Publishing in the New Millennium: A Forum on Publishing in the Biosciences: topic
This is a public forum
Current state of science publishing
Anna Kushnir
Tuesday, 16 October 2007 19:02 UTC
Are you satisfied with the current scientific publishing process?
-
Replies
-
That’s a very general question. The answers will depend on people’s own experiences of it. My own was very modest, before I left research and became a science editor. Since I have been a science editor (many years), I’ve received a lot of positive feedback from published authors, and a lot of negative feedback from authors whose work we could not offer to publish. Published authors tend to think that their work has been improved by the peer-review and publication process, but it is understandable that those who aren’t published are not so happy about it. However, my colleagues and I have received positive feedback about constructive decisions not to publish work, as they have helped people to improve their papers for publication elsewhere. So some good certainly can come out of a journal not being able to publish a paper, sometimes.
-
I think there is still a mystique to the act (art)of getting ones worked published that, despite public forums such as here at NN,will lead people to be a bit disatisfied. Specifically, while the peer-review process does a very good job of quality control people forget about the final “gate-keeper” role of the editor. That is people can make the assumption that just because the reviewers liked the article it will get published in that journal. Now correct me if I’m wrong, but the role of the editor is to decide what should be reviewed and then, after review, decide if the final product is really what is in the best interest of the journal. Unfortunately, it is this role of the editor that some mentors forget to educate their students about, let alone keep in mind themselves.
-
That is an excellent point, Craig. So far as the most competitive journals are concerned, what you write is certainly true. Sadly, papers are turned away by Nature on a regular basis because although referees have found them to be technically sound, they do not meet the editorial criteria. I think this process is not so marked in specialist journals, where competition for publication is less fierce, but it certainly can and does happen.
As, in the Nature journals’ case certainly, and I am sure for other highly competitive journals, editorial criteria are published in a freely available form on the journals’ websites, it would indeed make sense for mentors and teachers of relevant courses to include this advice as part of the standard process. -
The peer review process is one that remains opaque for many young scientists, like myself. The utility of the review process seems clear. While it may do a “very good job” as Craig Rowell writes—it is not self-evident whether this is best system one can conceive of. At a minimum , it does allow intelligent scientists the opportunity to dissect an article and provide critical (both positive and negative) feedback to the authors. Three questions, however, quickly arise:
1) Why is this process not blind to the reviewer? A double-blind process, although not perfect, might remove bias that enters due to lab pedigree, university-affiliation, etc. If the author hopes to remain anonymous, it might also cut down on the number of self-citations which litters the landscape!
2) How can you reduce the average time to publication? In many instances, the time between first submission and final publication can be several months. In some cases, this might be warranted. In others, it might not. There are several factors that clearly go into this measurement, but not many people would object to a more streamlined, faster process for submitting (and re-submitting) articles to the same (or different) journals. This leads to a corollary—why not have a common application with similar formatting guidelines for the first stage of review?
3) Can 3-4 reviewers accurately represent the views of a large community of scientists? The history of peer-review suggests that it was borne out of a close-knit community keeping tabs on each other—criticizing and helping each other along the way. With the number of scientists exploding, can we assume that the 3-4 reviewers that are chosen by a journal can accurately assess an article and represent the “field” well? What alternate mechanisms might do a better job?
These are just some questions that I have—more as a young interested observer—rather than a hardened critic. I would love to hear everyone’s thoughts on this topic!
-
I think you will find answers to your questions, and a good discussion, by visiting the “Ask the Editor” forum, where we have a section and discussion on peer-review and the publishing process. Also, Nature journals publish advice about peer review (how to do it, benefits of the system, and so on), and Nature last year ran a freely accessible peer review debate (archived on Peer to Peer blog). All these are linked at the Ask the Editor forum, so if you go there in the first instance, that will be a good start to finding out the answers to the questions you ask.
best wishes
Maxine. -
Kishore, here are a couple of more specific links for answers to some of your questions. Maxine gives a good response here about double-blinding peer review. Read this whole thread as a couple of other Nature editors have also weighed in. There’s good discussion related to your first question about bias.
And the peer review debate Maxine was referring to can be found here.
-
I realise I didn’t respond to questions (2) and (3).
On (2), reducing publication times, we can offer advance online publication as soon as the paper is ready for publication, so it does not have to wait until the next available slot in a printed edition. Obviously the web has helped enormously in other ways to reduce publication times, eg manuscripts are submitted online, can be peer-reviewed online and proofs can be sent by email or via online upload—all cutting out lots of time incurred in postal costs from previous years. Authors can also use web collaborative tools in creating their manuscripsts and in responding to the journal’s requests for revision during the peer-review process.
On (3), no, 3-4 reviewers can’t represent a large community of scientists in this sense, nor should they. As explained in our peer-review page (see link in my earlier response), the editor is seeking the very few people in the world who are uniquely technically qualified to assess that particular manuscript. There are going to be only a few scientists who can do that, in most cases.
The job of the journal editor, aided by the peer-reviewer, is to help the author to communicate that advance to a very wide audience, without hype and without underselling it. But before we can get to that stage, we need to find the very few specialists in the same subdisciplines who can independently assess the technical correctness of the work.
I hope this helps! -
To add to Maxine’s responses: here’s the discussion thread in our Ask the Editor forum that addresses Kishore’s question about a common application/submission form (someone asked the same question in that forum).
And this comes via Maxine: on Nature’s peer review policy page, there is a section called “The peer-review system” (scroll down about 3/4 of the way down the page), which goes into some of the issues you raise about alternate mechanisms. The editors at Nature have reviewed and experimented with peer review over the last few years and the results of that review are discussed there. Hope this helps.
-
Thank you so much for your comments and reading suggestions (and my apologies for the late reply but I was consumed with our conference that just wrapped up on Friday!)
From my initial reading, it seems that the question of double-blinded peer review is often cast aside for practical (i.e. “we will know who the author is based on the reference lists”) rather than ideological reasons. Although it is likely that a referee might be able to narrow down the likely authors to a handful—or in many cases just 1 or 2 possible labs—that doesn’t preclude that in certain cases it might actually shield the author from the referee.
To me, it seems that a combination of methods—peer review before publication followed by a journal-mediated commenting process would really help figure out which papers are in fact the ones that end up having the most impact.
On this question of impact factor, Dr. Varmus said at our conference that what we really would want is an “impact” factor for each article, not just the journal. With changing technology and the ability to record “hits” and “comments” along with citations—we might slowly see such a system evolve. Thus, the peer review system might turn out to be just the first part in a long journey of assessment.
-
On your last point, you can use Web of Science (ISI), Scopus (Elsevier) or Google Scholar (free) to look at “impact” of individual articles in this way—how many times the paper has been cited or downloaded, for example. You can also do this in services like PubMed but they are limited to particular disciplines (PubMed to biomedical sciences, for example).
You can go to this page of our author and reviewers’ website for direct links to these, and other, abstracting and indexing services.
-