Month: July 2021

Towards a more open scientific community: go beyond open data and FOSS

The word ‘open data’ is never new to researchers in the data science/ statistics field. They can rapidly find and reuse various types of high-quality datasets provided by global companies, institutions and universities via open data repositories, thanks to the FAIR (Findable, Accessible, Interoperable, Reuseable) principle. GitHub, which allows users to share their code as Free and Open Source Software (FOSS), facilitates them to publish code openly for peers review. Researchers also prefer to publish the original version of articles on the arXiv so the community identifies them as the author of articles. Is it reasonable to call such workflow using open access repositories throughout a good practice of open science (OS)?

The answer could be yes if OS is merely defined as FAIR data and shared knowledge. Plenty of software, platforms and communities together consist of an ecosystem in which other researchers can contribute and collaborate in every stage of the research cycle, with data, result, protocol, etc. For instance, the RIO helps the researchers sharing the proposal and being open at the very first stage of research; a shared literature repository like Zotero is becoming the common choice to research references for academia. Besides the mentioned GitHub, there are also AsPredicted for pre-registering, Zenodo for sharing data, Protocals.io to share workflow in the analysis step. This ecosystem also ensures the writing process and publishment open the community. The list goes on.

However, the OS movement cares more than open data and share knowledge. The OS is a paradigm that goes beyond the conceptions of FAIR and FOSS. As the UNESCO (United Nations Educational, Scientific and Cultural Organization) states in the Draft recommendations to OS (2021), OS is an inclusive construct aiming to make scientific knowledge openly available, accessible and reusable for everyone, not only the scientific community but the whole society. The purpose is to benefit academia and society by increasing collaboration and share, and to open the process of scientific knowledge creation, evaluation and communication to the societal actors. It builds on the following key pillars: open access to scientific knowledge, open science infrastructures, open science communication, open engagement of societal actors and open dialogue with other knowledge systems. This blog will discuss particular pillars from a personal perspective. 

Open access to scientific knowledge It is no doubt that every researcher, even student is a beneficiary. The researchers obtain inspiration from all open objects, rather than solely the journals and conferences in a traditional community. The students are closer to the state-of-art academic results more than ever. An open scientific community connects every member more closer such that everyone is not studying silently. But the challenges exist. The wildly used open publishment repository Sci-Hub was suspended by Twitter due to “counterfeit content”. Before this incident, it has been banned in several countries after Elsevier filed a series of lawsuits against it. Therefore, some archivists launched a rescue action on Reddit to save this open repository. Will the OS movement finally become to conflicts and compromises among academia, commercial companies and governments? In the initial stage of the open science movement, these mentioned platforms act as the pioneers to support the scientific community in converting more open. If someday the OS paradigm becomes the requirement of all research activities, will these free-using platforms turn to be profitable? Will the enthusiasm to open accessible science eventually creating more monopoly companies/organisations? 

Open science infrastructures You can never overemphasize the importance of infrastructure to research in the STEM field. But building and maintaining advanced science infrastructures is a heavy financial burden to a university or even a country. The shared infrastructures are therefore built, maintained and accessed by members inside of alliances. Personally, it is a fantastic pillar. With more science infrastructures open to the scientific community, researchers performing big-scale computing can save tremendous time and funds wasted on rent/built supercomputers. But when the infrastructure is no more supercomputers, but devices involving national security, such as space station if it is infrastructure, is the rule “as open as possible, as close as necessary” of open data applicable here? Certainly, open means no threat to safety. Providers reserve the right to close the access to infrastructures, but abuse of this right to set a boundary to open science, to maintain the technical advantages, will aggravate the Mathew effect between developed countries and developing countries.

Other than the discussed pillars, transforming scientific culture and aligning incentives is the action that interests me. The universal assessment metric to research is the publishment in the traditional scientific community. Therefore, researchers manually modify results or data supporting the conclusion in some academic fraud cases, aiming to publish their discoveries in top journals. Except for calling for academic integrity, exploring the new incentives is a crucial issue for the scientific community. When every stage of the research process is open to the community and society, the contributions on data collection, analysis can be tracked and evaluated. 

In my case, I have already uploaded the data to Zenodo and the code to Github. I also applied an ORCID and combined it with my Zenodo and GitHub repository. This is my first step in practising research that meets with the OS paradigm. But as the title indicated, OS is more than open data and FOSS. I will open more research outcomes at every stage of my research process such that I could contribute to other researcher’s work and the scientific community evolving to a more open one.

UC3M TICKET TO OPEN SCIENCE

What you have learned during the course

In this course we went through the “Open Science” topic. Briefly, Open Science is a regular change in the methods used in science and research. It’s a practice of science where everyone can collaborate or contribute, sharing any kind of data related to their research freely.

What is Open Science? | Open Science Twente

Among all the topics which have been discussed and explained in the course, I would take the Session 4 Disseminating your research: Open and FAIR data as my favourite one.

In this Session, we covered the aspects about the data you use, generate and share as a researcher. One of the key points was how to make your data FAIR (findable, accesible, interoperable and reusable).

Under this assumption, everything you produced in the course of your research could potentially be shared with the community. For this reason, not only you can win the trust of other researchers, but your results could be checked and confirmed, or even new insights could be derived from your studies.

This constant feedback improves the scientific community by acting as a security mechanism during the research and after the publication.

What is your personal opinion about Open Science and what are the drawbacks for practising OS it from your personal field/career.

As a researcher, I am convinced that investing in improving your “Open Science” skills is a fantastic idea, not only because it eases your work, but also because it contributes to improving the scientific community.

Personally, I will try to incorporate all the tools we have been provided during the course.

Indeed, Open Science is a great tool for researchers. Nevertheless, the amount of extra time and resources it needs to be accomplished should not be ignored.

As a researcher, one of your main jobs should be publishing your achievements and your studies. Especially, when you take in account that the funding sources are limited for the researchers.

This is why, in my opinion, you should not only focus in trying to be the most “Open Science” you can, but also focus in your own publications and achievements.

How are your plans to become an “Open Scientists” during your PhD and beyond.

Among all the ideas which have been presented during the course, I would like to highlight some of my favourites.

To begin with, I will start making my code and data in a way it follows the FAIR guidelines. This means, I’m supposed to correctly comment my code, detail the metadata files for everyone could use it in the future or create a Data Management Plan, and such like.

In addition, the use of Open Science tools (such as Tenzing or ORCID) will allow me to improve my researcher skills and outcomes.

In conclusion, I really liked the course and I would recommend it to any researcher. The speakers (including those in the Open Science Cafés) have been fantastic, and the material (videos, slides, links…) is extremely helpful.

I got to know what real OS is. And that really matters to me now.

I signed up for this course with the idea that I was already a fully compliant Open Science researcher, so the content of the schedule would somehow ‘validate’ my prior ideas and procedures. To be honest, I didn’t expect great revelations. I could use the course just as a reinforcement of my already ‘accurate’ practices and perhaps with fortune I would get to know a few tools with which my life as a researcher would be easier. But the various sessions and speakers of the course have proved me completely wrong. It is well documented that we often evaluate experiences as a function of our initial expectations, and mines were moderate. In fact, I had attended several courses of similar duration and they were sometimes interesting but most of them barely scraped the surface. Truth is I am impressed with the breadth and depth that this course has brought to me on various domains.

First, from an ethic standpoint: I’ve learnt that following real OS guidelines is not a zero-sum game. On the contrary, by following Platinum OS path only one ‘stakeholder’ can be worse off: big (for profit) publishers. All the other relevant players (researchers, universities and society as a whole) are clearly benefited. Obviously, this is more ideological than anything, but it seems to me that it is no longer bearable to have a scenario in which all the detrimental impact comes to the side of the research community. Moreover, following real OS procedures also can bring about motivation: researchers collaborate as to build stronger networks, better resources and accessible knowledge at every step of the process.

Second, a resource perspective. I’m quite impressed with the amount of OS tools, platforms and repositories that are available. And this leads me to think that practicing OS nowadays is even easier and more efficient than ‘traditional publishing’, if I may. As a result, I already created my own ORCID account (https://orcid.org/0000-0003-4119-2232) and checked on Sherpa Romeo whether the Journal to which I submitted my latest piece of research had an Embargo Policy. Given that it doesn’t have it, I’m now discussing with my coauthors to which preprint repository(/ies) we can upload our research outcomes. Personally, I’ve gotten to know the project OSF by the Center For Open Science and I’m keen on making our work available there.

And thirdly, a resource management viewpoint. I already believed that research pre-registration was a essential practice in order to avoid fraud. For instance, by creating ad hoc hypotheses that matched the final results. That wouldn’t be science/knowledge, but a sophisticated sort of storytelling. However, an OS approach such as carefully designing & implementing a Data Management Plan (DMP) and making it available helps with the aforementioned potential pitfall. In fact, I’m now working on making a more complete DMP myself.

Finally, aside from the materials and resources that we have been offered, I wanted to encourage UC3M to keep investing in OS initiatives (as explained by the head librarian in our final session) and make an effort to advert this UC3M Ticket to Open Science course also for undergrad students too. I firmly consider that this course is going to have an impact on my research career. Hopefully it will have it on others too.

How to make science open, how to make Open-Science.

All over the years I spent in college I have been involved in innovation and entrepreneurship movements. The main idea where every entrepreneur movement spins around relates to sharing your idea from the very beginning. It is necessary and essential in order to know its limitations, and whether you may success in its development.

It really shocked me when I started my researcher career inside UC3M, during my master’s thesis and my current post as a PhD student. I did not really understand the reasons why researching was so far away from this policy.

Science, in the way I understand it, refers to figuring out solutions in order to make the world a better place. It may sound childish and innocent, but I think this motivation should be the engine pushing every single step in the researching mechanism.

Science, in the way it actually works, refers to publishing more, choosing quantity over quality. When related with scholarships and grants for developing your PhD, only JRC publications are taken into account, and the more you got, the easier is to achieve that precious goal, no matter how many years and effort you have put into your project.

Ticket To Open Science has brought me closer to the reasons why I became a researcher at first, and a PhD student afterwards. My main goal is, and has always been, to bring engineering closer to social and medical issues and try to find out the way to solve them in low-price and feasible terms. I cannot see this happening through conventional research, where I cannot find the way to share my knowledge with those who could take it and turn it into something better; or make me realize where my weaknesses are in order to develop the best possible product for the final user. All through this seminar, I have understood that Open Science is the way to do it.

For those like me who were not familiar with this term, Open Science refers basically to a combination of procedures that make scientific knowledge openly available and reusable for everyone. Everything inside this project should be open, from your raw data and methodology to your results and discussions.

By sharing every step accomplished, a reproduction of your research is achievable. Hence, errors and updates can be made, and your project can be pushed to the next level, creating something bigger than what you could have done in your own, or simply surrounded by those members of your research group. Additionally, your ideas can serve as an inspiration and a guiding principle for future researchers all over the world.

Furthermore, society can be involved through citizen science, encouraging them to express their problems, test possible results, and learn about different studies that may help them in their daily issues. Researchers do benefit from citizen science as well, receiving opinions and even ideas and personal work from those interested.

As a researcher, I have learnt different repositories where to find papers in the Open, which has been really helpful for me. As long as I have been working with medical issues, where data has a high level of security and confidentiality, it has been specially difficult even to find basic information about joints and individual measures and characteristics. Through these repositories (such as “Dimensions”, etc), some information which I found difficult to find was, not only available, but also easier to gather and to organize.

I have also learnt the importance of publishing in the Open, not only the final papers and studies, but also my coding, raw data, methodology, etc. My first step as a PhD student in order to find my way into the Open Science has been to, along with my research team, send and try to publish our last study in an Open Journal (MDPI). Furthermore, I am starting to develop my Data Management Plan and trying to adapt it to Open Access. I may need to dig deeper in this concept as long as I will be working with medical information and confidentiality is of the upmost importance and difficulty.

I have included this slide because I find it really helpful for having a better understanding about what each researcher can do in order to push Open Science Up.

However, all that glitters is not gold, and Open Science has still some limitations and drawbacks. I find the rather high price needed to publish in Open Science one important limitation, as long as publishing will report a lower personal or institutional benefit in terms of status. Thus, institutions will rather publish in the close where (even though you have to pay too) higher benefits are received; such as JRC ranked publications which may serve you in order to receive financing. Furthermore, Open Science is not that widely spread, so people do not tend to search for specific papers in the Open Sources, and citations may decrease. From my point of view citations should not be that important and Open Science will undoubtedly increase citations in the near future but, being realistic, some measures, such us the h-factor, do represent our ranking as researchers and Open Science still needs a little push up. However, this type of drawbacks can be overcome as time goes by, and Open Science is pushed higher and higher.

So, long story short, I may say that, through this seminar, I have found in Open Science the perfect tool to turn my PhD experience into something society can hopefully benefit from while I grow up as an Open Researcher.

UC3M TICKET TO OPEN SCIENCE

Special Training for PhD Students

What you have learned during the course

Open Science Monitor : access to data and trends on open science | AIMS

The course contributed effectively by presenting concepts and strategies of fundamental relevance to my UC3 PhD Project.

Important concepts such as Open Science, Open Knowledge, Open Research and Open Scholarchip contributed to a broad and global reflection on my research area.

The Course adds a valuable contribution to the researcher.

What is your personal opinion about Open Science and what are the drawbacks for practicing OS it from your personal field/career.

As a researcher I believe and have invested in Open Science.

Open Science is a movement that proposes structural changes in the way scientific knowledge is produced, organized, shared and reused. It is a new way of doing science, more collaborative, transparent and sustainable.

What is Open Science? | Open Science Twente

How are your plans to become and “Open Scientists” during your PhD and beyond

The course presented several possible paths that the researcher can follow:

– How to become an Open Responsible Scientist

– Open Science Tools and colaborative research

– Planning Your Strategy to Open Science

– Working in a reproducible way

– Diversity, ethics and other researc standards

In conclusion, the course is very well structured from a content point of view, providing to the PhD student at UC3 Madrid a precise overview of Open Science.

The Open Science Café presents practical and applied views on the subject – an excellent and innovative initiative!

The course stimulated me to publish an article on the subject:  eSCIENCE AND SCIENTIFIC COMMUNICATION IN OPEN SCIENCE

Citizen Science as a tool for Open Science

The most interesting topic among the ones taught in this Open Science course was the one about citizen science. That’s why I decided to check Scistarter, a website that listed a lot of citizen science projects [1]. When looking for projects related to my field, Engineering, I found two interesting ones:

Potential Penguin [2]

Potential Penguin is a game that facilitates the learning of classical Physics concepts. By modifying the terrain, you have to help the penguin reach the goal with the indicated amount of potential and kinetic energy.

By playing the game, you are contributing to the project in two ways:

  • How visualization can help in learning physics concepts
  • What improves the performance in our quantum games

Project Sidewalk [3]

Project Sidewalk uses labels placed by users to improve city planning, build accessibility-aware mapping tools, and train machine learning algorithms to automatically find accessibility issues. There are two ways of collaborating with the project: placing labels or verifying labels placed by other users.

Example of label verification
Example of label placement

What am I doing to improve Open Science

To generally improve the progress of the Open Science initiative, I’ve sent two papers to two different journals as Open Access publications (paying fees up to 2500 euros before taxes …). These papers are part of the research done for the European project Labyrinth , part of the Horizon2020 initiative, which promotes Open Science.

References

  1. SciStarter – SciStarter. (2021). Retrieved 2 July 2021, from https://scistarter.org/
  2. Games, S. A. H. : C. S. (n.d.). ScienceAtHome: Games: Potential Penguin. ScienceAtHome.org. https://www.scienceathome.org/games/potential-penguin/.
  3. Project Sidewalk. (2021). Retrieved 2 July 2021, from https://sidewalk-sea.cs.washington.edu/
  4. Labyrinth 2020 – Ensuring drone traffic control and safety. (2021). Retrieved 2 July 2021, from http://labyrinth2020.eu/

Comments on Open Science Lecture

I see the OpenScience movement as the academic equivalent of the OpenSource movement in software development, where coders share their work to build something bigger and better amongst all the participants, newcomers and, lately, event multinational companies.

The interesting thing about OpenScience is to share methods, conclusions, measurements/datasets with the rest of the academic community so experiments can be replicated and improved by others. This issue is not trivial as according to Kostas Glinos, the head of the Open Science office at the European Commission, it could cost up to €300 billion/year the effect required to reproduce papers with not enough information, not sharing software or by correcting mistakes that the researcher may have introduced unintentionally. Even with the COVID crisis, most of the research data that was shared was not fully interoperable and reusable to help resolving the vaccine development in an effective way. So I agree that a change in the way we do research needs to be changed to reward team work or collaboration.

Current journal publishers are taking advantage of the need to publish papers as part of the PhD process to exploit the researchers, in my modest opinion. These publishers require you to provide the result of your work for free and then, if you are “lucky” then you can present it in a congress. This implies that you have to pay for the travel and the hotel expenses but, in most of the cases, you also have to register to the congress and pay the full tariff to present your work for free. In addition, not happy with that, they take ownership of your paper and sell it to the rest of the community and the author, in most of the cases, does not receive anything. If you are on the paper reviser side, you also need to dedicate your time to reading and making comments on paper for free, sometimes on very special dates as on Christmas holiday, as it happened to me very recently. That’s why sites like https://www.reasearchgate.com or https://www.arxiv.org are worth to collaborate with, as the create an open environment where your work can be shared for free with no third party taking benefits of your work.

On the other hand, these publishers also have an important contribution related to filtering the papers in terms of quality, style, length etc. Nowadays, millions of papers are submitted every year to the thousands of journals and publications available worldwide. The amount of information is so huge that it is nearly impossible to read everything. On the other hand, not all the published papers are relevant and, as PhD candidate, you can waste a lot of time reading papers with low innovation, small advantages or they are just a remake of previously seen papers. Sometimes I feed like Einstein where he was amazed by the amount of irrelevant information that was available at the beginning of the 20th century that he decided to start his thesis form scratch and firs principles. I am not that clever, not have such determination, but I prefer having a lag of 1 to years in the papers that I read to ensure that, as time goes by, they demonstrated that they were relevant to the advance of science. By doing that, I save a lot of time. I do not want to image what would be the situation if these publishers that we were mentioning before were not doing any kind of filtering in the information.

From the academic world point of view, current evaluation system is not optimum as we PhD students need to publish our research papers on high Impact Factor (IF) journals, as well as our tutors and thesis directors. Which leaves your research curriculum in the hands of the big academic publishers and their fees.

The idea sounds good but we need also to think about the Intellectual Property (IP) issues and when it is the right time to share our work with the rest of the community is also an important point to consider. Although sharing information is the way forward, as it happens in the software industry with the open source code, there must be also an effective way to protect authors against legitimate usage of the information when the license conditions are not respected. Because, in practice, it is very difficult to control if the user of the information is compliant with such license conditions. It is clear now to me that patents are not part of Open Science, but, on the other side, I think it is a good idea that all the data that are paid by public founds are available for other researchers.

At an international level, there is also significant effort to promote OpenScience. Some of the initiatives that we have reviewed during this seminar are:

  • ECSA (European Citizen Science Association) that promotes Citizen Science where everybody is encouraged to participate on scientific research and also contributing to research progress by connecting normal citizens with researchers.
  • Plan S from Science Europe where objective is to ll scientific publications resulting from publicly funded projects are published in Open Access as soon as …. 2021!!!!.
  • last, but not least, the European Commission, as mentioned above, that is developing a server where scientific publications, research data can be shared amongst the research community using the so called European Open Science Cloud.

In my particular case, I am going to open a github account where I will share with the community the partial results of my thesis investigation whenever I use public datasets or data, so everybody can take advantage of such work and improve it, which at the end of the end that’s the final target of Open Science philosophy. I have also created my ORCID account (https://orcid.org/0000-0002-3755-1661) with regards the paper that I have just submitted for publication to one of these big publishers (required as one of my thesis targets for final evaluation), I will also submit my paper to https://www.reasearchgate.com or https://www.arxiv.org as soon as the quarantine period will be over. And, finally, during this seminar I have also discovered new tools such as https://zenodo.org/ and https://www.openaire.eu/ to look for data or datasets to support my thesis. I have not been successful to find data to support my thesis, which is dealing with defects on castings and weldings X-ray images, but I will be trying.

Thanks to the organizers of this seminar to let us know these platforms of additional open source data and research !!!!.

Open Society And Open Science: Challenges And What We Can Do

Introduction

For all PhD students, “Open Science” courses are very necessary and helpful. From my own experience, this course benefit me at least in the following three points:

Resources – This course has introduced abundant open science resources to the students. For example, It tells students how to use the DORA, OPENIRE, and CORE and etc., which are very useful for students’ research practices.

Principles – This course has taught students what is “open science”, and explained the main principles of “open science”, in particular, It gives some specific strategic suggestions from the perspective of doctoral students’ scientific research and practice, when preparing to publish and store their academic results. It has also taught us how to use open science tools to make more people to be engaged in our research, and get more open data to be utilized in our study.

Thinking – This course has aroused my deeper critical thinking about “science”, “academic” and “society”. In the past, I was very interested in the philosophical concept of “open society”. After taking this course, I think that “open science” is one of the most important characteristics of “open society” in the development of academic research. In this blog post, I will discuss open society and open science briefly.

Open Society

The concept of “open society” was put forward by the French philosopher Bergson(1932). He defined it as a dynamic system inclined to moral universalism. And then during WWII, The famous British Austrian scholar Karl Popper further developed this term of “open society” in his masterpiece “The open society and its enemies“(1945). He saw it as “part of a historical continuum reaching from the organic, tribal, or closed society, through the open society (marked by a critical attitude to tradition) to the abstract or depersonalized society lacking all face-to-face interaction transactions”.

In Popper’s analysis framework, there are two streams about the term “Open society”. One is from the view of political science: in the open society, individual freedom and human rights are the cornerstones, the government should allow and accept criticism from the people, the political mechanism should be transparent and flexible, as opposed to authoritarianism. The other stream is from the perspective of epistemology: Popper believes that there is no so-called “ultimate truth”, critical rationalism is the philosophical basis of scientific skepticism, everything must be kept open.

Open Science

“Open Science” can be traced back to the Enlightenment period in the 17th century, but the exact phrase for “Open Science” was created by Steve Mann in 1998, when he registered the domain names openscience.com and openscience.org. For the term “open science”, different scholars and institutions have given various definitions. Although expressed in different ways, the concepts of “openness, cooperation, and sharing” are fully permeated. According to UNESCO(2021), open science builds on 5 key pillars: open access to scientific knowledge, open science infrastructures, open science communication, open engagement of societal actors and open dialogue with other knowledge systems. Currently, open science is leveraged and utilized wildly in academic research, which includes government management, business, education, publishing and other fields.

Significance of Open Science in An Open Society

Is today’s society an open society? There is no clear answer to this question. However, it is clear that with the development of information technology and the media industry, especially the rise of the Internet since the 1980s, our society has become more and more like an “open society”. The limitations of information dissemination and storage have been broken in both space and time. In the present time, within the internet everyone can freely express their own opinions and create personal content, such as videos, under the scope of the rules. Popper’s “open society” seems to be gradually becoming a reality on the Internet Era.

However, for academia, it is still very conservative. Especially in the academic publishing industry. According to a statistics conducted by the European Universities Association (EUA), the overall expenditure by 26 European countries for academic journals was €597 million (£515 million) in 2017, but 75% of that – some €451 million – was spent on subscriptions to journals published by the “big five” : Elsevier, Springer Nature, Wiley, Taylor & Francis and the American Chemical Society (ACS). Together they accounted for 56% of articles published (Mehta.A, 2019). With the monopoly advantages in the academic publishing industry, the “big five” have formed a de facto oligarchy. For commercial interests, the “big five” have built a high wall for publishing activities and access to products, which is not in line with the concept of open science.

From where I stand, as a part of the panorama of open society, open science is an irresistible trend. The development of new technologies represented by Internet technology will ultimately affect the academic publishing ecology, which is fully illustrated by the current facts. In February 2018, the European Open Science Cloud Initiative (EOSCI) proposed a framework for open science; on July 4, 2018, the French National Academy of Sciences released the National Open Science Program; on July 17, 2018, the National Academy of Sciences of USA released the plan of ” Open science by design”.

Two of The Most Biggest Challenge For Open Science

Why does the academic publishing industry present a closed feature? In my opinion, there are mainly the following two reasons:

Firstly, the current academic evaluation system is flawed. To get better recognized, most academic scholars have to submit and publish their research papers on high impact factor (IF) score journals, which are basically monopolized by the “big five” academic publishers. While, can we say those academic workouts, such as research book chapters or conference papers, have no value? Obviously the answer is no. Numerous scholars are the fundamental and decisive force in promoting the development of open science. Only when the open science work results of these researchers are fully recognized and directly linked to their performance evaluation, more and more scholars will have the motivation to participate in the open science initiative, and the open science movement can become more and more energetic, and finally succeed in practices.

Secondly, the profit model of academic publishing is too closed. According to the well-known laws of economics: “monopolistic enterprises can obtain excess profits through their monopoly position”, those “big five” companies, for their business interests, they are raising the subscription service fees of their published products year by year, which has led to protests from many universities, research institutions and scholars. Especially for scholars, they may have to pay high fees for publishing papers and when they want to get access to the published papers, they have to pay for it again. Some funding agencies like the European Research Council (ERC) also protested it, because when people want to get access to the results of the funded research, they may be involved in the issue of secondary payments. Academic publishing companies argue that their services price hikes were due to the increasing costs of publishing and printing, but this explanation does not convince most of the people. 

In order to solve the problem above, the governments or the public sector can increase investment and promote the open scientific infrastructure, which will strengthen competition in the academic publishing industry and guide the business model of academic publishing to a benign transformation. The government can also issue relevant policies or laws to reduce the fees for certain services in the academic publishing field, or the government can subsidize academic publishing companies instead of those companies directly charging scientific research relevant entities, or the government can build up academic publishing institutions for non-profit purposes. Because in general, breaking down the barriers to knowledge and science dissemination will promote economic growth and improve the overall level of welfare eventually.

What can we do?

As an early stage researcher, more specifically, a PhD student, we will go through the whole process of academic research and academic publishing. In my opinion, for becoming an “open scientists” during our PhD study and beyond, we can draw efforts from the following 4 points:

Above all, we should strengthen our belief in the concept of “open” in today’s society. In life, we should be open and embrace new technology, new information. And in academic research, we should be confident in the open science trends and internalize it into our inner values.

Besides, we should make good use of open scientific resources and tools. When we conduct the proposed research, we should make full use of the skills learned in this course and apply them to our research work. For example, when we are looking for literature or data relevant to our research, we can refer to the open platform like Dora and etc. And when we need to collect large scale data, we can encourage more people to get engaged in our research. These practical skills will be helpful to our own research.

And as for our research work outputs, we could try to publish and upload our papers, data and relevant materials in open science platforms. If we are not clear where these platforms are, we could ask the university library for help, they provide a variety of comprehensive services. And we will also submit the doctoral dissertation here at the end, which is an open science platform in some sense. And even though we plan to submit and publish our papers in “big five” journals, we can still use the pre-print version to relevant open science platforms. Only when more and more scholars continue to enrich the content of open platforms, can these platforms truly promote the dissemination of knowledge, inspire academic debates, and promote the development of academic science.

Last but not least, we should actively spread the idea of open science. Just as the value of money is to make it flow, so is the idea of open science. We pass on the knowledge of open science to more people, and more and more people will participate in the open science initiative. And if it’s possible, we can join a professional open science working group and use practical actions to promote the development of open science.

References:

  1. Henri Bergson ([1932] 1937). Les Deux Sources de la morale et de la religion, ch. I, pp. 1–103 and ch. IV, pp. 287–343. Félix Alcan.
  2. Popper, K. R. (1945). The open society and its enemies. London: Routledge & Kegan Paul.
  3. “openscience.com historical whois information – who.is”. who.is.
  4. Azoulay, A. (2021). UNESCO embraces open science to shape society’s future. Nature, 593(7859), 341-341.
  5. ANGELI MEHTA (2019). 75% of European spending on scientific journals goes to ‘big five’ publishers, Chemistry World, Available online:https://www.chemistryworld.com/news/75-of-european-spending-on-scientific-journals-goes-to-big-five-publishers/4010616.article(accessed on 1st July 2021).
  6. European Commission. The European Open Science Cloud Initiative (EOSCI) 2018. Available online:https://ec.europa.eu/info/research-and-innovation/strategy/strategy-2020-2024/our-digital-future/open-science/european-open-science-cloud-eosc_en. (accessed on 1st July 2021).
  7. European Open Science Cloud. France National Open Science Plan (2018).  Available online:https://www.enseignementsup-recherche.gouv.fr/cid132531/plan-national-pour-la-science-ouverte-discours-de-frederique-vidal.html. (accessed on 1st July 2021).
  8. Mike Henry (2018), National Academies Envisions ‘Open Science by Design, American Institute of Physicals (AIP), Available online: https://www.aip.org/fyi/2018/national-academies-envisions-%E2%80%98open-science-design%E2%80%99. (accessed on 1st July 2021).

A way to improve science

Before this course in Open Science I was unaware of most of its goals and even had some misconceptions about others. Now, after seeing what adopting Open Science can offer, I believe that this is a better way of doing research that the scientific community and the society as a whole needs. If science is all about trial and error until we get results, why not do the same with the research methodology itself?

A big problem in the current paradigm of research is the assessment of scientific production, which is heavily affected by each reasearcher’s amount of publications, citations and the journal they are published in. While this metric may be important, it fails to accurately determine the imapct of a researcher’s work. It can be argued that this encourages publishing many papers in detriment of their quality in order to not fall behind. This could happen to scientists in any stage of their carrers, but specially can affect young and not yet established researchers which need to fiercly compete with their peers. Some situations may generate fear and anxiety of being rejected by the community not based on the quality of the research but by other unfair reasons.
Even worse, some very important investigations may even never be undertaken because they are too hard, their results a priori uncertain or may take a lot of time. Another downside is that scientists may become ‘too busy’ to grow interest in other areas in contrast with the relevance of interdisciplinary research, which grows as does our knowledge. Because of this we might be missing out on wonderful science or even discouraging young bright minds to get into a research carrer.

The nature of science is coorperative, and that enriches it: the more widespread an idea is, the more new ideas it can generate. However it seems that the current research paradigm has deviated from that idea to become strongly competitive. It seems counterintuitive that this is happening in the era of Internet, which should make cooperation easier. It would be in the scientific spirit if all data, method, procedures – in short, science – were openly shared for anyone to see and use. This idea raises some alarms because in the traditional paradigm, opening up your research likely implies being able to publish less original papers or risking not getting the due credit. This indeed is a valid concern if the assessment metrics are the traditional ones, but the picture radically changes if the community starts valuing other criteria more.

In the same spirit as the previous criticism, it is evident that journals should change their policies regarding publication. The current business model usually includes a pay wall at some point or embargoes which work against openness. There are several ways a journal can treat the issue of open access, called open access routes. All journals need money to run but the difference lies in who and how much pays. For example, in diamond OA, users do not pay and the journal is funded by institutions or other stakeholders. In gold OA, authors usually pay a fee to publish. The goal would be to remove all payments from individual researchers or readers.
There also exists the issue of licensing: ideally all rights and credit belong to the author and journals do not own the research at any point. However, traditionally there are embargoes or other licensing practices that empower the journal over the researcher.
Plan S is an initiative aiming to address these issues. Its principles state, among others, that authors should keep the copyright of their works, that fees for publishing should be limited and payed by funders or insitutions. Some open access alternatives may even not be considered by researchers because in some areas it is almost mandatory to publish in a certain distinguished journal in order to be read or have any impact.
Beyond the problem of publishing, an open scientist should make sure that their research is FAIR, this is, findable, accessible, interoperable and reusable. This criteria is aimed towards making data as open as possible in a way that is useful for other researchers. Of course, there are many legal technicalities involved, such as sensitive information, but they can be handled via licensing, for instance.

In my opinion, for the reasons I mentioned above, Open Science is an alternative to the traditional paradigm that the scientific community desperately needs and that everyone would benefit from such a system. However, I think that the major challenge we face (and its weak spot) is the transition from the well established ‘Closed Science’ to Open Science. For instance, journals with high impact factor naturally draw a lot of publications, making it more difficult for other (maybe open access) journals to gain prestige. Another example could be that researchers focus their efforts to maximize the current assessment metrics (and not prioritizing better metrics) hence perpetuating the system.

For my professional career in mathematical physics, open science would be very beneficial for us scientists, and I think me and our group don’t have to change much our modus operandi. As for data, since we don’t work with physical experimental measurements and our works are always on a theoretical basis, we rerely work with extra information that we don’t already write in papers, and it is customary in our field to publish preprints in arXiv.org simultaneously to submitting them to an indexed journal. We also use ORCID for identification.
I have been lucky enough to do my PhD with a community of people that embrace open science principles and have passed them down to me. Before this course I could not spell out why they were good or beneficial and I even was unaware of some of the traditional practices that are considered ‘not open’, but thanks to it now I can fully identify and appreciate what and how they do science and I plan to stick to these principles during my PhD and throughout my career. There are still of course some points that I would like to be addressed, for example to increase the percentage of publication in open access journals. Although as a community we are not where I would like to be in terms of openness, we are advancing at a good pace. Personally I think it is reassuring for the future that a different, modern way of doing science is possible and I am eager to be a part of it.