Open access research: an early career researcher perspective
All the researchers disseminate their research through different media/documents, e.g. research papers, seminar/webinar in conferences/individual, patents, monographs. Fellow peers (academicians, research students/associates) read and work on these to have a better understanding to advance the research in a better way. For doing this, one need to have an access to these published pieces of work. The accessibility depends on the access status, open or need to be paid, of the particular document and the location (with/without subscription) from which it is being accessed. Open access (OA) document generally means that anyone can go retrieve it freely (without any monetary, technical or legal obstacle) otherwise one need to pay. Subscribed location means the server you are using to get the documents has subscription, either paid or through some contracts, for the documents.
In short, the accessibility to a published research document varies with many factors. Hence, this piece of discussion revolves around my experiences in two different continents (recent PhD from India and two years of PostDoc in UK) with different modes of research dissemination and access for the published research documents. Both the universities I been to, are one of the top research institutions in the respective countries (as per Times Higher Education) but have different approach towards OA research. Now, we will go through different ways in which an early career researcher (ECR) like me tries to access the published documents, the difficulties faced during the access, benefits with OA process and the mitigating solutions for OA research in the future.
Access through university/institute subscription?
Most of the researchers are affiliated with a university or an independent research lab/institute which have the paid subscription to different publishers e.g. Elsevier, Springer Nature, Wiley, IEEE, IOP publishing, Taylor and Francis; for an access (limited) to the published documents. Individual institutes (universities or labs) agree on a contract with the publishers with different terms like mode which is mostly electronic files in today’s digital age, duration i.e. time limit, limitation (journal volume, quality) of resources for access. These cost the institutes a whooping amount of money, e.g. universities in UK were paying about £192 million per year to publishers in 2008. If an inflation is considered, the financial burden gets increasing with time and number of journal titles/documents. If it costs too much, is it really necessary? It is the very first thing, a researcher will look for literature survey before starting any experiments. In short, for smooth access to the published documents to its researchers, an institute can’t avoid this cost.
Individual access — legitimate and other methods
What about the researchers who are not affiliated with the institutes having a subscription or individuals with independent laboratories? How they access the documents? People will first go to check if the required document is already OA (also depends Gold or Green OA), then no need to worry for the access. If not OA, it will cost one a few tens to hundreds of pounds to buy the document. The price also varies the way you want to have an access — only read/rental or fully download and print. Researchers with not much funds available or who does not want to spend money, try different things to have an access. First thing, one do is to search in Google Scholar if there is a free PDF available.
If not found on Google scholar, one checks on ResearchGate. It has provisions of sharing the full text file publicly or privately person to person. Many researchers having account on ResearchGate shares a private copy if someone requests for it. If no author of the document has a ResearchGate account, person email is sent with a share request of the article. It is quite fast for active ECRs to get response from but a long wait from busy professors. This time consuming task leads the researchers to the alternate methods of getting access to non-OA documents like use of ‘Sci-Hub’ for articles and ‘Library Genesis’ for books, an illegal method of file sharing.
Through informal chat with many researchers across different institutes, I found that people tended to use Sci-Hub very often before the provision of VPN or proxy networks from the affiliated institutes with subscription. The popularity of Sci-Hub in India can be understood from the recent lawsuit in an Indian court by Elsevier, Wiley and the American Chemical Society. The high use also comes due to ease of access where one can get access to all kind of documents just by putting title or digital object identifier (DOI) without need of any proxy servers. This also comes due to lack of training among the new research students and ECRs, regarding the status of article and which version (accepted or final printed) can be shared to avoid the copyright infringement at many instances.
Now comes the big question — what are the different ways through which the documents can be made OA? Laakso and co-workers talked about the conversion of subscription based journals to OA ones discussed a number of methods to it. There is also hybrid mode publishing offered by many subscription journals where authors choose to publish the work as subscription based or OA based with payment of article processing/publication charges (APCs). In last two decades, the number of OA publishers has increased where all the documents getting published are OA with an associated APCs. Non peer-reviewed preprint platform is another way for OA research dissemination. In the next part of discussion, we will see how these different approaches are beneficial and associated drawbacks if any.
OA publishing and recent moves from publishers
The most suitable way for free access to documents is to go for OA publishing mode where all the publications are OA from the beginning. The publication platform does not matter in this mode as you need to publish OA. For UK, most of the funders like EPSRC make it mandatory for OA publications and provides funds for APCs. Projects funded in Europe by European Commission in Horizon 2020 makes it mandatory for all research output to be OA. Going forward, European Commission also launched an OA publishing service — ‘Open Research Europe’ for grant recipients of Horizon 2020. On similar lines, other publishers launched numerous OA journals, e.g. different new from IEEE. Top 5 publishers handles more than 50% of the published papers in 2013 establishing an oligopoly. To encounter these, in last decade, other small publishing house came into market with new ones mostly based on OA journals e.g. MPDI, PLOS etc. Recently, IET (The Institution of Engineering and Technology) from UK announced new agreement with Wiley with transition of all hybrid subscription to gold OA journals.
OA — a burden for authors!
Then what are the obstacles in doing this? The first and biggest one is payment of APCs which is around 1–2 thousands of pounds per article which is generally supported by funders in developed countries but difficult for researchers in developing countries. As discussed above for UK and Europe, different funding bodies advocates for OA research outputs and support accordingly. Plan S was also launched advocating immediate OA. Even many universities like Manchester support OA publications but on first come first serve basis due to limited resources bringing all the burden to authors. But the grass is not green everywhere. In India, the biggest funder, DST or a university (except few exceptions) does not provide any funding for OA publications and authors are left with very few options to get support money for OA dissemination. This is helped with APCs waiver or discount schemes from the publishers but is not enough in all the cases. Though the OA is helpful to decrease the steeply incrementing subscription fee, it is also stopping researchers from low economy countries to publish at good venues.
OA agreements between publishers and institutes
To reduce the cost involved in accessing or publishing the OA documents, institutes are going for individual contracts with publishers for wide access in a group or consortium. In 2004, Elsevier and the University Grants Commission of India (UGC) agreed on e-access to documents on ScienceDirect to India’s top 50 universities. This was a great move to cut the cost with availability to a wide pool of the researchers. But question of free access remained unanswered for the low ranked institutes! Maybe the new ‘one nation, one subscription’ journal access plan an ultimate solution. Elsevier has OA agreements with few institutes across the world. Nature journals announced first OA agreement with Germany but the cost ($11,200 per article) is too exorbitant to avail except few institutions in the world. Recently, IRel consortium of Ireland signed an agreement with IOP Publishing for OA paper publications. These agreements save the institutes’ libraries a considerable amount of money to be spent on subscription and publication charges. Universities are also going to ‘unsub’ to cut the subscription fee through calculation of overlap between subscribed and OA published documents.
OA — preprint platforms?
To cut the fees associated with access and publication of articles, different preprint platforms are being used. Preprints are generally a full draft manuscript posted without any peer-review. These offer a rapid publication of the document with citable DOIs attracting many researchers to showcase a quick development in the research area without any financial burden, long awaiting peer-reviews and time involved in publishing with a journal. It became very important platform for all the research developments regarding COVID19 with 30,000 related articles posted on different preprint servers within 10 months of first confirmed case. It also helps to get the credit and establish the priority to the authors for the work carried out. It is very useful for visibility of the work supplemented with a possibility of wide peer-review.
Hurdles in OA — trustworthy peer review and published results
Now, let’s see the different obstacles in achieving OA research. First and foremost comes the quality of the research work presented through different documents. Any peer-reviewed published document goes to at least two experts of the corresponding research field for their comments after the submission. This also applies to drafts submitted to any preprint servers, these are subjected to peer-review otherwise it is dangerous to science. The reviewers can be anyone working in the field and the most opted review mode is single blind where the authors do not know the identity of the reviewers but the reviewers know about the authors. It offers a good perspective for judging the authors’ work with any prior document from the same research group. But it also leads to biasing towards well-known authors.
To avoid the favoritism, many journals use double-blind review process where nobody knows others’ identity but the reviewers tend to find the authors through the papers cited in the draft. Going further, open peer review was introduced where everyone knows each other’s identity but it is difficult to find the reviewers. How these are affecting OA publications? Most of the OA publishers as mentioned before advocates for accelerated publications within 8 weeks of submission, due to APCs involved and good for business, hence requiring a rapid review. In general, at least 4 weeks are given to the reviewers to submit the report. The reviewers are chosen by the Editor-in-chief or associate editors based on their expertise. This involves at least 1–2 week for editor to choose and send the invitation to review. To override this time lapse, the journals are using artificial intelligence (AI) to choose the reviewers which is based on the profiles at different platforms like GoogleScholar, personal homepages, Publons, ResearchGate, ORCID etc. and associated keyword. This was also confirmed by an editor of OA journal I volunteered to as a reviewer. The choice of reviewers with AI is also questionable.
Then comes the reduction in time given to reviewers for the comments to bring down the overall time in publication. I received review invitation to give comments in as less as 1 week of time which is obviously not enough with other commitments. From feedback of few colleagues, these 1 week requests mostly involve the OA journals and they don’t tend to entertain. So, journals are losing pool of some good experts due to cutting short of time duration. The peer-review gets completed with people having less expertise and the manuscripts passes to get published. This questions the ethics of the peer-review as well as the reliability of published data and results. It all left out to author’s ethics regarding correctness of data and the conclusions obtained.
To receive quality and timely comments from reviewers, journals started to reward the reviewers with discount coupons for book purchase or limited-time free access to published journals. There is a discussion on monetary payment to the reviewers like ‘The 450 Movement’ where $450 can be paid for each review carried out. European Open Science Publishing offers a paid peer-review model (7 Euro per approved review) for quality and timely comments from the reviewers. Is only cash payment a way to remunerate? Publons is another way to show the peer-review work done by researchers as it is quite important for ECRs like me to make it visible to have a proof of the academic activities. Global State of Peer Review report of 2018 from Publons took survey responses of ~12,000 researchers with findings of about 17% seem to be interested in peer-review in lieu of monetary benefits and about half wanted some sort of recognition from the research community and their employers.
Quality OA research — some solutions
To ensure stringent OA research outputs towards sustainable research community, it is very much necessary to get quality and timely review reports. This can be achieved by giving recognition to the review work carried out by the researchers by the affiliated institutes. It should be included in the promotion criteria for the ECRs along with document publications. Currently, there is no evident benefit from reviewing the manuscripts except some advancement in the knowledge. The time invested by ECRs in reviewing need to rewarded through proper channels to keep them motivated towards supporting OA research continuously.
The publishers or editors especially with OA platforms should think of a trade-off between the time given to reviewers and quality/timely reports. There are other commitments involved which should be taken care off along with reviews. Researchers tend to decline the review request citing ‘too busy’, but what will happen if nobody has time. Will the peer-review system be abolished? Journals may make a rule of reviewing 3 journals for authors from an accepted paper to keep a good pool of reviewers.
Not only good reviews are necessary, publishers also need to pay the cost of admin and official work. How to decrease the burden of OA APCs from the authors with a check on the expanses? The publishers may think of reduced APCs and funding bodies or the researchers’ institutes may put more money for the OA APCs. The authors may think of adding a section in the grant writing for OA research cost along with other costing of the project. These all relates to the money needed for payment of APCs. The government or different consortium may think of agreeing on some individual bulk payment to the publisher for all OA research going forward.
For free access to all the published documents for everyone, the respective government can go for ‘one nation, one subscription’ model where a nominated agency can go for an agreement having stipulated terms (finance, legal and technical). This can also be done on model based on free access to COVID19 journals and book content provided by different publishers.
Finally some takeaways……….
· Researchers need access to published documents to advance the research
· In absence of institute subscription, researchers try to get access through different means even some illegal ones
· There is a need of training for researchers towards sharing the correct version of document
· Authors should be least worried about APCs for quality OA and need to be ensured by funders or institutes
· Quality and timely review reports are needed for reliable OA research outputs and can be ensured by appropriate recognition to review process
· Institutes, funder bodies and government may take a combined effort with publishers to have free access to everyone
