Editor’s Note: Today’s post is by John Sack, Founding Director at HighWire and originator of the MECA collaboration.

A new academic publishing initiative known as Manuscript Exchange Common Approach (MECA) was accepted by members of the National Information Standards Organization (NISO) in May as a framework for best-practices development in manuscript transfer across systems. The idea behind the project is that the industry’s leading technology providers will work together on a more standardized approach to the transfer of manuscripts between and among systems, such as those in use by publishers and preprint servers.

Documents moving one folder to another

Currently, the way authors have to retype and reformat to submit and re-submit manuscripts to different publishers is counterproductive. It’s not just the authors that are losing out, reviewers are affected too — it’s estimated that around 15 million hours of researcher time is consumed each year, simply repeating reviews. These are the two issues that MECA seeks to address. What’s referred to as ‘publishing’s nasty secret’ could be partly solved if journals and publishers were able to transfer manuscripts between publications that are using different submission-tracking systems, without having to write new programs for each new pair-wise system exchange.

A lot of recent industry conversations have revolved around bottlenecks faced by researchers seeking to access existing knowledge — what I have called “Friction in the workflow” in a talk at APE and in a recent article — but there is still work to do on streamlining how articles are published in the first place. Too much time is wasted on the manuscript transfer and submission processes. For that reason, my company — HighWire — is thrilled to be working alongside Aries, eJournalPress, Clarivate, PLOS and now NISO on a solution that we hope will benefit everyone in the industry. As technology organizations, we are working to provide capabilities and tools that can reduce technology as a barrier.

MECA is now at the point where we have a working system functioning to transfer submissions and editorial material among about 15 journals, from three publishers, using two different manuscript management systems. Each of the collaborators has its own MECA-support timetable.  Joel Plotkin, CEO of eJournalPress noted that eJournalPress was part of the first pilot to transfer manuscripts between two different vendors — HighWire and eJP — and three publishers — Cold Spring Harbor Laboratory, Rockefeller University Press, and EMBO — as a part of the submission and peer review process. Joel said that for eJournalPress, “Going forward, MECA will be used for all future integrations / manuscript transfers, where possible”. Chris Heid, Product Lead for ScholarOne at Clarivate Analytics said, “we are proud of our early partnership in the project and we look forward to contributing as part of the new NISO working group. MECA’s goal of facilitating the exchange of manuscripts and peer review metadata across publishers and platforms aligns with our publisher-neutral values and we look forward to incorporating these emerging standards into the ScholarOne Manuscripts platform.”

As a group, we felt it was essential to get into operation — to create a proof of service. It was one of the group’s principles that we would do what we can, now, in practical terms, rather than discuss options that might not be feasible for years. The support of NISO, which will provide a neutral forum for MECA to be discussed and refined, will be crucial to enable this collaboration to expand to a larger group of stakeholders. It’s important that we hear the voices and concerns of all interested parties in order to continually refine MECA’s standards to ensure industry relevance.

Obviously industry-wide collaborations such as MECA come with a whole set of challenges that need to be overcome. With the many different stakeholders and interested parties involved, achieving agreement on a set of standards is complex, but we feel that we have made important first steps. There is already one fully-operational implementation of MECA in production, and this will serve as a base for documentation and elaboration through the NISO review and approval process.

In order to be of value to all parties in the ecosystem, there are different scenarios that need to be considered. There are four use cases we set our sights on in scoping MECA:

  1. Transfer manuscripts and reviews among journals
  2. Transfer papers between preprint servers and journals
  3. Import papers from authoring systems/tools to submission systems
  4. Export papers from submission/publishing systems to other services (e.g., repositories, compositors, etc.)

In our work, we focused on seven areas where standardization was necessary to make MECA work:

  • Vocabulary: providing a standard nomenclature
  • Packaging: a simple, flexible, standard way to assemble files
  • Tagging: being able to pass submission information from system to system
  • Peer review: being able to pass review information from system to system
  • Transfer: enabling the transfer of information from system to system
  • Identity: a unique, consistent identity for document instances across systems
  • Transmission: a simple, consistent way to send the information across systems

Craig Jurney, one of HighWire’s architects working on MECA, compared MECA to the process of introducing standardized containers to the shipping industry, which enabled all sorts of efficiencies.

There are some friction points around the editorial process and practices.  Our hope is that MECA will eliminate technology as a limitation so publishers and researchers can choose to address editorial concerns. NISO’s best practices approach could choose to address these friction points: manuscript formatting, triage “unmasking”, and review portability.

Manuscript formatting:

What we’re not yet trying to do is to introduce standards that would simplify formatting across different journals. As Charlie Rapple points out in her Scholarly Kitchen article from August last year, this is a pain point for researchers. Specific formats might be used by journals to retain their own identities. Another reason is to improve efficiency for editors, reviewers and production staff who have to handle many different manuscripts and would find it a burden to go hunt for the same information in different places and different formats. Some journals have taken an approach to this problem called “format-neutral submissions” where they encourage initial submissions in any format, requiring adherence to journal-specific standards only when an article has been accepted for publication. Authors are pleased by this and an increasing number of journals are adopting this practice.

Triage “unmasking”:

In terms of what MECA is trying to achieve, there are some voices that question whether authors would want to transparently submit a paper to a journal after that paper had previously been rejected by another journal. The author might — quite understandably — not want the second journal to know that it wasn’t the first choice, or who the prior rejection was from. This is an entirely valid concern. We’ve spoken to authors and publishers about this, and it is our aim to ensure that all submission decisions using MECA give editorial groups the ability to make decisions that aren’t limited. MECA can help overcome any technological hurdles that authors face when re-submitting work.

Review portability:

Another concern that has been raised concerns the process of transferring peer reviews of a manuscript between different journals. Authors might not want to do this for the reasons outlined above, and there are further questions about how issues of subjectivity and bias might be recognized and addressed. Different journals have different scopes and review criteria, after all, so transferring a review from one journal to another might not make sense.

If there can be a level of agreement among author, editor and reviewer about whether reviews can or should be transferred, this would come under the remit of a set of editorial best practices outside of MECA. MECA’s role is to ensure that if this is a desirable outcome, it can be achieved without technological hindrance. We hope that best-practices for editorial coordination emerge rapidly, but it is more likely that for a while, a period of experimentation with various options will occur. Evidence from different approaches should be gathered and shared, informally now, and more formally at the next Peer Review Congress.

That brings me to the big picture: what we are trying to do with MECA is minimize the friction caused by the technologies used in the submission and review process.

When the technology is no longer the source of friction, we can move to focus on the process itself, with new degrees of freedom – about how we want the actual system to work. From there, journals can differentiate to take advantage of authors, editors, reviewers, time or talent.

For more information about MECA, see the website and/or write to info@manuscriptexchange.org.

John Sack

John Sack. John is the Founding Director of HighWire, which facilitates the digital dissemination of more than 3000 journals, books, reference works, and proceedings. John is also the Co-Director of the International Congress on Peer Review and Scientific Publication.

Discussion

10 Thoughts on "Guest Post: Manuscript Exchange — What MECA Can Do for the Academic Publishing World — And What it Can’t"

To me the big question here is whether there is room for an author to change/update a manuscript before it is transferred. If the article is rejected, it is usually for a reason. We want authors to listen to peer reviewers, and take their suggestions for improvement to heart. If the article is to be directly transferred from Journal X to Journal Y after rejection, then when do authors get that opportunity?

Assuming the author has addressed reviewer concerns, then it is a different article than was originally reviewed, so why would you want those old, no longer relevant reviews to accompany this new version?

It has long been my experience that authors want each journal to think they were the author’s first choice, and that they want new reviewers to look at the revised work with fresh eyes, rather than being biased by reviews done on a previous version. I’m concerned that this will result in low uptake for MECA services, just as we’ve seen for manuscript transfer consortia that have been put together in the past.

Excellent point David. Additionally, why would a Publisher not only reveal reviewer names but provide those reports and names to another Publisher to use next time around (peer review costs money!)? With respect to data protection issues, you would have to get permission from reviewers to share their reports in this way, creating extra work for the rejecting publisher and further delays — plus reviewers may say no to passing on their reports. I am already helping the author through peer review and providing reviewer reports,hopefully the author will consider those reports and resubmit again after further work. What is the incentive for me to help another publisher get that improved work instead of it coming back to the publisher who has put in all the (preliminary) work?

MECA enables all documents associated with an article to move across system boundaries in a common way. But it does not require that that happen. Which documents move would be the area of business practice that should be layered on top of MECA.

The number of articles rejected without review is substantial enough that even before you ask whether reviews be transferred, you should have a substantial benefit to the authors.

But the question of what practices — do you ask the authors about transferring reviews? do you ask the reviewers? do you send the reviewers’ names? do you send the decision letter(s)? the authors’ responses? — are ones that right now are being decided within the boundaries of single manuscript systems and publishers (i.e., a single publisher might be moving some of all of these documents now within their single manuscript system). MECA enables this across system boundaries.

I have heard, anecdotally only, from medium sized publishers who are still evolving their transfer practices intra-publisher. For example, do you consult with another journal’s editor before offering transfer? do you obtain a promise of any sort (e.g., “it will go out for review” or “you will get a decision in xx days” or such) from the receiving journal to give to the authors? Undoubtedly some practices lead to higher transfer-offer-accepted numbers.

Does anyone have a list of several manuscript transfer consortia? The only one I know of is the Neuroscience Consortium. The new journal, “Life Science Alliance” — from Rockefeller, Cold Spring Harbor, and EMBO — is not a “consortium” of that sort I would say. And is it our first implementation of MECA since these three publishers use two different manuscript systems for transfers. But as noted there are several additional (not transfer) use cases for MECA and we have some of those underway.

Axios did one iteration of acting as a third party review transfer service – some authors approached us to say that their paper had been reviewed quite favourably at journal X, but rejected for lack of novelty. They wanted those reviews to be carried forward to journal Y, preferably with the reviewer identities from X, so that the Editor at Y could be sure they were from competent reviewers.

We contacted the Chief Editor at X, who agreed to pass along the original decision letter (to rule out any possibility of the authors editing it), then contacted the reviewers to see whether they were OK for journal X to release their identities to Axios and ultimately to journal Y.

The reviewers all agreed, and we passed the whole decision package (original decision letter + reviewer ID’s, the authors’ response to those reviews, and the revised manuscript) to journal Y. They accepted it without further external review. It was important for the authors to revise the article, as that demonstrated to journal Y that the reviewers’ concerns could be dealt with.

The only rate limiting step was manually obtaining the reviewers’ permission to release the identities – if a journal was part of a consortium doing this on a regular basis the review request email should just tell the potential reviewer that their ID might be passed to another journal via a third party. We had a very similar clause in the Axios review request letter and it never seemed to create any issues.

I really like the idea of MECA, although I don’t think it will be used much to transfer articles from one publisher to another for the non-technical reasons already mentioned. Some publishers use several different systems internally, including multiple editorial platforms, invoicing platforms, collaborations with preprint servers, production vendors, etc, etc. A standard way to link these together would be a step forward. There’s probably not a great deal of direct benefit for authors, but it could save a lot of time and effort for publishers.

Personal data seems to be an issue here. Info about authors, reviewers and editors is potentially being transferred across entities (business and legal). Not unmanageable but nor ignorable,

And at some point there is likely to be an outlet that just receives rejected content it will publish with low threshold at the expense of those ‘higher up’ the food chain. How are costs to be shared equtiably? I’m guessing very little will transfer upwards…

Notwithstanding the challenges that previous attempts at manuscript transfer have faced, waterfalling within a publisher’s own family is something that does happen and is presented as a feature/author benefit so even tech improves in that would be useful (I’m probably not the only person to experience different workflows happening for different journals within a given publisher’s family). Anyway … I continue to imagine a service sold to university research offices that manages manuscript placement … MECA would be very beneficial.

I think this has already been done: https://xkcd.com/927/

Of course, that’s *slightly* tongue in cheek, but I’ve never had to re-type or re-format a manuscript to send it to a different publisher, because this is a problem that’s already been solved (at least, within my field).

It’s not just the authors that are losing out, reviewers are affected too — it’s estimated that around 15 million hours of researcher time is consumed each year, simply repeating reviews.

John, I’m sorry to jump on this statement, but it gets repeated over and over again by every by every commercial entity that wants to disrupt/transform scholarly publishing. This passing phrase treats peer review like a commodity (the price of soy beans) or a check-box (yes/no). If more eyes on a manuscript results in better science–an argument routinely used to support preprints–then sending a manuscript through another round of review shouldn’t be considered a waste of precious time.

I have no doubt that MECA will save some clerical time on the part of the author. It also appears to create other clerical and administrative hurdles, as highlighted in the comment section above. Make your argument on the basis of clerical/administrative efficiencies and please stop undermining the value of peer review.

A good author is going to respond to reviewer comments, creating a new manuscript that’s different than the original one that was reviewed. Is it now a waste of time to review that new manuscript, or should we rely on the old reviews of something that may be entirely different? While there’s some redundancy, the old reviews are no longer fully relevant, even before we figure in the context of the journal for which the manuscript is being reviewed.

As Phil notes, there’s a frustratingly common notion that we can magically streamline the processes of science. We see the same thing for negative research results. Sometimes an experiment fails because the concept is wrong, sometimes it fails because of experimenter error. How can you tell the difference between the two? You can’t, and that’s why we need some level of redundancy in science. If we only allow one try at any hypothesis, we’re going to miss out on a lot of discoveries because people make mistakes.

Redundancy is a strength of the system, not something to be eliminated.

Comments are closed.