Ever since Springer Nature and ResearchGate announced their cooperative agreement this past April, many have wondered what exactly the “sharing of articles on the scholarly collaboration platform in a way that protects the rights of authors and publishers” might look like.
Today, we get our first glimpse. Springer Nature and ResearchGate have announced that “full-text articles published in select Nature journals since November 2017 will be rolled out to researchers’ ResearchGate profiles starting now and completed by March 7, making it easier to read or download research on or off campus from that moment on.” I had a chance to speak yesterday with Steven Inchcoombe, Chief Publishing Officer at Springer Nature, and Ijad Madisch, CEO of ResearchGate, about this project.
Though small in scope, the importance of this project should not be overlooked. This pilot project represents the first significant experiment with the syndication of publisher content to a content supercontinent. My fellow Scholarly Kitchen contributor, Roger Schonfeld, has been tracking this emerging strategy and exploring it in recent months.
The Project
Approximately 6,000 articles published from November 2017 onward in 23 Nature research journals will be uploaded to ResearchGate to the relevant author profiles. The journals included are all subscription-only titles that do not offer an article-level Gold APC option. The open access option for these titles is green deposit with a six month embargo. But, when the articles in this pilot are uploaded to ResearchGate, the copies uploaded will be the version of record and they will be available during the pilot without any access controls.
The journal titles are:
- Nature
- Nature Astronomy
- Nature Biomedical Engineering
- Nature Cell Biology
- Nature Chemical Biology
- Nature Chemistry
- Nature Climate Change
- Nature Ecology & Evolution
- Nature Energy
- Nature Genetics
- Nature Geoscience
- Nature Human Behaviour
- Nature Immunology
- Nature Materials
- Nature Medicine
- Nature Methods
- Nature Microbiology
- Nature Nanotechnology
- Nature Neuroscience
- Nature Photonics
- Nature Physics
- Nature Plants
- Nature Structural & Molecular Biology
The initial pilot is planned for 3 months though it may be extended based on how it progresses. During the pilot, Springer Nature and ResearchGate will gather data about how and how often the articles are discovered, accessed, and used as well as feedback from authors and readers about their perceptions and experiences. To stave off any worries for conscientious authors aware of their copyright terms with Springer Nature, authors will be alerted that these copies are legitimate to have publicly accessible from their ResearchGate profiles. And, to respect author freedom, authors will have the option to remove them from their profiles if they prefer.
Negotiating Power
In my view, the most notable aspect of this arrangement is that Springer Nature is not only allowing publicly accessible copies of subscription-only content but also allowing the upload of the version of record copy onto ResearchGate. This means that anyone can also download the version of record; it is now “in the wild” through ResearchGate.
Springer Nature is uniquely positioned to have taken another path. Springer Nature already offers the SharedIt service, which is powered by Digital Science’s ReadCube, for making copies of specific articles available for public access and reading without uploading them onto other platforms. For this reason, I was personally predicting that this collaboration would utilize an inline reading copy built off of SharedIt. It is not at all clear why Springer Nature would prefer to have a downloadable version of record available through ResearchGate rather than a more controlled inline version.
As such, I have to wonder what this signals about the strength of each partner’s negotiating position. Though a top publisher, Springer Nature is facing challenges that all publishers are currently facing, e.g., Sci-Hub piracy and the open access demands of Plan S, but also some that are unique. In contrast, ResearchGate is the most visited science website in the world, with significantly more traffic than Google Scholar, and it has more than 15 million researchers in its social network. That is a tremendous position of strength from which to develop projects that are aligned with its interests. While nothing in this pilot obviously precludes a future based on SharedIt, it is striking that the parties went ahead with a more radically open approach.
Legitimizing ResearchGate
So, what does ResearchGate get from this collaboration with Springer Nature? In some ways, the answer here is obvious; ResearchGate gets content and content drives more engagement on ResearchGate. Madisch emphasized to me that ResearchGate needs to work with publishers in order to serve researchers, observing that publishers play an important role in the spread of scientific information and the journal article continues to be the mechanism for formal dissemination of results.
Of course, the value for ResearchGate goes well beyond being able to load a few thousand pieces of content. Working with a well-established publisher legitimizes ResearchGate and distances it from a piracy reputation. This may be particularly useful as ResearchGate defends itself against ongoing lawsuits filed by Elsevier and the American Chemical Society.
Opening Springer Nature
More puzzling may be the question of what Springer Nature gets from this collaboration. Springer Nature is clearly very committed to this partnership. Inchcoombe emphatically stated to me that “Springer Nature’s view is that ResearchGate is a legitimate platform and a platform we want to work with.”
As I read Springer Nature’s Plan S feedback, I had a strong sense that it is seeking to position itself for the future as the leading publisher in open science/open access. But, the feedback also makes clear that there is not a financial model that supports Springer Nature making an immediate transition from subscription to open access publishing and that there is a need to drive demand for open access publishing.
That is admirable positioning, but what is really going on strategically? Is Springer Nature observing the traffic leakage to sites like ResearchGate and looking at how to account for this usage in order to ensure that the value of its library licenses do not appear to decline precipitously? Or, is Springer Nature actually playing a longer term game of integration with the scholarly communications networks in contrast to Elsevier’s choice simply to buy Mendeley and integrate it into the stack? A rumor has been spreading that Springer Nature might one day purchase ResearchGate. But, if it did that, would it find it just as challenging to position ResearchGate as a neutral community space as Elsevier has found it to be with its Mendeley? Instead, might Springer Nature be looking to build up ResearchGate over time as a neutral site — a supercontinent — where all publishers can place their content and receive the benefits of ResearchGate’s massive traffic?
Balancing the business model realities with their longer term aspirations is perhaps a tricky position for Springer Nature to be in — wanting to drive open channels while also maintaining the library subscription channel for as long as possible. Could it be that experimenting with content syndication to ResearchGate, while holding out the possibility for adding usage controls in the future, is indeed the strategic move?
Note: My thanks to Roger Schonfeld for detailing out the syndication model over the past few months in posts here on the Scholarly Kitchen and his helpful comments on a draft of this essay.
Discussion
58 Thoughts on "Springer Nature Syndicates Content to ResearchGate"
As a short term pilot project, we can be assured that strategists at Springer Nature will be carefully measuring the impact of this undertaking before determining how to proceed.
For this reason, the design of the pilot is especially interesting. With subscription content from two dozen Nature journals being made freely available on another platform, there are two major variables at work. First, there is the enormous amount of traffic moving through and across ResearchGate, far more so than on Nature.com. Second, there is the move to open these articles to public access when previously they had been subscription only.
This seems like a significant choice, in terms of the experimental design of the pilot. In the standard syndication model we would not expect articles to be made available on a public basis but rather only to users with subscription entitlements. At a minimum, as mentioned in Lisa’s excellent article, we might expect inline reading rather than downloadable files. What exactly is Springer Nature testing out in this pilot then?
At a minimum, this should give Springer Nature clear evidence of the effect on their own platform usage when VOR versions are made available at a third party site. A hypothesis would be that on-platform usage will decline, replaced to some degree with Research Gate traffic. Of course testing such a proposition might have value, because if that does not prove to be the case, then all the current fretting about traffic leakage may be misplaced. And in such a scenario, if usage data on ResearchGate can be gathered and monetized through content licensing, without paying the price of declining on-platform use which might threaten those very licenses, then there could be an enormous win in all this for a publisher.
Is it altogether too far-fetched to think this might be on the table? If not, why is the pilot not using an entitlements service or inline reading?
Inline reading is in direct contradiction of the business model of sites like ResearchGate. If your business model is entirely predicated on showing users advertising and spying on users, then any functionality that takes the user off of your site means less ads shown and less user data collected. That’s why they’re generally insistent on hosting the content themselves.
Inline reading leaves the user on the site though. And, I’m pretty sure that ReadCube, for example, collects an awful lot of analytics … probably to what pages are viewed of the whole, etc. If RG can share data with SN, presumably the data could flow the other way too. Which is what got me to wonder about negotiating position.
Does it? My understanding from what I’ve seen is that the content is framed within ReadCube, removes journal branding, and shows the reader ReadCube advertisements and links. Perhaps RG has more sophisticated data capture ability (given they’re the entity with whom the reader has a membership)? If the ad revenue and the data are RG’s bread and butter, how much could they possibly share with publisher partners (given that at some point they need to make money)?
Perhaps worth noting that RG’s terms of service prohibit users from posting anything “whose primary purpose is to drive Users to links outside of ResearchGate”, so essentially any non RG links are not allowed.
Of course that is not a statement about what RG itself is allowed to do.
Right, but it speaks to their overall model which is built around keeping traffic within the confines of their own advertising and data collection arena. It’s why offers to RG to provide free access to linked articles within journals were rebuffed.
Oh, yes, that is definitely the model. But, now I’m confused, what offers to linked articles are you referring to?
I can’t speak for other publishers, but in many cases where we’ve done takedowns, we have offered both the author and RG a direct link to a freely available PDF version of the article in the journal, but at least so far, there’s been no interest in using them to supply the same functionality to users.
More confused. You (your publisher employer) are sending takedown notices for open access articles?
Freely available, but only the OUP platform so you can capture the usage data, etc.?
Wouldn’t that be a perfect situation for syndicating the content and implementing distributed usage logging? Likely more reading while OUP still gets the analytics? Of course, I am aware that OUP is in the Coalition for Responsible Sharing with Elsevier and the American Chemical Society so I imagine there is some aspect of this that doesn’t work with the OUP business model.
DUL would be nice, but remember that providing COUNTER stats to subscribing institutions only represents part of the value of website traffic. There are also things like advertising (usually impression-based), collecting usage data on users from institutions that are reading the articles but not subscribing (as potential sales targets), offering related articles and recommendations that can either help drive non-subscribers to subscribe to the journal or subscribers to find new products that they might also subscribe to, not to mention all the companies out there training their AIs and building their own data analytics products.
Third parties reusing one’s product to drive further revenue can be a good thing, but those third parties should then help support the generation of the content upon which their business relies. It makes sense long-term, as poisoning the well and destroying the sources of material upon which you rely will eventually leave you without anything to offer your own customers. But more importantly from the not-for-profit academic community, any money we can bring in from outside of academia means cheaper prices for libraries, funders, and individual researchers. Every penny that a commercial, for-profit third party extracts from the system means a penny that has to be made up for from an academic source.
Syndication could work if one could arrange a real partnership with shared costs and shared benefits, but so far I don’t see that happening here.
It’s an interesting time to be an observer of this. I suspect Springer Nature is not anticipating that this pilot resulting in “poisoning the well and destroying the sources of material upon which you rely” but there is definitely risk in any new venture. Only time will tell if other publishers will be relieved that they didn’t take this risk or regret their lack of strategic action!
As you speculate in your post above, I think this is a data-gathering exercise (if we do this, what happens?), rather than a glimpse at the actual strategy that the data would lead to.
I think that Roger was detailing that out more specifically in his comment on how the assessment might be designed than I was emphasizing that in my post. But, yes, clearly there is data gathering. I don’t think it is disconnected from a strategy though, specifically that of bolstering/preserving the library channel long enough for open access business models to become sustainable in order to emerge as the leading publisher in open science/open access.
I can no longer reply to the later messages in this subthread so I am not sure if my reply will mess up the threading. Sorry to be late to the party here today.
Let me just say that there ABSOLUTELY are trade-offs in contemplating a syndication model, including the balance between losing out on analytics/advertising vs. increasing tracked usage. And these will naturally differ across different categories of publishers, based on their existing portfolios, ability to generate starting point usage traffic, and other factors. And the strategy here is not just a question of whether to participate as a source of syndicated — many publishers will wish to examine whether they themselves might become supercontinents themselves. I explored all this in the piece this fall. Here are two key paragraphs:
“Nonetheless, a major publisher would see that there are dilemmas associated with enabling the creation of these new businesses. For one thing, there could be seen to be an opportunity to turn every publisher site into a Supercontinent, so I asked Henning Schoenenberger, Springer Nature’s Director of Product Data and Metadata, whether publishers will eventually distribute competitors’ content through their own sites, i.e., turning SpringerLink into a multi-publisher content access platform. He said, “Springer Nature can’t speak for other publishers but fully intends to continue to invest in its own platforms (ie SpringerLink and Nature.com) and will continue to develop services and tools that will add value to the specific content that we publish.”
“I wondered about some additional risks. I asked Schoenenberger if he was concerned about giving an analytics advantage to platform providers who will have more granular usage data than publishers. He said, “You can say that there is some risk that we open up some advantage to a competitor but this is an industry initiative to facilitate access and transparency in research. That balance is well done.” From conversations with other publishers as well, I am convinced that there is broad agreement that an appropriate balance can be struck.”
https://scholarlykitchen.sspnet.org/2018/10/15/syndicate-content/
This is a very interesting development. My feeling is that Springer Nature might still just go ahead and buy out ResearchGate (not sure of the possibilities) to tap into its huge traffic. With Open Access (read as declining revenues) cracking on their heels, publishers will continue to look for alternative routes to engage (AND monetize) the scientific community, and ResearchGate presents a golden opportunity. Once Open Access becomes the norm in the near- to mid-term future, the entire game will probably be about which platform is more innovative and efficient in providing search and access to different output formats for scientific communication, regardless of the publisher.
I am reading this just after reading that the University of California has decided not to renew its subscriptions with Elsevier. It’s easy to underestimate the possibility of large, powerful commercial enterprises to shrink into insignificance.
There is real fear of losing the library channel altogether due to leakage to SciHub, ResearchGate, etc. This is what has happened, at least for now, for Elsevier content with Germany and California. While Elsevier has sued ResearchGate, that hasn’t even yet had the effect on leakage of putting a finger in the dike. Through this pilot, depending on how it develops into an eventual model, Springer Nature provides a real alternative to leakage. That’s what syndication is all about.
The question would be the business model behind that syndication. We know that RG runs at a deficit. How would Springer Nature continue as a business if they’re essentially giving away their publications?
Some might wonder how publishers will continue long term if they don’t?
David, the question is how it develops as an eventual model. I speculated in another comment that SN may be testing out the effects of free distribution through RG on their S/N platform usage and value proposition. If the effects are minimal, then maybe this actually can work. On the other hand, I think many of us believe that an entitlements service will eventually be layered on to content hubs like RG, ensuring that only users with licensed entitlements can access the VoR. Those are two routes through which syndication can preserve the subscription business.
Or, at least preserves the subscription channel for awhile – and from the publisher view hopefully for long enough for open business models to fully develop.
FWIW, indeed I did say we are on the path to a global flip – it is a matter of when and how not if in my mind: https://scholarlykitchen.sspnet.org/2019/01/22/celebrating-30000-open-access-articles/
One other thing that makes this particular experiment less risky than it might seem. Unpaywall, as an example, but there are others, refuses to index ResearchGate as a target for its links to public content (see https://unpaywall.org/faq). Unless these tools – which are used by libraries to connect users to open content – start to index ResearchGate, the library-to-content pathway will not result in a user connecting to these uploaded PDFs. That potentially helps preserve the library channel a bit more than it might initially seem.
Trying to picture the endgame here. I suspect it’s a question of how one sees ResearchGate, to paraphrase Joe Esposito, is it a business, a product, or a feature (https://scholarlykitchen.sspnet.org/2015/01/27/when-is-a-feature-a-product-and-a-product-a-business/)?
Right now we know that RG runs at a deficit, and that their publicly stated business model is built around advertising. If one looks at the entire advertising market for scholarly journals, it’s around $400M annually (https://www.stm-assoc.org/2018_10_04_STM_Report_2018.pdf). If it were able to capture 100% of the market, could you run RG annually at a profit on $400M revenue per year? An estimated 90% of that ad revenue is from print advertising. Could you run RG annually at a profit on $40M per year? Neither of those figures offers much in terms of paying to license lots of content from publishers or sharing revenue.
As an alternative model, surveillance capitalism is a possibility. RG could (and likely does) spy on users and collect data about what they’re reading and saying to one another. Is there a market for this data? Given that Facebook and Google and Amazon are likely already spying on and collecting much of that same data from RG’s users, can RG offer much that the big players in the market can’t?
Another confounding factor — Article 13 of the new European Copyright Directive puts much more liability on platforms like RG for the illegal content they host. If it passes as written, this is going to put a major crimp in their plans and potentially require them to pre-emptively filter content as it is uploaded.
Which leaves me wondering where this is all going. Is RG going to be worth the price that it would likely take to purchase it (given over $100M in investment). Will the lawsuits and Article 13 drop RG’s value enough to make it a reasonable acquisition target? And once you owned it, what would you do with it? Also, selling yourself off to a commercial publisher seems like a pretty steep drop-off from original plans to win a Nobel Prize (https://heureka-conference.com/researchgate/).
David, the endgame of syndication is that library licenses to primary publishers remain in place for subscription content while user access takes place (based on entitlements) at third party sites like ResearchGate. As I wrote in my piece earlier this week, the idea is to separate user access from the business arrangements while respecting licensed entitlements.
So just to say it a little more directly, the idea is NOT necessarily to put an end to subscriptions and flip the whole model to advertising. That may be how RG would support its own distribution/hosting costs, but not how the whole publishing system that lies behind it would be supported.
My question then is how the third party site gains access to the subscription content and the access control information. What would be in it for a publisher to grant those permissions? Would it just be a matter of saving the costs of running your own platform? Or would the syndicator be expected to pay a license fee or share some percentage of revenues?
Interesting also to think about this in light of RA21.
Maybe it’s Friday, but this seems like an overcooked and breathless response to a ho-hum story. Publishers have been syndicating content to what you call “content supercontinents” for decades, and risking all sorts of potential downsides in exchange for potential new revenues and reach. Ovid, EBSCO, and many others are “content supercontinents” that don’t freak publishers out (too much), and are integrated into the publishing economics of most publishers. Back in the late-1990s, Ovid especially freaked publishers out, but analysis showed the effect of syndication was generally neutral or positive when revenues and reach were netted out. Chances are this trial will generate a similar cost:benefit analysis.
ResearchGate being treated as a potential syndicator via this trial is interesting to me because it appears to put them in a box that SpringerNature (and other publishers) might like in the future — that is, another potential source of revenue in exchange for content. The test will reveal whether there’s something there for both parties to formalize, but this seems like a nice way to declaw ResearchGate and bring them into a more responsible and familiar mode via a potential financial and syndication deal.
It’s a syndication test. Welcome to Snoozeville.
I think you’re confusing syndication and aggregation. This is fundamentally different than EBSCO et al aggregating content and paying publishers to do so.
Kent. Please read my post from the other day for a distinction drawn between aggregation and syndication. Once you understand the different models let’s return to the discussion.
Would you define the key difference as aggregation being secondary availability of materials (in addition to the publisher hosting the primary distribution point) and syndication as the third parties replacing the publisher as the primary distribution point?
Aggregation: Content hub pays publisher for a license to reuse content and makes its aggregation available to readers based on its own sales to libraries.
Syndication: Content hub respects publishers’ library licenses and only providers access to users who are entitled to access based on those licenses. Money may not change hands. Usage data from the comment hub is added to the publisher’s Counter usage reporting to libraries, authors, etc.
I’ve never heard that definition of “syndication.” You think that comic strips are “syndicated” to newspapers in this way? I think you’ve made up a definition out of thin air.
Aggregation is the cumulative result of syndication in my book. Publishers syndicate their content to aggregators. Aggregators aggregate syndicated content. I think that’s a pretty normal view of how this part of the business works.
Yes, Kent, I named a strategy that is developing in our community. You should be crediting me for that rather than accusing me of making something up out of thin air.
What you’re talking about I’d call more precisely “authorization syndication,” which differs from traditional content syndication because what’s being syndicated is the authorization to view the content. To just call it “syndication” is wrong. So, I’ll make up something out of thin air, and that’s what I think you’re talking about with more precise language. Make sense?
You can call it whatever you like. I am not going to engage in a semantic argument with you. If you’d like to speak about the substance of the strategy here, now that you seem to understand it, I’m all ears.
I still don’t understand it. The more I think about the speculation here, the less feasible it all seems. I think this is much simpler than it’s being made out here. I think it’s a syndication trial, and in the normal, traditional sense. I don’t think there’s much more here, and it doesn’t seem exciting. The more I think about it, the less I see here. Have a good weekend!
Roger’s writings on how things might be unfolding are what led me to recognize immediately this seemingly small pilot as much more consequential than it might appear at first glance.
It is a path forward. It fits with what a lot of other vanguard thinkers about the future of the Internet are thinking. I’ll write about that all separately.
For everyone who appears to be confused about this emerging syndication model, I created a brief explainer: https://sr.ithaka.org/blog/what-is-content-syndication/
One small comment: there is a statement to the effect that SpringerNature might buy ResearchGate. I have no idea about this, but it seems more plausible that the acquisition would go in the opposite direction, assuming RG escapes its legal issues. I am not defending valuations or valuation methods, but investors are likely to pay a higher premium for ResearchGate’s growth prospects than for SN’s cash flow (enormous though it is).
I’ll go with the easier theory (Occam’s Razor comes to mind) that Springer has just lost their mind. I don’t buy the strategic thinking, nor the playing the long game, theories. This does legitimize non payment for articles that others are paying for. Losing their mind could be useful for scholars in the short term. The rest of us need to think about long-term preservation of the scholarly record.
Scott, could you say a bit more about how this threatens the long-term preservation of the scholarly record in your view? The copies on ResearchGate aren’t the only copies and I don’t see any effect on the preservation commitments Springer Nature has for its content.
A point worrying to me and that I don’t see much focus being made here is that this pilot is preventing the researcher experience from happening at the publisher site but moving it away to RG.
If Springer abstracts would be made available at RG profiles, readers could scan and then link to journal/article at Springer, I would fully get it as a win win collaboration.
This pilot of disrupting your reader experience by placing one of your most important high quality assets in someone else’s platform (which happens to be an emerging competitor who nobody fully trusts and some other publishers are sueing) looks demential, naive or extremely strategic to me.
I beg everybody to find a different term from “syndication” to describe the Springer-ResearchGate arrangement. “Syndication” is a term of art in the media world, including publishing. This is not syndication and it confuses people, myself included, when that term is used. RG may wish to call it “syndication” as it makes their presumptive model a handsome arrangement for publishers, but syndication without the historical exchange of funds, and a wide network of syndicatees, is not what this is about. This comment is purely about the term, not the underlying arrangement.
Just to be clear, ResearchGate did not call this syndication. Roger called it syndication when he observed signs the model could be emerging — posted last October: https://scholarlykitchen.sspnet.org/2018/10/15/syndicate-content/ — and I adopted Roger’s term when I identified this Springer Nature pilot as an example of what he was predicting.
With that aside, I’ve been pondering why the naming has been such a point of contention in the comments on my post and if the issue is that multiple communities of practice use the same term in different ways. Wikipedia is one of my go-to sources for documentation of different uses and I discovered that there is a disambiguation page for syndication (https://en.wikipedia.org/wiki/Syndication).
Indeed the term syndication has been used in broadcast and print media as some have argued its meaning here. But, in addition to this use in the publishing industry, it is also used in the Internet content industry other ways. It seems we’ve come upon another one of those terms (open access, anyone?) that means what it means — until it doesn’t.
I think that is accurate. My point is, Why make life more difficult? You wouldn’t expect someone to call Atypon an ILS; it’s simply confusing. Why not start anew? Some years ago I had a hobby of Googling for “open access.” Most of the links were for signing up for health care.
Twp answers and an observation plus a P.S.:
1) This term is already in use in the Internet content industry. Roger is observing, in my mind, a specific case of a more general phenomenon.
2) I’m not in the practice of citing other people’s analysis but then giving it a new name. That might lead people to think that I did the initial analysis and I wouldn’t want to erase a colleague that way.
3) Finally, I am really quite puzzled why there was not a single objection to this terminology when Roger first offered it last October or discussed it in detail in his post that was, quite literally, just two days before this one. The time to suggest changing the term might have been last October or earlier in the week – but that didn’t happen.
P.S. While I’m writing … I didn’t make up the term “supercontinent” either though it seems that also may be being credited to me in the comments – to ensure that is clear, that term is also from Roger.
Do you think there’s any significance in that none of the journals included in this trial offer an open access option for authors?
Since this is a pilot to see what happens, I think it leaves the assessment design a little tidier to not have a confounding variable of individual articles that may be Gold open whereas others that are not?
Yes, perfectly sensible. Keep it simple, keep it clean. Get results, analyze, iterate.