“PEER COMMUNITY IN” PROJECT
FUNCTIONING OF THE VARIOUS PEER COMMUNITY IN X (PCI X)
PCI X STRUCTURE
SELECTION OF ARTICLES
ARTICLE EVALUATION PROCESS
TRANSPARENCY AND ETHICS
FATE AND CITATIONS OF RECOMMENDED ARTICLES
COMMENTING ON ARTICLE RECOMMENDATIONS
“PEER COMMUNITY IN” (PCI) PROJECT
The PCI project is a non-profit scientific organization that aims to create specific communities of researchers reviewing and recommending preprints (unpublished articles) and, to a lesser extent, recommending postprints (published articles) in their field. These specific communities are entitled Peer Community in X, e.g. Peer Community in Evolutionary Biology.
The goal of the PCIs is to highlight and recommend preprints as of particular interest to the community concerned. The preprints recommended by PCIs are complete articles of high value that do not necessarily need to be published in traditional journals.
The aim of PCI is to offer scientists a free, stimulating, transparent and non-exclusive way to validate and promote their scientific output, by removing this monopoly from journals.
PCIs publicly highlight and recommend high-quality preprints. The preprints recommended by PCIs are complete articles of high value, that are reliable and citable without the need for publication in traditional journals.
Recommenders evaluate, and may choose to recommend preprints that have not been published by or submitted to a journal. To a lesser extent, they may also recommend postprints, i.e. articles already published in journals. Recommender is a role similar to that of a journal editor (finding reviewers, collecting reviews, making editorial decisions based on reviews), with the possibility of recommending the preprint after one or several rounds of review. When recommenders decide to recommend a preprint, they write a “recommendation”, which has its own DOI and is published in the electronic journal of the PCI in X. The recommendation is a short article, similar to a News & Views piece. It has its own title, contains between about 300 and 1500 words, describes the context and explains why the preprint is particularly interesting. The limitations of the preprint may also be discussed. This text also contains references (citing at least the recommended preprint). The recommendation of a postprint is more straightforward: the recommender selects a postprint that he/she wishes to recommend and invites a second recommender from PCI in X to write a joint recommendation text.
A “recommendation” is a short article written by one or several recommenders, describing why an article (preprint or postprint) is particularly interesting. It has a DOI and is published in the electronic journal of the PCI. The recommendation is a short article, similar to a News & Views piece. It has its own title, contains between about 300 and 1500 words, describes the context, explains why the preprint is particularly interesting and contains references (at least one, to the preprint recommended). The limitations of the preprint may also be discussed.
Stimulating: Peer Community in X recommends remarkable articles.
Free: there are no fees associated with the evaluation process, and no charge for access to the comments and recommendations. The website is freely accessible.
Transparent: Reviews and recommendations (for preprints) and recommendations (for postprints) are freely available for consultation. Recommendations are signed by the recommenders. Reviewers may choose to remain anonymous or to sign their reviews.
Based on sound and independent evaluations: Recommenders and reviewers must declare that they have no conflict of interest with the authors or the content of the preprint they evaluate and recommend. The Managing Board performs a quality control check on the format and deontology of reviews and recommendations.
Not exclusive: A preprint may be recommended by different Peer Communities in X (a feature of particular interest for articles relating to multidisciplinary studies) and may even subsequently be published in a traditional journal (although this is not the goal of the PCIs).
Several Peer Communities in X will exist in various scientific fields (e.g. phytopathology, ecology, cancer research, etc.). The goal is not to set up a monopoly, and several alternative recommending systems may coexist with the PCI project. Preprints recommended by PCIs would not necessarily require subsequent publication in classic journals, because they are high-quality complete articles that are reliable and citable in their own right. In time, we expect PCIs to become the major medium for the publication of scientific output.
F1000Prime: Readers have to pay to read F1000 recommendations. F1000Prime is a for-profit business and is not intended to replace the current system based on journal publications. F1000Prime recommendations concern only published articles. F1000Prime does not, to our knowledge, wish to recommend preprints.
Winnover is “an open access online scholarly publishing platform that employs open post-publication peer review”. There is a fee ($25 per DOI) and it does not provide any recommendations. Winnover allows authors 1) to upload an article onto their platform and then encourages researchers, colleagues, and other scientists to make critical comments on the article over a given period of time and 2) to revise the article on the basis of the comments received and to provide the new version with a DOI. It is therefore not a “recommendation” as such, but an open process of critical review without a given threshold determining whether an article may be considered scientifically “valid”.
The Peerage of Science initiative operates upstream from the publication system and provides support to existing scientific journals. It is not, therefore, intended to replace the current system. The goal is to submit an article actively to obtain constructive criticism before submitting it (and the responses to the criticisms received) to a scientific journal. It is stated that “Authors may accept a live publishing offer from a subscription-only journal, or may choose to export the peer reviews to any journal of their choice.” and that “The revenues of Peerage of Science come from organizations wishing to purchase the peer review services for use in their decision-making, such as publishers, funding organization, and universities.” Again, this is a very different model from the PCI project.
Episcience has the objective of favoring the emergence of “epijournals”, electronic open-access journals containing items deposited in open archives, such as arXiv, and not published elsewhere. Our project is not intended to create a journal. It is, instead, a system for recognizing remarkable articles by awarding them a recommendation, and this recommending process is possible for both articles that have already been published and for articles deposited in open archives.
eLife is a journal publishing original articles with publication fees. “A fee of $2,500 is collected for published papers” (https://submit.elifesciences.org/html/elifeauthorinstructions.html#fees).
PeerJ: Authors have to pay to publish in PeerJ. Either they pay $1,095 to publish a paper or each author pays a one-off fee of $399 (or more) allowing them to publish 1 (or more) paper/year in the journal (https://peerj.com/pricing/#apc-membership-pricing).
The non-profit PCI organization is responsible for the creation and functioning of the various specific Peer Communities in X. The members of the Managing Boards of each Peer Community in X will also be members of the non-profit “Peer Community in” organization. Hence, representatives of all existing Peer Communities in X would collectively take decisions concerning the creation of each new Peer Community in X. If you would like to launch a new Peer Community in X, you should contact a Managing Board member and explain your project.
Yes, and this is one of the chief advantages of PCI. The recommendation process is not exclusive and articles of interest to several different Peer Communities in X could be recommended by all those communities. This aspect is of particular interest for articles dealing with multidisciplinary studies. There would be no a priori hierarchy of communities, although some would be highly generalist (e.g. PCI Mathematics) whereas others would be more specialized (e.g. PCI Entomology). However, to avoid recommendations of various versions of an article, a preprint already recommended by a Peer Community in X could only be recommended by another Peer Community in X as it stands. In other words, once a Peer Community in X has recommended a preprint, this latter must be considered reviewed (i.e. like a published article) by all the other Peer Communities in X interested by recommending it.
The PCI project is an original idea of Denis Bourguet, Benoit Facon and Thomas Guillemaud, working at INRA (the French National Institute for Agricultural Research) in France.
FUNCTIONING OF PEER COMMUNITY IN X (PCI X)
Each recommender is expected to manage the evaluation process for one or two articles per year, on average. Thus, becoming a recommender for PCI X is not associated with a substantial workload. Recommenders evaluate and may recommend preprints that have not been published by or submitted to a journal. To a lesser extent, they can also recommend postprints (i.e. articles published in journals). Evaluating a preprint means playing a role similar to that of a journal editor (finding reviewers, collecting reviews, making editorial decisions based on reviews) and, possibly, recommending the preprint after several rounds of review. Recommenders deciding to recommend a particular preprint write a “recommendation” for that preprint, which has its own DOI and is published in the electronic journal of the PCI in X. The recommendation is a short article, similar to a News & Views piece. It has its own title, contains between about 300 and 1500 words, describes the context and explains why the preprint is particularly interesting. The limitations of the preprint can also be discussed. This text also contains references (referring at least to the preprint recommended). Recommending a postprint is more straightforward: the recommender chooses a postprint that he/she wishes to recommend and invites a second recommender from PCI in X to write a joint recommendation text.
New recommenders are nominated by current recommenders and approved by the Managing Board. If you are interested in becoming a recommender, please contact a current recommender in your field.
Each recommender manages the evaluation process for one or two articles per year, on average. No recommender is allowed to handle more than five articles per year, to minimize the risk of a few recommenders dominating the recommendations made.
Scientists will care because the recommendations are attributed by a recognized group of colleagues, free of charge, and colleagues, employers and funding agencies will almost certainly recognize this recommendation as a mark of quality.
PCI X STRUCTURE
We expect PCI X to bring together several hundred recommenders, but there will be no restriction on numbers. This size would be sufficient to recommend a large number of articles, even if each recommender recommends only one or two articles per year.
The Managing Board of PCI X is a group of recommenders from this community. Its principal function is approving the nomination of new recommenders for PCI X. The Managing Board also deals with problems arising between authors and the recommenders responsible for evaluating and/or recommending articles. It detects and deals with dysfunctions of PCI X, and may exclude recommenders, if necessary. It also performs quality control checks on the format and deontology of the reviews and recommendations published by PCI X. Finally, members of the Managing Board of PCI X are part of the non-profit organization “Peer Community in”. This non-profit organization is responsible for the creation and functioning of the various specific Peer Communities in X.
The Managing Board of PCI X has between about 5 and 15 members chosen among the recommenders. The modality of choice depends on the community.
The Managing Board of PCI X is mostly responsible for nominating new recommenders for PCI X. It also deals with problems arising between authors and the recommenders responsible for evaluating and/or recommending articles. The Managing Board detects and deals with dysfunctions of PCI X, and may exclude recommenders from this community, if necessary. It also performs quality control checks on the format and deontology of the reviews and recommendations published by PCI X. Finally, members of the Managing Board of PCI X belong to the non-profit organization “Peer Community in”. This non-profit organization is responsible for the creation and functioning of the various specific PCI X.
- PCI X is dependent on the “Peer Community in” non-profit organization (see https://peercommunityin.org).
- Preprint authors do not have to pay to have their articles evaluated or recommended by PCI X.
- Readers do not have to pay for access to preprints, recommendations, peer reviews and editorial correspondence published by PCI X.
- Reviewers and recommenders are not paid for their work for PCI X (just like the reviewers and associate editors of journals published by for-profit companies).
- The founders and members of the Managing Board are not paid by PCI X.
- The scientific and administrative work underlying the functioning of PCI X corresponds to approximately one full-time position (comparable to the work of an Editor-in-Chief, usually paid by his or her research institution, and possibly the work of both a managing editor and an editorial assistant of a journal), or two full-time positions if we include the development of the global PCI project.
- The maintenance of the web site and the management of PCI X e-mail addresses corresponds to approximately 0.1 full-time positions, or 0.5 full-time positions if we include the development of the global PCI project.
- The people currently performing these tasks are employed by a French public research institute (The French National Institute for Agricultural Research, INRA).
- The annual (non-salary) operating budget of PCI X is about €5,000 (equivalent to the costs to authors of publishing two to four open-access articles in major evolutionary biology journals), or €20,000 if we include the development of the global PCI project.
- This budget is currently funded by grants from research institutes (e.g. INRA), research laboratories and societies (e.g. SSE).
No, they are not.
Yes, the Managing board can exclude recommenders if their recommendations are of insufficient quality or if they do not respect the code of ethical conduct of PCI X.
Yes. PCI X regularly backs up its data in several mirror web sites. Recommendations and peer-reviews are deposited in HAL open archive.
SELECTION OF ARTICLES
Recommended articles may have diverse formats: reviews, comments, opinion articles, research articles, data articles, technical notes, computer notes, movies, etc. No editing, formatting or proof-reading of the recommended articles is required. We only ask the authors of recommended articles to add a cover page to their preprint and a sentence at the beginning of their abstract stating that their preprint has been peer-reviewed and recommended by PCI X.
The goal of PCI X is to recommend preprints deposited in open online archives and, to a lesser extent, articles already published in scientific journals. Peer Communities in Evolutionary Biology recommenders are encouraged to recommend articles in the following descending order of priority: 1) preprints deposited in open online archives (hence freely accessible) and not yet published in scientific journals (to encourage the expansion of the recommendation process), 2) articles already published in scientific journals but published with open access (and preferably free of charge for authors) and 3) articles published in scientific journals but not with open access.
Any preprint can be recommended, provided a PCI X recommender is willing to manage its evaluation and to recommend it. However, not all submitted preprints are recommended, because not all preprints achieve the required quality.
Authors are invited to submit their preprints to PCI X via a dedicated tool on the PCI X website. For submission, the preprint must not have been published or be under consideration for publication in a traditional journal. If a PCI X recommender is interested by evaluating the preprint he/she initiates the evaluation process.
Yes, probably. Authors are invited to submit their preprint to PCI X. But depending on the size of PCI X, the number of preprints awaiting evaluation, and their quality, a fraction of those preprints may not be considered.
The processes of publication and recommendation are not exclusive: a preprint evaluated by PCI X can subsequently be submitted to a journal. However, concurrent submission to a journal and to PCI X – i.e. submission to PCI X at the same time as submission to a traditional scientific journal – is not allowed.
No, parallel submission is not possible because:
– Parallel reviewing might decrease the importance of the PCI review in the authors’ eyes. If a paper is accepted for publication in a journal before recommendation by PCI X, the authors may decide not to make the modifications requested by the PCI reviewers. This might also decrease the willingness of researchers to act as recommenders and reviewers for PCI X.
– Recommenders and reviewers may consider it a waste of time recommending or reviewing preprints that have already been submitted to a journal. Recommenders are likely to prefer to wait for publication of the article before its recommendation for PCI X (without the need for further review).
– A growing number of journals are accepting PCI X evaluations as part of their own editorial process.
– Sequential evaluation (PCI X first and then submission to a journal) is more likely to result in scientific improvement of the paper.
Yes, it is possible to recommend old articles. However, PCI X is a new system that focuses principally on the evaluation of new and unpublished preprints.
ARTICLE EVALUATION PROCESS
1. For preprints (i.e. articles neither published nor under consideration by a scientific journal) deposited in open archives online: One of the authors requests preprint recommendation from PCI X and can suggest recommenders. If a PCI X recommender is interested by the preprint, he/she initiates the evaluation process. This process is very similar to the process of evaluation for publication in a journal, and includes at least two high-quality peer reviews. Based on these reviews, the recommender decides to reject or to recommend the preprint as it stands, or asks the authors to revise their preprint. Several rounds of reviews may be required before a recommender decides to reject or recommend a preprint. When the recommender is satisfied by the reply and changes made by the authors, he/she can decide to recommend the preprint. This requires the recommender to write a “recommendation”, a short article, similar to a News & Views piece. This recommendation has its own title, contains between about 300 and 1500 words, describes the context, contains references (a reference to the preprint recommended at the very least) and explains why the preprint is particularly interesting. The limitations of the preprint may also be discussed. Once validated by the Managing Board, the “recommendation” and all the editorial correspondence (reviews, your decisions, authors’ replies) are then published by PCI X.
2. For postprints (i.e. articles already reviewed and published in scientific journals): Two recommenders of PCI X consider a postprint as particularly interesting and worth recommending to PCI X. They cowrite a “recommendation”, which is a short article, similar to a News & Views piece. It has its own title, contains between about 300 and 1500 words, describes the context, provides a link to the DOI of the article, contains references (a reference to the preprint recommended, at least) and explains why the postprint is particularly interesting. The limitations of the postprint may also be discussed. Once validated by the Managing Board, this “recommendation”, is published by PCI X.
3. The recommendation text, signed by two recommenders, provides a link to the DOI of the article and is published.
You must first deposit your preprint in an open archive. You then log onto the PCI X website or sign up if you do not yet have an account. You then click on the green button “Submit a preprint” and follow the procedure.
Any recommender of PCI X can perform this task, provided that he/she follows the code of ethical conduct of PCI X.
A “recommendation” is a short article, similar to a News & Views piece, written by one or several recommenders and describing why an article is particularly interesting. It has a DOI and is published in pdf and html formats in the electronic journal of PCI X. Each recommendation has its own title, contains between about 300 and 1500 words, includes references, describes the context and explains why the recommended article is particularly interesting. The limitations of the article may also be discussed.
TRANSPARENCY AND ETHICS
Bias, cronyism, retaliation and flattery are limited by i) the transparency of the reviews, which are freely available and may be signed, and ii) the transparency of recommendation texts, which are freely available and must be signed. In addition, PCI X has established a code of ethical conduct (no conflict of interest, no recommending of articles authored by recent coauthors and/or friends, etc.) to be followed by its recommenders. The Managing Board of PCI X performs quality control checks to ensure that these ethical standards are adhered to for reviews and recommendations.
All information leading to the recommendation of an article is made public: the name of the recommender responsible for recommending the article, his/her editorial decisions and recommendation text, the reviews and suggested corrections and the authors’ replies are available from the PCI X website, and the consecutive versions of the preprint are deposited in open archives. Only the name of the additional reviewers may be withheld.
No, only reviews and comments leading to the attribution of a recommendation (positive, but with criticisms and suggestions for improvement) are published. When a paper is rejected, the reviews and comments are sent to the authors but are not published.
Yes, all authors, whether or not they belong to PCI X, can submit their preprints to PCI X.
FATE AND CITATIONS OF RECOMMENDED ARTICLES
Each recommendation by PCI X has a DOI and can therefore be cited in your CV and in manuscripts. For example:
- Bravo IG (2017) Unmasking the delusive appearance of negative frequency-dependent selection. Peer Community in Evolutionary Biology, 100024. doi: 10.24072/pci.evolbiol.100024
When a preprint is recommended by PCI Evol Biol, you can cite it as follows, by indicating which version of the preprint has been peer-reviewed and recommended:
- Gallet R, Froissart R, Ravigné V. (2017) Things softly attained are long retained: Dissecting the impacts of selection regimes on polymorphism maintenance in experimental spatially heterogeneous environments. biorXiv, 100743, ver. 4 peer-reviewed by Peer Community In Evolutionary Biology. doi: 10.1101/100743
No, no editing (formatting) of the articles is carried out and articles are recommended without modification of their format. Unpublished versions of recommended preprint deposited in an open archive are not edited. We ask only that the authors of recommended preprints add a cover page to their preprint, together with a sentence at the beginning of their abstract stating that their preprint has been recommended by PCI X – e.g. see https://www.biorxiv.org/content/biorxiv/early/2017/09/28/141127.full.pdf.
Yes, recommended preprints are indexed. Google Scholar indexes all sorts of documents (articles, books, reports, etc.), including preprints deposited in repositories such as arXiv, bioRxiv, and Hal. These platforms therefore record preprint citations in the same way as they record citations of articles published in journals. An author’s profile in Google Scholar would therefore take into account recommended articles, whether the articles concerned are preprints from repositories or articles published in journals.
Most, if not all journals already accept the citation of articles not published in peer-reviewed journals (e.g. book chapters and reports). As the preprints recommended have been peer-reviewed, we see no reason why traditional journals would refuse to consider them valid.
More and more journals are accepting articles deposited as preprints in open archives for submission. See http://www.sherpa.ac.uk/romeo/index.php.
First, increasing numbers of journals are accepting articles deposited as preprints in open archives for submission. See http://www.sherpa.ac.uk/romeo/index.php.
Second, a preprint reviewed and recommended by PCI X is a preprint with a guaranteed level of quality. It is not a published paper. There is, therefore, no reason for a journal to refuse the submission of recommended preprints, quite the opposite in fact.
Third, concerning PCI Evol Biol, our list of recommenders includes numerous associate editors, editors and editors-in-chief of major evolutionary biology journals.
Fourth, concerning PCI Evol Biol, the Editors-in-Chief of Ecology Letters (Tim Coulson), Evolution (Mohamed Noor), Oikos (Dries Bonte), Evolutionary Ecology (Matthew Symonds), Evolutionary Applications (Louis Bernatchez), Molecular Ecology (Loren Rieseberg), Journal of Biogeography (Peter Linder), Frontiers of Biogeography (Joaquin Hortal), BMC Evol Biol (Christopher Foote), Genetica (Pierre Capy & Juan Bouzat) and Journal of Evolutionary Biology (Wolf Blanckenhorn) have indicated they will consider submissions of recommended preprints and that they may use PCI reviews and recommendations for their own review processes, if appropriate.
The recommendation refers only to the version of the preprint that has been recommended.
COMMENTING ON ARTICLE RECOMMENDATIONS
Yes, everyone, including authors and readers, can comment on recommendations. All comments are welcome, provided that they deal with the science, are signed and are respectful to the authors, the recommenders who made the recommendations and the other commentators. Comments considered as abusive can be notified to the Managing Board, which may decide to withdraw them.
Yes, everyone, including authors and readers, can comment on recommendations, comments, and the corresponding article. All comments are welcome, provided they deal with the science, are signed and respectful to the authors, the recommenders who made the recommendations and the other commentators. Replies to a comment not respecting these rules can be notified to the Managing Board, which may decide to withdraw them.
If a reader disagrees with a recommendation or with any comments on an article, he can write a comment. This comment will be published, provided that it is signed and is respectful to the authors, the recommenders who made the recommendations and the other commentators. Comments not respecting these rules can be notified to the Managing Board, which may decide to withdraw them.