Author: Renee Hoch, Managing Editor, PLOS Publication Ethics Team Journals’ and publishers’ authorship policies set expectations with regard to credit and responsibilities…
Imagining a Transformed Scientific Publication Landscape
Open Science is not a finish line, but rather a means to an end. An underlying goal behind the movement towards Open Science is to conduct and publish more reliable and thoroughly reported research. Increasing the transparency, reusability and connectivity of scientific outputs is a common desire shared among publishers and researchers, but progress can seem slow and implementation far from widespread.
Akin to how scientific understanding is often achieved through incremental progress, system-wide changes toward Open Science will only be achieved through earnest collaboration among funders, institutions, publishers and researchers. Looking at both pragmatic solutions and the underlying ideals, we imagine changes to scientific publishing in the context of four fundamental functions that publishers should provide: dissemination, verification, recognition and community building.
How can the future provide more usable, accessible and relevant dissemination?
The term “research article” is largely used as a catch-all for published discoveries. A research article is typically consumed as a static PDF that’s sent to publishers after results have been accrued and conclusions drawn. Readers must wait until the study is complete and the associated paper published before they can access the methodology, which may have been finalized months or years prior.
Breaking down the research article into more digestible and structured parts—or modules—can provide researchers with more opportunities to reuse and improve research. Incomplete reporting of methods in traditional articles can prevent research from being reused or reproduced. More opportunities to publish detailed and enhanced methods are the two article types coming soon to PLOS ONE: Lab Protocols and Study Protocols.
Lab Protocols, developed in partnership with protocols.io, share the step-by-step instructions for verified methodologies and computational techniques. Utilizing the specialised features of protocols.io’s platform in conjunction with a peer-reviewed article on PLOS ONE that contextualizes the presented methodology, Lab Protocols will provide readers with the information needed to replicate a validated method in a precise and accessible format. Study Protocols are open for planned research studies, giving readers an opportunity to see design and analysis plans and helping reduce waste and bias in research.
In the future, studies should be presented as a collection of interlinked research objects and associated metadata. Data, code, protocols, and other elements such as reagents can all be independently identified and verified, allowing for readers to more easily find and access the elements most relevant to them.
How can widespread adoption of Open Science provide more reliable and reproducible research?
The ultimate test for the reporting and validity of a study is whether it can be fully reproduced by different researchers. The materials used, the steps of the methods and the experimental conditions must be precisely described for a study to be reproducible. Reproduction of previous research can be a jumping-off point for contemporary researchers that leads to new discoveries. In short, proper reporting helps advance the field and accelerate further discoveries.
On platforms such as protocols.io, researchers can verify if published protocols work for them. Alongside participating researchers, publishers are important in the verification of publications by organising peer review, which can be customised for different research outputs. For Lab Protocols, PLOS will be organizing peer review of submitted manuscripts and content hosted on protocols.io: a small step toward extending the value publishers can offer in verifying the validity of research and improving trust in new research formats.
Transparency enables verification. Preregistration enables interrogation of what was planned and what was carried out, and publishing peer reviews and preprints allow readers to view changes that occur during peer review. Direct links to deposited data and executable code gives an opportunity to work with the same material as the original authors.
Having these elements—protocols, data, code, review history—available and linked to research articles reduces dependencies on the original authors of a study, promoting efficiency. A study that can be replicated with available and easy-to-find information allows the research to be usable far in the future.
How can reward and incentive systems be improved to provide researchers with credit for more of their research?
The perceived success of a researcher has traditionally been dependent on the credit they receive through publications. Career advancement and other rewards can be determined by how research outputs are valued and recognized, and thus recognizing diverse and more granular research contributions beyond publications is essential for researchers to be rewarded for their work appropriately.
The CRediT—required by all PLOS authors and publications—offers a more nuanced view of authorship and contributorship. By providing new opportunities for peer-reviewed publications that reward critical but sometimes overlooked contributions, such as a methods development, publishers can provide opportunities for diverse elements of the research process to be better recognized.
Peer-reviewed protocols give authors a formal publication that enables their methods to be cited, shared and used as a signal of accomplishment. Similar opportunities exist for sharing, describing and citing research datasets and software, which need to be more widely adopted by the wider scientific community, furthering chances for additional recognition.
Publishers are just part of the solution. Institutions and funders can work with publishers to recognise and reward researchers for working transparently and reproducibly. Researchers can benefit personally from presenting their research in open, reusable formats as this can drive collaboration and reuse—and can consider Open Science principles when assessing studies by their peers.
How can Open Science provide more opportunities to collaborate and connect?
The scientific community has made great progress in making Coronavirus research as Open as possible. With authors across disciplines depositing their articles as preprints and publishers making relevant research fully accessible, the benefits of Open Science have been made even clearer. By connecting researchers through peer review and providing structures to make relevant research shareable and accessible without barriers, journals help researchers build and maintain their communities. The role journals have in supporting these communities will continue to evolve as innovative new ways for researchers to collaborate and connect develop.
If these sweeping changes have helped improve scientific discovery in a period of global need, they could be adopted in other areas of research. Comprehensive commitment to Open Science requires the full scientific community, as publishers alone cannot persuade such practices to be adopted. For research to be more efficiently disseminated, verified and credited, system-wide changes toward Open Access must be embraced across the scientific community.
Sharing protocols and research design ahead of executing the research is a recipe for disaster. Other people will steal the ideas. Yes, publications matter in career advancement, but being the FIRST to publish something matters a lot in this competitive environment. Much effort, time and organization–and preliminary research— can go in to developing a study protocol, so why hand it over to others for free? They’ll take your work and your ideas, and carry out the work with this jump-start done by you for them. They’ll beat you to the patent office, so to speak, figuratively or literally. Researchers already know how to design studies without sharing the design with everyone else; or they can consult with colleagues in-house. Bad protocol development surely happens occasionally, but there are always complexities in studies that can only be seen in retrospect. Peer reviewing of protocols probably won’t do any better, and presents serious risks . In my experience submitting manuscripts, the peer reviewers are often way off the mark anyway. The inability to thoroughly understand or replicate some work is often the fault of the journals which don’t insist on clarity of the process being described, or allow poorly explained statistics, or allow authors’ failures to address bias or limitations of their work or why they didn’t do something a certain way. Those are the common problems with manuscripts which I peer-review. And god knows we don’t need yet more unpaid peer review chores to do.
I agree all and it must be open access
Thanks you and best regards
Please remember that it is almost impossible to duplicate ecological studies which are costly to perform and occur at one point in time and space. One can seek to verify results through careful statistical analyses and modeling, but similar trends/patterns are not the ‘ultimate test’ you advocate above.
In addition in the laboratory, verification by duplication really leaves the verifying laboratory with nothing new to publish. Will PLoS publish these duplicative verifications?
Thank you so much for your wonderful efforts in serving science and humanity
In many country especially developing, many researches suffered from massive reduction in facilities and reagents, for example I try to measure mycotoxins in feeds, but I lost time searching for old techniques that my suited to my country capability,
If there will real efforts to arranging and sorting all the protocols from old to new according to subject e.i mycotoxin measurments, from old techniques such as uv vis and TLC gradually to the GS/MS/MS and so on for the different
I think this my give the researchers from different level an opportunity for academic continuation and for development of new and easy protocols
While I believe science should be accessible to a greater audience, fee for publishing disadvantages small labs and PUIs where the focus is on undergrads rather than the endless chase for grants. It gives readers greater access, but restricts publication opportunities to larger and better funded labs. Until we can fix both problems, I find it unfair to criticize those authors that cannot afford open access options.
It has become apparent in the era of Merchants of Doubt, that inconvenient truths often lose to well-funded science denial. And implementation of results (e.g. climate science) depends on public opinion, winning battles for truth versus propaganda. So I think you should explore adding a step to your vision. What Bill Schlesinger calls “translational” science. Repackaging results in formats accessible to the public. More scope for science journalism. More outlets that reach wider publics. More integration with education systems. More funding for truth.
I congratulate on your altruist endeavors. As an applied linguist, I specialize in researching scientific terminology from a lexicographic point of view. Many of your research papers have served as a basis for analysis through out my work and I find them quite reliable. I am most interested in their proper translation into Mexican Spanish, particularly the abstracts and keywords. There are a lot of complications when it comes to consulting or retrieving them in Spanish on line. Keep the good work on going.
Many of your suggestions do not seem feasible to me.
I’m not sure what the intended goal of this post is. There have been many such efforts in the past, and the biggest issue is always one of diminishing returns. It is far more useful from a research standpoint to move to a new topic or the extension of published work, rather than put the extra effort into tracking provenance and protocols. It is unclear as the weather “open” COVID papers are of any use whatsoever. Perhaps it makes the public feel better, but the research community have always had access to the avenues of publications, unfortunately, those who do not, are unlikely to have valid inputs into experimental and theoretical studies.
A major point which is not adequately covered, is how the incentivisation scheme fits into the existing systems dominated by point metrics like h and i indices.
1. Modern science arose 500-odd years ago primarily from invention of mechanical printing. The technology permitted modern science, but its physical limitations provided constraints to thought and practice. The new online technologies provide opportunities to expand that thought and practice. Your initiative is an excellent exploration of the possibilities by removing some of the hidden constraints.
2. A limitation of what you are proposing is that there is no explicit addressing of one of the large constraints in the current way that medical science (at least) is conducted. There is no direct and continuous connection between the choices made by researchers about what problems are addresses by the research effort, the resources expended by the community on such research, and the value placed by the community of solving the research problems as receivers of the research output. How do we close the loop of ideas, uses and usefulness?
3. Stated another way, when a doctor and patient discuss the use of a treatment, there is currently no guarantee that all information required to make a fully informed choice is freely available and accessible at that decision-making point. Your initiative would form part of that process, but it is not sufficient in itself for adequate informed consent to occur.
4. What is required is that no secret/confidential/inaccessible information is permitted to be promoted for use at that decision-making point.
5. This ethical consideration, at an individual and community level, might be seen as a useful road-map to achieving this aim. It would help the process of plotting out what additional components of the system you are starting by your initiative, and to identify what has not yet been thought of.
This is a new way of thinking to wards the future of communicating science globally. My fear is the issue of ownership for specific discoveries. What about it?
That is all we need, the academy’s response speed – can science be changed? Yes, the time has come for us to be agile in many actions. In Brazil, with the loss of space and investment, the path is very narrow, and at that moment, joining efforts was the only way out for my Research Group. Today we are 11 months with a product at the beginning of a clinical trial, despite all the setbacks that we still live in some institutions.
The cost of open access is too high for most of us in developing countries. Grants are few and far between and the cost is high even when research is financially supported in local currency. As it is, even good research from developing countries is viewed with suspicion, the cost of open access has put them at a further disadvantage.
COMMUNITY BUILDING: I think this is the most critical aspect of the ‘mega’ journals. But, the proposal doesn’t address an essential limitation of ‘mega’ journals the lack of field-specific communities that the well-established journals with a good tradition provide. The community aspect in terms of the pool of reviewers as well as of readers is a major advantage for the quality of feedback during reviews that can improve manuscripts and is highly relevant for recognition within the field.
I’d suggest establishing or experiment with field-specific sections under the PLoS One umbrella, e.g., PLoS One Respiratory Medicine and Imaging with a section editor or associate editor(s) that are established in the field and can build a community. Among the promoted collections as an alternative to specialized journals, I have not found an example that is running over a long time and represents a community of returning authors and researchers — it seems that unstructured individual initiatives without mechanisms for recognition and other incentives do not result in self-sustaining contributions. I think it is essential for PLoS One to re-establish communities and identities around specialized sections or additional PLoS journals. Otherwise, I think that the fast-growing number of open access journals linked to established communities and high-impact journals will attract a growing numbers of potential PLoS authors.
I think a necessary aspect of this intensive digital collaboration would be a development of a joint research strategy within each discipline: which questions are important for that discipline, what tasks should be done in the nearest 3/5/10 years. And this strategy should be understandable and available for everyone in the community. Otherwise researchers are just not encouraged to reveral their plans and ideas which simply can be stolen by others – and this blocks collaboration. I discuss this in more details here: https://www.researchgate.net/post/Science_20_Is_global_research_strategy_the_missing_piece_we_need?_ec=topicPostOverviewAuthoredQuestions&_sg=5P2YC2M61Hraczhxzqs0x3ElT0s7HEQsP8sAPdnCGX4fEFoge7mlQfJ-hEJ9uloaKU6smYTQR9RrnZTj