There was big news a few weeks back in the Humanities and Social Science community when it was announced that Elsevier purchased SSRN — a popular preprint server for this area of research. There was much wringing of hands by some users of SSRN, despite Elsevier’s assurances that the site would remain largely as it exists. All SSRN staff will be maintained and it will still be free to post and free to read.
SSRN has nearly 640,000 articles and last month promoted over 100 million downloads. This is not small potatoes and much has been written about the strategic advantage that owning a large preprint server gives to any commercial publisher.
It’s important to note that the objection to Elsevier’s acquisition of SSRN is exactly that Elsevier is the purchaser. SSRN was a for-profit entity before Elsevier entered the picture.
Adding more fuel to the fire of the former SSRN supporters, some users have found that their posted PDFs were removed recently due to copyright concerns. It appears that some of these removals may have been in error but just as many that have hit the public eye seem justified.
Scholarly Kitchen readers know this but I’ll state it again. Authors may not legally post versions of content that they do not have the right to post. Each publisher can/will have different policies around what can be shared on an unrestricted website. Generally speaking, authors are free to do whatever they want with the author’s original version (preprint version) of their manuscript, and permitted to share the accepted manuscript version of their published papers after an embargo and with some restrictions on commercial use. Unless the paper is under a license that allows the final published version of record to be posted online, this is typically not allowed without a licensing arrangement, such as paying for open access, or the host site signing a deal with the paper’s publisher.
So back to SSRN — much like when Elsevier purchased Mendeley, a clean-up of materials posted in violation of copyright is going to happen. Because Elsevier has their own set of article sharing policies which they expect others to follow, so too they make all efforts to follow the policies of other publishers.
Some folks are complaining that this is a change to the SSRN policies. In the May 2016 announcement of the sale of SSRN to Elsevier, SSRN Chairman Michael C. Jensen wrote, “and our copyright policies are not in conflict — our policy has always been to host only papers that do not infringe on copyrights.”
Copyright and Trade Marks: Any person who provides any material for inclusion on SSRN is responsible for ensuring that this material does not violate other parties’ copyright or other proprietary rights and does not otherwise violate law or applicable SSRN policy. As part of its general right to remove material, SSEP reserves the right in its sole discretion to delete or make inaccessible files that do or may contain material that violates law, or applicable SSRN policy, or the rights of third parties.
What seems to have changed is not the policy but the desire to enforce the policy. Safe Harbor laws allow websites to claim ignorance when copyrighted material is posted on their pages by random users. However, once the legal copyright holder notifies the site that materials under their copyright are posted, the site is required to remove it in a timely manner.
According to SSRN on the post linked above, they are taking steps to review papers at submission for any potential copyright violation. Users of SSRN on the same discussion board would rather SSRN allow these materials to be posted and wait for the copyright owner to send a take-down notice.
In the wake of these pseudo-controversies, we have SocArXiv. While it has been stated that SocArXiv was in the works prior to the announcement of the SSRN sale to Elsevier, their timing could not be better. They are hoping to capitalize on the growing discontent with SSRN.
SocArXiv was described by founder Philip Cohen of the University of Maryland as such:
… there remains a need for a new general, open-access, open-source, paper server for the social sciences, one that encourages linking and sharing data and code, that serves its research to an open metadata system, and that provides the foundation for a post-publication review system. Once it’s built, anyone will be able to use it to organize their own peer-review community, to select and publish papers (though not exclusively), to review and comment on each other’s work — and to discover, cite, value, and share research unimpeded.
SocArXiv is built on the Open Science Framework (OSF) platform, though the current site is labeled as “temporary.” The University of Maryland, where Cohen is employed, is called a “host.” In this interview with Richard Poynder, Cohen called it a “Program of the University of Maryland.” On the University of Maryland site, you can make a donation to support it. In the future, Cohen explained, the hope is that universities and research agencies will fund the project along with the grants and funding that OSF gets through the Center for Open Science. The cost for maintaining the site will not be insignificant. Annual operating expenses for arXiv, albeit a much larger site serving primarily physics, are reportedly over $820,000 and a new fundraising campaign is underway to grow capital for improvements. Funding for physics has been higher than for social sciences so it’s not clear that grants will be as plentiful for funding SocArXiv.
Despite the descriptions, there was some confusion that arose last week when it was pointed out to SocArXiv that they were hosting the final PDF of a paper under copyright with Taylor & Francis. A discussion with Rebecca Kennison (a member of the SocArXiv steering committee) the SocArXive official Twitter handle, myself, and others ensued and things got a little murky. Here is what was said:
On the subject of whether anyone at SocArXiv is checking material before it is posted, the answer is no — not at this time. Okay, but the ingestion process does seem to indicate that there are human beings involved. I submitted a paper, which is done entirely by email. The requirements for the email are to include the title of the paper in the subject line and then the abstract in the body with the paper attached. In my profile, my affiliation information was parsed from the email signature line and the paper. An affiliation is not a required element of the submission email.
It is possible that there is an automated ingestion process pulling these things into the system but that seems unlikely given that the affiliation was not a requirement in the email and that is the one that was used to create my account.
As stated earlier, SocArXiv is running off the OSF Preprints platform, which is listed as “coming soon” on the OSF site. The not-yet-available nature of the platform means that there isn’t a lot of information about how it works.
If there is a person actually processing papers, one would think that a check on whether the user has the authority to post the attachments sent would be part of the check, thus avoiding the final published PDFs of articles.
As of today, the T&F paper is still posted. Kennison stated that SocArXiv “will respect all legitimate take-down notices” but went on to say that in her personal opinion publishers have to “prove they have copyright, not just claim they do.” By law, the requirements for a “legitimate take-down notice” do not require such proof.
This conversation evolved to another question. If SocArXiv is a preprint server, why would there be published papers? This was the response:
So despite the fact that it’s called a preprint server here, here, and six times on the submission page, it’s not a preprint server. It’s an archive that will host anything a user wants to post and eventually, you can add comments (a.k.a. post-publication peer review).
The other touted benefit is that users receive a persistent ID for citations. The persistent ID is a URL. Here is mine: https://osf.io/d2kz9/. Given that every description of the site says that the OSF Preprints is a “temporary” home, where will it go and what will happen to these persistent identifiers? One of the main reasons we use DOIs as persistent identifiers is because of the transitory nature of URLs.
I have to say, this feels like a hasty launch. Cohen said that the plans were in the works well before SSRN’s sale to Elsevier was announced and I have no doubt that this is true. We don’t know what the status of the plan was at the time of the announcement. But the entire thing does not seem ready for prime time. If OSF’s platform was not ready, which seems to be the case with the entire OSF Preprint platform labeled as “coming soon”, then what was the rush?
I believe that the answer is Elsevier. The steering committee for SocArXiv have some really grand plans and the use of the OSF platform creates an end-to-end solution — maybe. Having one platform for project plans, data, figures, presentations, preprints, and (ambitiously) post-publication reviews that are accepted by the community in lieu of journal articles could be a game changer in the future. But me “getting” this, required two days of research to figure out why things didn’t seem to be what they were called.
The confused launch and blurry vision may slow the ambitions of SocArXiv. There are other alternatives for those no longer enamored by SSRN (despite its new infusion of cash that will sustain and likely add functionality to the site). In the meantime, as with any new endeavor, I would suggest that authors proceed with caution and keep back-up copies. This project will be one to watch.