The last few weeks have brought a series of opportunities to think “well now that’s interesting – what’s going on?” about the relationships between ResearchGate and scholarly publishers and today was no exception. I’m going to use this blog post to chronicle the events for my own clarity of thought and then share some observations/ask some questions. Let me say at the outset that I have no insider knowledge – everything I know about these relationships is based on public documents and articles.

STM Letter to ResearchGate and the Principles

On September 15, the International Association of Scientific Technical and Medical Publishers (STM) sent ResearchGate a letter stating that “STM’s members now have devised a solution for you that would enable ResearchGate to operate in a way that would be consistent with these principles” (the principles being those articulated in the STM Voluntary Principles on Article Sharing [PDF]). After more detail, the letter concludes “If you fail to accede to this proposal by 22 September 2017, then STM will be leaving the path open for its individual members to follow up with you separately, whether individually or in groups sharing a similar interest and approach, as they may see fit.”

As I read the letter, I couldn’t help but ask why ResearchGate would want to comply with the principles. As far as I can tell, ResearchGate did not participate in the consultation that led up to the development of the principles (disclosure – I did submit comments to the process) and has not made any moves towards signing on to them. This could be taken as evidence that ResearchGate is taking the “voluntary” name seriously and declining to volunteer to be regulated by the principles?

Is there an assumption that these principles have gained some sort of universal acceptance or is this an attempt to assert that the principles have a kind of objective status? Publishers have indeed endorsed the principles but I doubt that most scholars or even journal editors are aware of them at all. I’ve written about some of the issues I see with the principles from a library perspective (Substantial and Enduring Roles for Libraries in Article Sharing – Part 1 and Part 2) and have been asking about whether there will be any auditing of publishers who have made endorsements to see if their policies comply (current answer is that STM will review a publisher upon request, which to me isn’t an audit, but I digress). I see that there are scholarly collaboration networks that have endorsed the principles; however, as SSRN and Mendeley are both owned by Elsevier, the level of acceptable among scholarly collaboration networks generally seems less than universal.

I also noted that the STM letter itself was not hosted on the STM website but instead on Elsevier’s and on ACS’s (American Chemical Society). Also, I noticed the law firm that sent the letter on behalf of STM is associated with STM’s Copyright and Legal Affairs Committee. Does this mean that STM no longer has a working group on scholarly sharing and the principles? I was unable to locate one by browsing or searching just now. I notice that STM does have an Enforcement Task Force that is the “enforcement arm” of the Copyright and Legal Affairs Committee. Are the principles now being treated as an enforcement mechanism?

ResearchGate Response

I’m not sure it is accurate to use the word “response” in that mostly it appears that, at least publicly, ResearchGate did not respond to the letter. None of the articles that I saw reporting on the letter included any comment from ResearchGate. The Times Higher Education reported that ResearchGate declined to comment and resorted to quoting from a previous statement of ResearchGate’s founder and chief executive, Ijad Madisch:  “has previously said that he “wouldn’t mind” if copyrighted material was removed from the site, as researchers could continue to share papers privately.”

An undated STM webpage states that “The International Association of Scientific, Technical and Medical Publishers (STM) has written to ResearchGate, the scholarly collaboration network (SCN), to make an offer to work with the platform collaboratively to bring the site into compliance with copyright. It is with regret that the core proposal suggested in this letter has been rejected by ResearchGate.” I’m unclear from the phrasing whether ResearchGate stated its regrets or that STM regrets to report the rejection this to its membership?

Enter the Coalition

The STM webpage also states that “The Association supports all of its members as they take forward their own discussions and actions directly with ResearchGate.” Granted, the original letter from STM to ResearchGate stated that “If you fail to accede to this proposal by 22 September 2017, STM will be leaving the path open for its individual members to follow up with you separately, whether individually or in groups sharing a similar interest and approach, as they may see fit.” Yet, this still seemed a rather quick retreat by STM. Did STM expect its proposal to be rejected? Was the letter only intended to lay groundwork or to try and get ResearchGate to tip its hand?

Regardless of the letter’s purpose, as a next act in the process, a group of STM members did indeed follow up and establish the Coalition for Responsible Sharing. According to the Coalition Statement, “The coalition members include the American Chemical Society, Brill, Elsevier, Wiley and Wolters Kluwer. These organizations will begin to issue takedown notices to ResearchGate requesting that infringing content be removed from the site.” Interestingly, the Coalition Statement also says that “Concurrently, The American Chemical Society and Elsevier are asking the courts to clarify ResearchGate’s copyright responsibility.” I think that means that ACS and Elsevier are pursuing this separately from the Coalition’s collaborative project of sending takedown notices? I’ll note here that Robert Harington wrote a useful piece on Scholarly Kitchen about the Coalition as well based on an interview with James Milne, Senior Vice President of ACS and chair of the Coalition.

Was I the only person who wondered why the Coalition was made up of so few STM members and so few endorsers of the principles? I think of STM as the organization through which publishers choose to collaborate because of aligned interests even though they are in heavy competition with one another. Does this signal a change in status of STM as having a leadership role in the industry?

With respect to the takedown notices, is the underlying claim that there is so much infringement on the ResearchGate platform it is an undue burden on the publishers to have to issue takedown notices? Is this an attempt to begin to overturn the established (through multiple court cases and laws) and also generally process for addressing copyright infringing content – i.e., that a platform is not required to pre-screen but instead the requirement is to respond to notices of infringement?

Given that the original STM letter posited a system for monitoring copyright compliance “could be implemented within 30-60 days,” how defensible is the claim that sending takedown notices is not a viable solution? I wonder if takedown notices would be “highly disruptive to the research community” as the Coalition Statement says? Given how often scholars already cope with intermittent access to publications, I can imagine it might drive more scholars to SciHub; however, I wonder if it would be any more disruptive than when libraries are forced to drop subscriptions because of rising journal prices?

I also think Harington is correct that “the popularity of ResearchGate with users is a potential public relations problem” for the publishers issuing takedown notices.  Is it likely that scholars would put the blame on ResearchGate or on the publishers? Thinking back to Elsevier sending takedown notices to Academia.edu, it seems scholars put the blame on Elsevier, no doubt encouraged in that view by the messaging from Academia.edu? One could imagine ResearchGate might have its own messaging as takedown notices are received?

Copyright, Document Integrity, and Analytics

I don’t think it takes much time browsing on ResearchGate to conclude that there are many PDFs posted on the site that likely infringe copyright ownership of publishers. To upload a file to ResearchGate and make it public, a scholar must click agreement that they have the rights to do so, a process similar to uploading to library institutional repositories as well. One suspects that scholars are not reading, not understanding, or not caring as they click to agree. I have no doubt that the takedown notices, once sent, will result in content being removed from public view.

I have found the additional charge against ResearchGate more puzzling. Harington’s Scholarly Kitchen piece states that “there are reports that ResearchGate has stripped out metadata from papers, rebranded them, and altered links in papers to point to their own hosted versions of papers, rather than the original journal” and the Coalition Statement says “ResearchGate often also substantively alters articles for the same purpose [to generate traffic to its site], and where corrections or retractions are issued, it fails to update articles accordingly on its site, undermining research integrity.”

Interestingly, bepress (recently acquired by Elsevier) also rebrands PDFs with a cover page upon download (side note – this can be toggled on/off by a scholar preference setting on ResearchGate but, in my experience, is not a scholar controlled option in bepress). It will be interesting to see if bepress practice is changed now that it is Elsevier owned and Elsevier is involved in issuing this complaint against ResearchGate.

I’m not sure what to make of the claim that ResearchGate alters links in papers. I have not encountered this myself and no examples have been provided in any of the write-ups. It is the case that ResearchGate, like other platforms (e.g. Microsoft Academic), does list the references from a publication and links to records on the platform for them; however, unless I am missing something that does not alter links in the PDF. Has anyone encountered altered links in a PDF? I would be very interested in an example so I can understand this charge. The claim about not tracking and uploading corrections/retractions seems a bit strange given the overarching charge is that ResearchGate hosts copyrighted content without rights and this then charges that ResearchGate fails to do so (i.e., find and upload copyrighted content) when it should?

I wonder if another issue driving concern from the STM/Coalition members is that the publishers are not able to access and monetize the analytics that ResearchGate use generates?  The principles state that “Publishers and libraries should be able to measure the amount and type of sharing, using standards such as COUNTER, to better understand the habits of their readers and quantify the value of the services they provide.” One can imagine that ResearchGate also sees the value in these data given its calculation of an RG Score for each scholar, a score that reflects not only citations to the work but also a scholar’s activity on the ResearchGate platform. Positing a future in which copyright material is only shared privately on ResearchGate, which is in keeping with the principles, would publishers be interested in purchasing this analytics data from ResearchGate? Is this an eventual source of revenue for ResearchGate if it can ride out the copyright takedown and lawsuit actions?

Enter the Springer Nature/ResearchGate Press Release

So, we finally make it to today’s revelation. In a joint press release, ResearchGate and Springer Nature issued this statement: “ResearchGate and Springer Nature have been in serious discussions for some time about finding solutions to sharing scientific journal articles online, while at the same time protecting intellectual property rights. The companies are cautiously optimistic that a solution can be found, and we invite other publishers and societies to join the talks.”

Given that is the entirety of the joint statement, it is not surprising that questions start forming immediately. What does “for some time” indicate? As a reminder, the letter from STM to ResearchGate was dated September 15, which was only 24 days ago. Do these discussions pre-date the letter? What to make of the claim of “serious discussions” rather than only “discussions”? Mild speculations of implications for strategic acquisition and mergers also arose.

I couldn’t help but recall Harington’s question and Milne’s response in the Scholarly Kitchen essay: “I asked Milne if there were any similarities to the early days of YouTube, and whether the ultimate solution would be a revenue-sharing arrangement along the lines of that site. Milne said that none of the publishers involved have explored revenue generating arrangements under agreements to license content to ResearchGate.” Could it be the case that the Coalition members are not exploring revenue sharing but that Springer Nature is?

Perhaps the most humorous response I saw on Twitter today: “What will Elsevier think?” I was also wondering “so just how tense is the next meeting of the STM Copyright and Legal Affairs Committee going to be given Elsevier, ACS, and Springer Nature as members?”

Not an Apologist

Lest I leave an inaccurate impression, I want to also add here that I’m not an apologist for ResearchGate.

I have had a ResearchGate account since April 28, 2014. I’ve saved almost every email I’ve received from ResearchGate since that time. I have similar archives related to my profiles across the major scholarly collaboration networks because I teach Manage Your Online Scholarly Identity to Maximize the Reach and Impact of Your Work as an open workshop in the University Library at the University of Illinois at Urbana-Champaign and also provide customized versions for departments, research groups, etc. The only effective way that I have found to be able to teach faculty, postdocs, graduate students, etc. about these systems is to use them.

I have questions about the development trajectories of all of the platforms, questions about how they communicate with users and potential users, questions about how they position themselves, questions about how they compete with library/institutional repositories, etc. I definitely have these questions about ResearchGate. I also wonder if publishers might turn their attention to library/institutional repositories, sending letters offering collaboration before strategic takedown notices and lawsuits.

Ultimately, I also recognize that ResearchGate, more than any other scholarly collaboration platform at the moment, is seen by the scholars with whom I interact as providing great value. Even when they are annoyed by ResearchGate, they value it and tell me the annoyance is worth it for what they get out of it. If that value is disrupted, I do not assume that they will turn to librarians for guidance but rather that librarians will need to be poised with outreach and strategies to meet faculty needs. As such, I’m not a ResearchGate apologist but I do find it critical to track on these developments.

What will tomorrow bring?

Addendum: 

Not “tomorrow” but a few days later (October 16) my email brought me this response from Jim Milne, Chair of the Coalition for Responsible Sharing. I publish it here as an addendum to this blog post with his permission. As context, the “separate note” that Jim mentions was a posting I made to the LibLicense-L discussion forum.

As Chair of the Coalition, I also have questions—mostly around what ResearchGate is doing and thinking with respect to journal articles, given recent events—but perhaps we will all learn their thinking soon. There are a lot of things that I can’t comment on because of the legal process, but there are a few things I think I can say.

On the STM association Voluntary Principles—we don’t see these as an onerous demand on Internet sites that want to host or post non-OA articles published in scholarly journals. In return for being able to post the articles, such sites would provide the journals with usage data to enable this data to be included in Counter reports. I do agree that more sites should genuinely consider modeling their business processes on the offer outlined in the Principles.

However, I’d like to also point out that the STM offer, which was outlined in the 16 September letter sent by STM’s outside adviser, went far beyond asking ResearchGate to adopt the voluntary principles—it also provided a compromise allowing some existing material to stay available while proposing an automated checking system for newly uploaded content.

Regarding modification—I believe you’ve seen the comments on Scholarly Kitchen and an example of how they change article content. ResearchGate’s activities continue to change over time, and at the moment, a lot of the modification pertains to the cover pages ResearchGate attaches to each article.

I have also asked about your point regarding bepress; it has a standard feature that enables institutions, if they wish, to add a cover page on preprints and manuscripts posted on their repositories. The cover page is created by the institution/author; on the contrary, ResearchGate adds cover pages itself that include prominently the ResearchGate brand, detracting from the publisher and journal representation.

I also saw your separate note questioning how Coalition members “noticed” copyright infringing articles that have recently been taken down by ResearchGate; you also asked about what constitutes a “significant” amount of articles that have been removed. To answer the first question: we regularly check which articles are available on ResearchGate, and in doing so, we noticed the change. We believe ResearchGate has recently taken down thousands of copyright infringing articles. Exact numbers are difficult to ascertain, as they vary greatly from publisher to publisher. They also keep changing on an ongoing basis, and ResearchGate has given neither authors nor the Coalition any information about what content or how much it has taken down.

Finally, a quick note to reiterate that as a group of societies and publishers, the Coalition believes that we all want the same thing—for scholarly collaboration networks to understand and respect the important value publishers bring to the record of science.

Hope this response has been helpful.

Best, Jim