prizes. But in the two decades in between, he didn't really publish very much. He's gone down as one of the greatest scientists of the 20th century, but today he would probably just have been sacked.

, is an interventional cardiologist, internal medicine doctor, and university researcher who makes science videos and bad jokes. Offbeat topics you won't find elsewhere, enriched with a government-mandated dose of humor. Trained in Cambridge; now PhD-ing in London.