ResearchGate, Academia.edu, and bigger problems with scholarly publishing..

So Discover Mag just published an article of mine, a sort of response to a piece in Forbes calling for academics to delete their ResearchGate and Academia.edu accounts. I thought the original piece was awesome and informative, but didn’t go far enough for me, and in fact many of the problems it pointed out where just parts of much bigger issues in the scholarly publishing ecosystem.

Here’s the full, unedited version, which has a bit more information and snark included. Comments on a postcard.


ResearchGateGate. It doesn’t really have a memorable ring to it, but is something we might be hearing more about in the future. A recent article published by Prof. Sarah Bond at Forbes encouraged researchers to remove all of their research articles from the for-profit company, Academia.edu. This has led to a wave of account deletions at the site, and also at ResearchGate, two of the most popular social networks that duel with each other as the ‘Facebook for academics’ of our time.

The issue raised in the article is essentially this: Why should for-profit companies be allowed to generate profits from your research with little transparency? It’s a good question.

Well, actually, this sounds suspiciously like our entire scholarly publishing ecosystem to me, and it is not clear why Academia.edu is being treated in this way. For decades, for-profit companies have been making vast sums of money from the work of researchers, and often with profit margins in excess of 35%, greater than those even of Google (25%) Apple (29%) and even the largest oil companies like Rio Tinto (23%).

The traditional scholarly publishing market is worth an estimated $25.2 billion USD each year, most of which is generated  through publicly-funded researchers giving their work to free for publishers, having that work reviewed for free by their peers, and then having publishers sell each piece of research for around $40 a copy. Researchers in exchange get to have an extra line on their CV and a trip to the pub to celebrate contributing to publisher’s vast profits, I mean, humanity’s corpus of knowledge. This is a vast, global ecosystem that researchers fuel every day, and one that is undergoing quite a state of upheaval at the moment as more and more researchers realise just how daft the whole thing is.

So why are people treating ResearchGate and Academia.edu differently?

ResearchGate are renowned as the ultimate academic spam email machine, often sending unsolicited invitations to be co-author on articles you had nothing to do with, or vague comments about how a grad student from Estonia accidentally downloaded one of your datasets. More recently, emails have even got more ‘clickbaity’, not actually revealing anything useful unless you click through to the site, and much of the time still nothing informative awaits.

ResearchGate even has its own score, the originally named ‘ResearchGate score’, which is perceived by some as a measure of scientific reputation. A study (also available on ResearchGate if you want a tinge of irony) in 2015 found that the score is essentially non-transparent and non-reproducible, fulfilling two of the criteria for academic nonsense, and even incorporates everyone’s favourite metric, the Journal Impact Factor (another proprietary metric), into its calculation. Creating one flawed metric using another flawed one seems like a grand way to go about researcher assessment, doesn’t it. It’s not entirely clear what the score is there for, who uses it, or how it’s used, but there are better ways of doing it than leaving research assessment to a for-profit company, and their proprietary and generally weird metrics.

Academia.edu made a bit of a faux pas last year too, when it was let slip that they were sending out emails to some of their members about the potential for a new service in which papers hosted on their site could be ‘recommended’ by website editors for a small fee. I was actually contacted by someone from the product marketing team about this, being one of them ‘open advocate’ types, apparently. We discussed the idea a bit, and I basically said ‘Don’t do it.’ Scholarly publishing is in an era of massive upheaval at the moment, and I didn’t see this sort of ‘service’ of basically paying for promotion of your work (but isn’t that what we do with scholarly journals anyway..?) being met too well when more important shake-ups were happening. Like that whole ‘Open Access’ thing.

While the site didn’t seem to pursue this idea further, it did lead to the usual ‘Twitter outrage’ over such things, and another wave of account deletions, with the hashtag #DeleteAcademiaEdu.

This isn’t the first time that issues have been raised with either platform, then, so why the new wave of attraction?

The Forbes article states “Moving our papers away from Academia.edu is then about taking possession of our work and deciding what we do with it, rather than allowing a private company to use our scholarship for profit.” Again, it seems to me that this assault specifically on Academia.edu is a bit odd, as the vast majority of our publishing system is exactly that: private companies making money by taking researcher copyright and work and then selling it.

One of the key arguments put forward is that in December 2016, Academia.edu revealed their alternative business model in the form of a premium feature. This provides additional information to users such as who is reading your work, what their academic role, geographic location and university are, as well as the source directing them to your work are. The Forbes article argues that this promotes academic class politics and hierarchical stratification even more, and is quite right in doing so. It does not and should not matter about the ‘rank’ of who is using your work, but how they are using it and why. That’s the sort of service worth paying for, and actually something tools like Altmetric already give you for free.

Additionally, the article argues that the platform now has a policy which means that the site can collect and evaluate data provided by users, and possibly then sell onwards. Again, it is not clear how this is any different from any publisher or journal which harvests data based on the content researchers freely provide for it. Except that ResearchGate and Academia.edu are free, make no demands on author rights, and certainly do not sell your research articles.

But by all means, if this concerns you then delete your accounts. But you should probably also then stop giving your research away for free to private publishing companies too. And delete your Facebook and Twitter accounts too, while you’re at it.

If anything, these data analytics on both platforms provides a valuable service to researchers, pending what I said about the ResearchGate score above. Both provide metrics on article re-use that are useful for researchers in seeing how their work is being digested by the community. ResearchGate even provides citation scores now too for researchers, similar to Google Scholar and other for-profit platforms like ScienceOpen. And all of them do this for free to users, removing some of the domination over citation metrics that Web of Science and Scopus, both premium and privately owned services, used to have.

Richard Price, the CEO of Academia.edu, has even statedThe goal is to provide trending research data to R&D institutions that can improve the quality of their decisions by 10-20%.” That sounds pretty useful to me for a lot of different stakeholders, including researchers themselves.

And I guess one question is, so what if they are making money from publishing data? If someone sees an opportunity in making large-scale assessments about scholarly publishing and research in general, isn’t that a good thing? One of the main reasons why we publish is so that other people can re-use our work, including on a large-scale. Publishers don’t pay researchers for giving them their work, so it remains unclear to me again why this should be viewed as different for ResearchGate and Academia.edu. Except that these platforms seem to legitimately give something of value in return beyond a brand name.

As such, I find the arguments in the Forbes article not particularly convincing against either platform. Any argument against both, and their relationship to academia and research in general, seems to ignore the context and the bigger picture of our enormously broken scholarly publishing system. Things have got so bad, that whole universities and countries are now taking a stand against the profiteering nature of some publishers. Making money while improving the overall system of scholarly communication is feasible, and none of the arguments put forward convince me that either platform is in actual conflict about this.

Dark sharing versus Open Access

ResearchGate now call themselves an ‘online information society service’. Whatever that means. I can’t see researchers exactly stating that as the primary reason why they use it. Many researchers see it more as a digital CV or business card (you can request an article discussing this via ResearchGate, the original is paywalled). Most also use it for sharing their research articles, data, or projects.

It’s no great secret that a large chunk of the articles both platforms host are done so illegally. The final published ‘version of record’ is usually prohibited by publishers from being publicly shared, even by the original authors, as many require that authors transfer over all of their rights in order to be published. While this practice is in itself questionable, this does not legally justify the large-scale copyright infringement that is so apparent on either site, irrespective of how useful it might be to authors. This is even enhanced by Google Scholar, whose search algorithms preferentially point you to free versions available on either platform.

A consequence of this is that a couple of years ago, one of the biggest scholarly publishers, Elsevier, and one perhaps not held in the highest regard by many academics, sent 2,800 DMCA takedown requests of articles it published that were illegally hosted on Academia.edu. While this was a generally bad PR move for both Elsevier and Academia.edu, Elsevier were technically fully within their rights to do so. One over-arching problem here is that ResearchGate and Academia.edu are not accountable to anyone but their shareholders. When questioned about this illegal file hosting, they can simply wave their hands and say it has nothing to do with them and it’s down to the individual choices of their members. Responsible, isn’t it. In the meantime, they can both keep using this illegal content to enhance their data analytics, which is perhaps more of an issue than what they then choose to do with such data.

One major issue here is that you can easily be fooled into thinking that this sort of ‘dark sharing’ with ResearchGate and Academia.edu is a good compromise for doing Open Access publishing right. Well, it’s not. If anything, it can detract from Open Access developments by undermining the impetus for it by providing a quick shortcut that superficially resembles the same thing, but entirely lacks the stability and management of a journal or repository system. This also means that Academia.edu is governed by different politics and ethics to that of an institutional repository, although it is not immediately clear how this non-equivalence is necessarily a bad thing at the present.

But at the end of the day, even Open Access is being used as a way for publishers to make additional money from your work. While around 70% of journals indexed by the Directory of Open Access journals do not charge to publish, the majority of large publishers who control the journals which researchers often have to publish in to receive formal recognition, often charge in excess of $3000 to not make money off your work through subscriptions. That doesn’t make much sense either, does it?

Posting to Academia.edu is no more difficult than freely posting to an institutional repository, yet with more than 47 million members at Academia.edu (apparently), you have to suspect that ease actually has little to do with it. In fact, it is probably the ‘Facebook-ness’ of Academia.edu that makes it so appealing – academics precariously use it as a professional advertising tool, and in an academic environment where egotism and self-marketing is rewarded more than sharing, it is easy to perhaps see why one is more popular than the other. After all, the best reward for self-archiving your work is only that warm feeling knowing that you’ve done the right thing, and you can’t put that on your CV.

All is not lost

There are a swathe of institutional repositories out there that have one job: to make your work Open Access in a manner that is compliant with research funding mandates and publisher policies. The great thing about this is that it’s easy, and everyone benefits from it without a penny exchanging hands. Except for the librarians and support staff that administer the repository, anyway. The institutional repository system isn’t without its flaws, with some still requiring institutional logins to access, for example, but they certainly offer a more sustainable option for researchers at the present.

There are a host of other subject-specific or cross-disciplinary repositories too. These include Zenodo, a non-profit and funded by OpenAIRE, the European Organisation for Nuclear Research, and CERN, to help super-collide your research, or the arXiv, which has been hosting articles since 1991. Ethan Gruber has even recently launched a tool that transfers all of your content between Academia.edu and Zenodo, for those interested.

For me, I deleted both of my accounts through redundancy as I was just not seeing the value in them. I have already made all of my research openly available through my institutional repository at Imperial College, as required, or available at Open Access journals.

Similarly, Guillaume Cabanac of Université de Toulouse, who originally alerted me to the Forbes article, said “All of my papers are available through my lab’s repository and other local and national repositories. I also try to keep a CV up to date with all DOIs and links to self-archived PDFs on my webpage. I don’t see the point of pushing them on these privately owned platforms too. Now I’m educated about Open Access, I decided to leave.  The fate of my profile was in my hands.”

For those not in such a position to be able to do this, there might still be a great appeal in using both platforms though for sharing of articles. Academia.edu and ResearchGate can be an option to help level the field a bit: for sharing non-traditional research outputs, for authors who can’t afford to publish Open Access, for those who don’t have access to an institutional repository or have an institutional affiliation, for those who don’t remain within academia but still want to preserve their research articles in a community space.

Being in a position where you can delete your accounts, therefore, is actually a position of academic privilege, and telling other authors to do so could be inconsiderate of their position and status. And for authors in situations like those above, I get the feeling that they won’t be too concerned about complaints from publishers about how they’re allowed to use their own work, or that the platforms are for-profit.

So I think a lot of the angst towards Academia.edu and ResearchGate might be better placed elsewhere. We have an entire scholarly publishing system that is largely fuelled by taxpayer money, but governed and constrained by private interests, and that should be something of much deeper concern. ResearchGate and Academia.edu are really just small-fry in this vast sea of profit-seeking.

The real question is what do we do when private interests actually start to interfere with those of the public, as many scholarly publishers actively do by prohibiting access to research – indeed, this is how they make their money. I don’t see Academia.edu and ResearchGate doing that, or at least not to an extent that is greater than any other for-profit company involved in scholarly publishing.

If you want to actually do something useful, choose Open Access, and share your research far and wide. Just don’t lock it up.

Author note: Thanks to Lisa Matthias and Penny Andrews for discussions on this topic.


Leave a Reply