My article on open access in the New Scientist provoked an email from copy-editor Miranda Potter. Starting from the article’s mention of my recent paper in PLoS ONE, she raises the question of who is going to pay for copy-editing services if the drive to OA is accompanied by downward pressure on price (the $1350 fee I paid to the journal did not include any provision for copy-editing).
This re-ignites some of the arguments about quality that were discussed in my post on PLoS ONE, but it is an important topic so I was glad that she agreed to allow me to post her email here. Please feel free to respond in the comments.
From: Miranda Potter
Subject: Open access publishing“Did part of the fee that you paid to the open access journal go to a scientific copy-editor, so that you received a high-quality proof for comment and correction? Regardless of whether papers go through the review process, I have yet to see a paper that could not be improved upon by the services of a native English-speaking, highly qualified copy-editor.
The reviewers have many tasks and cannot possibly be expected to copy-edit every poorly written article. These roles are very different; the reviewer is a specialist in the subject area, whereas the copy-editor is a generalist who spends their time reading many scientific manuscripts, sometimes in very different fields. The reviewer should have sufficient in-depth knowledge to be able to provide constructive critcism of the scientific issues presented and help the authors to ensure that their article is high quality.
In contrast, the copy-editor ensures that the final article conforms to journal style, which is a more mundane, technical role. More importantly, the copy-editor reads every word of each manuscript and helps the authors by ensuring that any little glitches that the reviewers don’t have time to spot are picked up before the article appears in press, thereby preventing embarassing errata. You will find that for the journals with higher impact factors, the copy-editors will check that every panel of every figure matches its description both in the main text and in the figure legends, right down to the last asterisk. In fact, you would be amazed at how many meritorious authors cannot even manage to spell the names and addresses of their colleagues correctly.
I’m sure you’re aware that a paper written by a native Spanish speaker is likely to use a very different English linguistic style from one written by a native Polish or a Japanese speaker. If left without copy-editing, such papers might be very difficult for the worldwide readership to understand. I’m afraid there is a reason why scientific writing tends to be rather dry and structured in a certain way; it facilitates general comprehension. Isn’t that the most important thing?
If journals/authors opt out of paying for copy-editing services, I fear that science published directly on the Internet may degenerate into international gobbledegook and lose credibility. Who will stand up for the scientific copy-editors of the world?”
That is a really interesting point. I know nothing really about academic publishing, but I do know about commercial publishing. If you look at the difference between a self-published book and one from a ‘proper’ publishing house, the differences that stand out are design, layout, editing and copy editing/proof reading. (I make the distinction, because there are distinct roles for an editor, primarily about readibility of the content, and copy editor/proof reader, primarily concerned with style/errors/typos etc.) These are all hugely valuable in making the book more readable, and I would hope that journals do provide most of these services, which do cost money.
You may argue that an academic paper doesn’t need the polish of a commercial book, but surely it would benefit from being readable and with minimum errors. Perhaps the way to go about this would be invert the model. Instead of readers paying to read the journal, perhaps the academic institution the author works for should pay for the editing etc. to ensure that output that is ascribed to their institution reaches a minimum standard?
‘”You hear it said that the public has paid for it, the public should have access to it for free,” said Campbell. “At that point, I draw back and say, what do you mean by ‘it’? You mean something that has got quality in it, that has been selected and copy-edited. Somebody has got to do all that stuff, so the ‘it’ you want usually has involved somebody acting like a publisher to make it useable and comprehensible.”‘
[‘Open access to research is inevitable, says Nature editor-in-chief’ in Friday 8th’s Guardian.]
Obviously, it’s the value added that counts. Pushing that thought a bit further: what if traditional publishers continue as before, but also offer truncated/rough/raw representations of papers for free to satisfy calls for OA?
Basically we’re talking about a PubMed/whatever record ++: better-structured (so this isn’t about sharing draft papers), also adding ‘aims’, ‘findings’ and so on, semantically tagged as far as possible, but nonetheless a version like starvation rations compared to the full [paid] one with developed arguments, flowing text, nice figures etc..
David Shotton at Oxford has such a draft prescription for short descriptions of papers[*], which (and afaik we’ve both had this thought independently) are a sitter for a first draft of the truncated (largely author-submitted, ‘raw’, free) version of a paper described above.
Overall I think something like this, while requiring a little effort, would be rather beneficial for all. We seem to be heading this way anyway, with more content like reference lists and textual corpora (though far from what the OA crowd demand, which is increasingly part of funding conditions — Wellcome, NIH, etc.). Access to linked data increases paper citations and visibility will increase anyway as the Web2.0 stuff gradually gets going, which is all good for increasing impact and ad impressions, so we get OA and publishers still get money.
Basically, it’s the ‘freemium’ business model: barebones for free; flesh for a fee.
[*] Try repeating “Shotton’s Short Research Reports” after a few beers…
Chris, you’re basically describing Green Open Access.
And that is much closer to the final, published version of the paper than you’re envisaging. You talk about “a version like starvation rations compared to the full [paid] one with developed arguments, flowing text, nice figures etc.”, but it’s no part of an academic editor’s job to help the author develop arguments. It’s the author’s job to do all of that, and 99 times out of 100 to create the illustrations, too. The editor’s job in academic publishing is much less an artistic one, and more to do with scientific vetting and administration.
95% of the value that gets added during the publication process is the peer-review. And that of course is provided by other authors (on a volunteer basis) rather than by the journal itself. On that basis, if you could give me manuscripts that have been through peer-review but have otherwise not benefitted from the journal’s editorial process, then that would be everything I needed.
Hiya,
Yes arguments, figures etc. come from the author — the bulk of the value to which the journal adds (more or less) and on which it trades.
The short report, available on gold terms probably would just be ‘the facts’, also entered by the author (for example, for ‘aims’ consider the requirement from some journals for an author-generated strap line for an article; same thing for findings etc.). The journal handles it, but as with the main piece the content would come from the author. Yes it’s another hoop through which to jump, but as data collection and management improve it will come more naturally, and most importantly it gives us the chance to hang (most likely public whatever the state of the paper) data off an open and reliable thing (the short record serving as an anchor with its own DOI [and itself carrying the DOI of the full piece, inter alia]).
So it requires nothing but handling by the journal (though each can of course choose to curate/edit). The problem (apart from getting it going in the first place) is really the preprint one — can such a thing be so unattractive (while remaining useful) that journals won’t see it as competing with the full fat cash-generating version. (Though some evidence suggests that’s beneficial anyway).
C.
I have to say that, at first sight, I’m not sure what this proposal adds to the green OA route, which has the advantage of not involving any significant additional effort on the part of the author. But maybe I’ve misunderstood?
But it’s gold OA not green (i.e., would be done through publishers, and ideally libre at that), but with a summary version of the work. Nothing is being kept in anyone’s institutional shed here — we’d be exploiting existing publishing and search infrastructure and ensuring that even pay-only papers can have their data shared and contextualised. It’d let everyone carry on more or less as they are without breaking any rules or requirements. And then there’s all the semantic web thing, but that’s a bit more out there.
Well, quite. Copy-editing is an essential but often undervalued/misunderstood/ignored part of publishing, and yet another of those things that Open-Access enthusaists would probably rather we didn’t think about.
But do you think the above model, where you (e.g., NPG) charge only for the value when added, providing the raw content OA, is viable? I can’t see the world changing that quickly so compromise seems appealing. I know NPG are recruiting a ‘data editor’ — something like this would complement that nicely (as they’ll presumably be collecting metadata in a more structured way anyway).
Henry – You forgot to add ‘present company excepted.’ 😉
(You can have that copy-edit for free!)
I offer two perspectives on this — one as a reader, and one as an author.
First, there’s this: in reading PLoS ONE papers, I’ve not been aware of any deficiency in the style or correctness of the English, beyond what I routinely see in my discipline’s more traditional journals (Journal of Vertebrate Paleontology, Paleobiology, etc.) I don’t know whether that just means that the established journals are also not using copy-editors, or whether it means that PLoS ONE authors are very careful, or whether those that need copy-editing are paying for it before submission. But one way or another, that’s what I am seeing. (I should note that I am a very pedantic reader and do tend to pick up on minor errors.)
Second, speaking as an author, I’ve never been aware, in any of my published papers, of my language having been improved by copy-editing. (Of course that doesn’t mean it hasn’t happened — I could have missed the changes.) I have on occasion seen new errors introduced by copy-editing, or by some other step in the production process. Again, it might just mean that palaeo journals don’t use copy-editors — I don’t know. But my guess is just that it’s because I am a careful writer and reviser. Of course, others’ mileage may vary.
My conclusion from all this is that it’s sometimes a waste of time and money for journals to send articles through a copy-editing phase. I prefer the decoupled approach taken by PLoS ONE: that if you need copy-editing (for example because you’re not a native English speaker), you can buy that service yourself; and if not, you needn’t.
There’s a role for copy-editing, for sure — just as there’s a role for scientific artists. But people who don’t need those services are ill-served by journals that force authors to buy them (by increasing their own costs and passing them on either to subscribers or to Gold-OA authors.)
As journals move online, edits proposed by readers should be able to replace a dedicated copy editor – or at a minimum make them 100x more productive since they just have to approve or deny rather than reading everything. Without a print deadline it doesn’t have to be perfect at day one and most academics are natural pedants anyway…
I have seen something similar proposed before (not sure where) — that a published online version should be up for comments and editable for the first month or so (by the author) before being locked down as the version of record. I wouldn’t support wikipedia-style eternal editing…
I’d have to agree with Mike Taylor here: as a reader, I have never noticed a difference between supposedly copy-edited papers and those without. In fact, if anything, the papers that are the absolute worst to read are those in the journals with the ‘digest’ articles: If these papers are not directly in my field, I’m basically lost after the third sentence in the abstract, so I don’t really read any of these journals any more, other than for skimming the News-and-Views articles. Everything else published there is pretty much useless, if it’s not directly in my field, because it is so ridiculously condensed. Either it’s me or maybe the should employ undergraduates or high-school students as copy-editors?
As an author, I’ve noticed a copyeditor twice: once the person marked a bunch of sentences in my review article and asked me to rephrase them (i.e., more work for me, but potentially helpful). The other time (Elsevier journal), they introduced numerous errors in the conversion from my manuscript. Upon inspection, it turned out they didn’t use my txt file, but rather my PDF file, which meant that where there was a line-break in the PDF a ‘space’ was missing in the proofs. Unfortunately, not all of these errors were fixed in the final paper and are still in the version of record.
I don’t think a copyeditor was involved in our Science paper or any of the (few) other ‘hi-impact’ papers of ours, at least not that I could tell.
So, from my perspective, we can drop copyediting from scientific communication as it already is essentially absent.
Or we introduce it in a way that’s actually noticeable and worth paying for.
Maybe you’re just a gifted writer Bjoern… but I believe Miranda when she says that many manuscripts have passed through her hands and benefitted from the process. However, I can’t really speak to the scale of the problem. In any case, I would favour shifting the responsibility to the authors to sort out copy-editing, as a mechanism to keep APCs down and as an incentive to authors to write clearly.
On the other hand, most academics aren’t wikipedian basement-dwellers with plenty of liberty for crawling other people’s work, especially when there’s no obvious return on the effort.
I should perhaps admit that I’m writing from my basement. It’s my own basement, though. Not my mom’s.
I’m with Mike and Bjoern on this.
I can certainly see how good copy editing is useful (even essential) for authors who don’t speak English as their native language, or who have other reasons to need careful proof reading. But – frankly – I have generally found the whole process of copy editing quite unnecessary, even annoying. In my papers it has mostly just added errors (sometimes not even clearly flagged in the proofs), which is frustrating to say the least – especially when they sneak through to publication.
That said, we did recently have a very positive experience with the copy editors at Psychological Science, who made a lot of intelligent suggestions for rephrasing certain parts of a complex paper. Their standard of copy editing was a cut above the rest I’ve experienced, e.g. through Elsevier.
The PloS One approach definitely seems the right one to take for now. Ideally we should try and move to a form of OA in which the entire publication process is free of additional costs. We already get free peer review and editing. So – just a thought – but perhaps a small fraction of research grant income or HEFC funding could be put aside so that every department employs a professional copy editor (much like many departments used to employ graphic artists), trained to proof read English (where necessary) and format accepted manuscripts into a simple journal style.
I’ve posted some thoughts on this over on my blog but the TL;DR is that I feel that these costs should be pushed to the author, that most articles don’t need it, as they are either already good enough, will be unread or are uninteresting, and that ultimately discussing copy editing cost is a bit of a red herring when thinking about the cost of production overall, when there is a chance that we could see costs being reduced by an order of magnitude within the next few years if projects like PeerJ or scholastica can gain acceptance as venues of publication.
http://partiallyattended.com/2012/07/12/the-cost-of-production/
Most articles “will be unread or are uninteresting” is sadly true.
I have to agree with most of what you said, especially about how most articles will be unread. It’s sad to see how most people don’t understand how much a copy editor is needed!
Thanks for the post 🙂