[Image: Jean-François Millet: Le Vanneur. Source: WikiPedia]
I have been writing a few posts lately about open peer review in scientific publishing (Open Review of Scientific Literature, Openness And Anonymity in Peer Review, Third-party review platforms). As I have mentioned, quite a few platforms experimenting with open post-publication peer review have been appearing around us recently.
Now it seems there is an actual journal on its way, embracing open review and open access from the very beginning to an extent I have not seen yet. It sounds like a very brave and exciting initiative. According to their own description it is going to be a journal for all disciplines of science. You can read more about the ideas behind the journal on their blog: The Winnower. It was also featured recently here.
Curious about this new journal as I am, I have been talking to its founder, Josh Nicholson, online on a few occasions lately to find out more about the journal. I have decided to publish this Q&A correspondance here in case others are interested.
Q&A with Josh Nicholson
2013/10/04 – on Google+:
As I understand, you will publish manuscripts immediately and publish the accompanying reviews of them when ready. Will these manuscripts be open to review by anyone, will you find reviewers, or a combination thereof?
In principle, it would be “most open” to allow reviews by anyone, but specifically when some paper is not “popular” enough to attract reviewers spontaneously, I guess it might also be necessary to actively engage reviewers? If so, do you consider somehow paying (monetarily or otherwise) reviewers?
The papers will indeed be open to review by anyone. We want it to be completely transparent and open. We also wish to be completely clear that papers without reviews linked to them have not been reviewed and should be viewed accordingly. We would like to engage reviewers with different incentives in the future and will explore the best ways to do that as we move forward. Our system will in essence be quite similar to “pre-prints” where authors are allowed to solicit reviews and anyone is allowed to review but it will all occur in the open. We will charge $100 per publication so that we can sustain the site without relying in grants. We would love to hear more of your feedback should you have any!
I have been considering for some time how an open peer review system can attract reviewers and possibly encourage them to identify themselves to “earn” reputation.
The Stackexhange network, among others, seems to be quite popular and it seems to me that one of the things driving users to contribute is the reputation system where a reputation score becomes the “currency” of the site. Users can vote other users’ questions and answers up or down. This lets other users quickly assess which questions and answers are “good”. Votes earn the poster of the question or answer reputation points and this encourages posters to make an effort to write good questions and answers.
It seems to me that such a system could be used more or less directly on a peer review platform. It would both encourage users to write reviews and let other users assess and score reviews (review of reviews).
We agree with you 100%. We would even like to offer the “best” reviewers, as judged by the community, free publishing rights. Ultimately we would also like to make the reviews citeable. Some of these features will not be present in the initial launch but will be expanded upon and rolled out over time. We hope you will consider submitting in 2014 and reviewing!
We have a few other select features that will be present in the initial build to attract reviews. Some of these will be discussed in future blog posts.
2013/10/06 – in blog comment:
Have you at The Winnower considered if you could make use of third-party reviewer platforms for your publishing?
We have briefly communicated with LIBRE and are indeed open to reviews from third-party platforms. We are happy to work with anyone towards the goal of making reviewing more transparent.
2013/10/11 – on Twitter:
Will you have any sort of editorial endorsement of papers you publish or will the open reviews be the only “stamp of approval”?
Open reviews will serve as ‘stamp of approval.’ We hope papers will accumulate many reviews.
Papers can be organized based by content as well as various metrics including most reviewed etc.
2013/10/11 – in blog comment:
I am very excited about your new journal – that’s why I keep asking all sorts of questions about it here and there ;-)
In terms of archiving papers, what will you do to ensure that the papers you have published do not disappear in the event that the Winnower should be out of business? Do you have any mutual archival agreements with other journals or institutional repositories?
We are happy you are excited about The Winnower. Please keep the questions and comments coming!
We are currently looking at what is the best way to preserve papers published in The Winnower should The Winnower not survive. We are looking to participate in CLOCKSS but have not made any agreements as of yet.
Another one: under which terms are you going to license the published manuscripts? For example, I have heard authors express concern about third-party commercial reuse of papers without consent under CC-BY. I am not sure yet what to think about that.
Content published with The Winnower will be licensed under a CC BY license. Commercial reuse of work, as we understand it, must cite the original work. We want to open the exchange ideas and information.
2013/10/18 – in blog comment:
Here goes another one of my questions: will your platform employ versioning of manuscripts?
I imagine that authors of a paper may want to revise their paper in response to relevant review comments. Just like it often happens in traditional pre-publication review – here we just get the whole story out in the open. If so, I think there should be a mechanism in place to keep track of different versions of the paper – all of which should remain open to readers. As a consequence of this, there will also be a need to keep track of which version of a paper specific comments relate to.
Rating: will it be possible to rate/score papers in addition to reviewing/commenting? While a simple score may seem a crude measure I think there is a possibility that it could help readers sift more efficiently through the posted papers. In a publishing model like yours, it is going to be harder for, e.g. funding agencies or hiring committees to assess an author’s work, because they cannot simply judge it by where it was published (that may be the wrong way to do it anyway, but that is not what I am aiming to discuss here). A simple score might make the transition to your proposed publishing process “easier” for some stakeholders. I am a bit reluctant about it myself, but in order not to make it too superficial, maybe scoring/rating should only be possible after having provided a proper review comment. This should make it difficult for readers to score the paper without making a proper effort in assessing the paper.
We are happy to have your questions! There will indeed be an option to revise manuscripts after a MS has collected reviews. We are however a bit uneasy about hosting multiple versions of the paper as we think it may become quite confusing. We are happy to explore this option in the future but currently we believe that the comments along with the responses should be sufficient to inform the reader what was changed.
Do you agree?
Our reviews will be structured meaning that there will be prompts which allow different aspects of the papers to be rated.
So you plan to allow revision, but previous versions are “lost”?
I see the point about the possible confusion, but what if a commenter points out some flaw about some details in the paper, the author acknowledges it and revises the paper? Now, future readers can no longer see in the paper what that comment was about. Well, they can see the comment, but they can no longer see for themselves in the actual paper what the flawed part originally said.
Could the platform perhaps always display the most recent version of the paper but show links to previous versions somewhere along with the metadata, abstract etc. that I assume you will be displaying on a paper’s “landing page”? The actual PDF of an outdated version of a paper could have a prominently displayed text or “stamp” saying that this is an outdated version kept for the record and that a newer version is available?
Perhaps links to older versions of a paper would only be visible in a specific comment that refers to an earlier version of the paper?
These are good points. We will discuss some of these and see what approach will work best and what we are capable of. If it is not too much confusion and not too much to implement this into the build this could be quite useful as you point out. Thanks!
By the way, I think it would be interesting if it were possible not only to comment on papers, but to annotate the actual text in-line. I think it would be great if readers could mark up parts of text and write comments directly next to them. Would it seem too draft-like?
I am not sure how this could be done, technically, but it seems like the technology http://hypothes.is/ are brewing could enable something like this.
We agree that inline comments are quite interesting and we have this as a possible tool to build in the platform in the future. We however have limited funding for the initial build and want to focus on features that are critical first and complimentary second. But this is a great idea and definitely something we will be exploring in the future.
Good point. It is best to get the essential features working well first.