Join 3,523 readers in helping fund MetaFilter (Hide)


The best of the web - that'll be $30, please
March 29, 2013 5:37 AM   Subscribe

Open access: The true cost of science publishing
posted by Gyan (45 comments total) 19 users marked this as a favorite

 
They say that their commercial operations are in fact quite efficient, so that if a switch to open-access publishing led scientists to drive down fees by choosing cheaper journals, it would undermine important values such as editorial quality.

Or it would put market pressures on journals to work within the economic reality rather than charging amazing rents to "add value" to value they receive for free (or, sometimes, less than free).

But Philip Campbell, editor-in-chief of Nature, estimates his journal's internal costs at £20,000–30,000 ($30,000–40,000) per paper.

That doesn't match the "quite efficient" quote above. Campbell is asking me to believe that it costs his journal more per article to process each article than it costs the universities to pay the faculty who write the things? (Most faculty publish more than two articles per year, after all, and they are also (at least theoretically) teaching and doing service.

All of this is also tied to the idea that "Open Access" mean "Gold Open Access." The "Green Open Access" model is tied to the idea that scholarly authors (and their institutions) have a right to make their material available as they see fit rather than giving it as captive production to commercial interests.

Have you been to a scholarly trade show? The big scientific publisher have a lot of money to throw around. While I would miss the occasional lavish dinner (and the somewhat less occasional useful tchotchke (really, how many tote bags do I need?), I would be happier if we could bring down the price of journals and make material that is often paid for by public funds (at least at second hand) more available.
posted by GenjiandProust at 5:57 AM on March 29, 2013 [8 favorites]


Oh man. This is a firestorm waiting to happen. Probably the single biggest debate among the scientific community as a whole. The issue is that the community is conservative: people publish to get jobs and get tenure. The judges of hiring and tenure are older, and maybe less likely to accept a PLoS publication than a J Neuro, for example. Plus, productive & top-tier researchers will always* have grant money and will continue to be able to publish in the traditionally-good journals. Everyone else may be pushed to open-access, but the top-flight stuff from prestigious labs will still be in Nature and Science, I think. (Again: conservatism; radical ideas from upstarts won't be in these journals.) This will change if libraries (especially smaller, lower-tier-university ones) revolt and stop getting subscriptions.

* The other thing that would cause a sea change is an across-the -board reduction in government funding. Fingers crossed that doesn't happen.

All speculation. I'm interested to see what will happen, especially every time I get a $3000 invoice from OUP or Elsevier.
posted by supercres at 6:04 AM on March 29, 2013 [4 favorites]


What stood out to me was the profit per article: $5,000 on average. This is money that is coming out of the budgets of universities whose libraries purchase subscriptions to bulk sets of non-open access journals. It really seems like there is something crazy going on when universities have to pay private companies for work that their own salaried employees create and facilitate. In my own field (linguistics), I almost never use paper research materials; most of the time I'm reading pdfs of articles which I either get via a university subscription online or from the authors themselves. We do have a good peer-reviewed journal, Semantics and Pragmatics, which uses an opensource publishing system (http://semprag.org/about/aboutThisPublishingSystem) and exists exclusively online. The cost to run is little, given that a huge part of the value they add is peer reviewing, something that scholars do for free anyway. I hope this is the wave of the future, but I imagine the big academic publishers do not want to lose their tight grip on this revenue stream.

One additional point: in the article, the publishers claim that they add editorial value. I've worked on a few manuscripts with large (and famous) academic publishers who were entirely unhelpful, wanted "camera-ready" documents, and were not even able to provide a consistent formatting policy. My impression of their contribution to the process was that they allowed their name to be on the spine, and the rest of the work was meant to be done by the authors and editors (the academics putting together the chapters, not the editors employed by the publishers, who might not even exists as far as I could tell). The only reason we had to go along with this was because you need to have "reputable" publications on your CV if you want to be successful in academia. When the shift can be made to viewing online-only open source publications as "reputable" by a critical mass in the field, we will hopefully see some kind of sea change.
posted by tractorfeed at 6:15 AM on March 29, 2013 [10 favorites]


tractorfeed, while I mostly agree with you, it's worth remembering that moving online does not actually save much money -- while it lets you get rid of a warehouse and some shipping people, editorial costs remain the same, and the tech costs you add pretty much eat up any savings, as I understand it. So electronic or not isn't as much of a factor as people assume, especially with the "online saves money" meme, which really needs to get excised from everyone's minds (especially administrators, who use it to hide the real message "we will give you one-time money to establish something, but upkeep and maintenance come out of your budget -- double amusement points when it's their initiative).
posted by GenjiandProust at 6:21 AM on March 29, 2013 [1 favorite]


> "What stood out to me was the profit per article: $5,000 on average."

For the sake of fairness, the linked piece did NOT say there was a $5,000 profit per article. It said there was $5,000 revenue per article, with the costs to produce each one being around $3,500 to $4,000, meaning a profit of $1,000 to $1,500 per article.
posted by kyrademon at 6:24 AM on March 29, 2013


However, the vast majority of authors don't self-archive their manuscripts unless prompted by university or funder mandates.

I really wish this weren't true, but I know professors who will just re-request their articles through ILL or through the university subscription packages rather than keep an off-print or pdf for themselves, because it's "easier." Has anyone found a good way to encourage deposition in university depositories, or have any of the professors here had a good experience with one?
posted by jetlagaddict at 6:35 AM on March 29, 2013 [1 favorite]


For the sake of fairness, the linked piece did NOT say there was a $5,000 profit per article.

Good point; I read too quickly, but still I think the profit per article is much higher than what authors would guess.

As to the online savings, my point was not so much that there is a cost to paper publishing (which I'm sure there is) but that the most important part of the entire process is already done for free. In the case of the Semantics and Pragmatics journal, they use unpaid reviewers just like every other academic journal I'm aware of, and the technology investment is covered by the sponsoring institutions, including two large research universities that already have a substantial tech infrastructure in place. I'm sure that MIT and UT Austin find that their sponsorship of S&P is a much better deal than paying a private company a large fee (hidden by non-disclosure agreements) for a set of journals, some of which they might not actually want.

Looking at the production flowchart for the journal I've been mentioning, it seems that most of the labor being done is already done for free for most journals, except possibly for the case of the copy editing, proofreading, and layout, although as I said many private academic journals expect "camera-ready" files meaning they are not providing this kind of work to begin with. And to be sure, the organization of the endeavor is no small amount of work, but again being the editor (the person who coordinates the logistics of handing out submissions to reviewers, getting the reviews back, corresponding with authors, etc) is often an unpaid job, even when the journal itself is published by a private company.
posted by tractorfeed at 7:00 AM on March 29, 2013 [1 favorite]


There are more efficient was of doing this. Many of the same universities with a decent number of tenured and tenure track faculty members already have university presses. If the organizations that are serious about open access want to change publishing, using our own existing programs might be a place to consider. Granted STEM publishing isn't the same as humanities publishing, which is what most university presses currently do, but enough of the infrastructure is there that if the parent institution and its library invested enough to get it started, they could easily adapt a new and related work flow. The primary differences between humanities journals publishing and STEM journals publishing involve editorial approach and scale, and building new programs would need to create processes to incorporate those differences. But there are efficiencies to be had by sharing infrastructure, in both digital and in print production. And by bringing it in-house, universities can reduce middlemen and the need for high margins. These things will have a cost, and what they require is investment, but in the end, money saved by the library will more than make up for the additional cost involved in a university publishing a significant amount of its own scholarship and research.

Other issues that would need to be addressed include the foolish reliance on publishing programs to generate the majority of the income of professional organizations like the AAA and the ACS, and the free-rider problem, where university presses currently subsidize the vetting of tenure publications, and the improvement of the reputation of colleges and universities that do not themselves have a publishing program. As for the professional societies' desire for revenue issue, I have no answer for that and the solution must come from the members themselves of those kinds of organizations. Until they take the governance of those organizations more seriously, this will continue. And as for the free-rider issue, either faculty at other institutions pay a premium author fee for publication in OA, or that college or university suck it up, and finally open their own publishing program.

FWIW.
posted by Toekneesan at 7:23 AM on March 29, 2013


I've attended meetings where the publisher buys champagne and hors d'oeuvres for several thousand scientists. They certainly have money to throw around to protect their business model.
posted by Philosopher Dirtbike at 7:27 AM on March 29, 2013 [4 favorites]


I just published a paper in mathematics. As far as I can tell, most (current) papers in mathematics are read off the arxiv preprint server; that was true for my research at least. Paper is for older work. The ironic thing is that in order to get my paper peer reviewed, I had to give up my copyright and take down the paper from the arxiv, making it much less likely to be read by anyone.

In mathematics, the whole issue could be solved (at least in the US) if the AMS officially sanctioned an editor/reviewer overlay to the arxiv. The value in publishing is in which editor selects it for "publication" i.e. says that it is noteworthy and which respected academic peer reviews it i.e. certifies it correct and probably refers it to peers. The whole idea of a journal is senseless with electronic distribution and the "open access" journal thing just doesn't work for young academics.
posted by ennui.bz at 7:43 AM on March 29, 2013 [1 favorite]


As tractorfeed noted, there is zero editorial value added by the publishers. An academic journal's editors and referees are academics paid by their university, not the journal.

There are technologically inept fields where people write their articles in Microsoft Word, which the publisher then pretties up by running through Adobe's typesetting software. These academics would simply learn more powerful features in Word if this "service" disappeared tomorrow. And fields like math, physics, comp. sci., afaik linguists, etc. avoid this problem by writing in LaTeX anyways.
posted by jeffburdges at 7:50 AM on March 29, 2013


Are there still math journals who protest papers appearing on arXiv, ennui.bz? I thought they'd all stopped that? If some still do, they need to face a boycott.
posted by jeffburdges at 7:52 AM on March 29, 2013


(reading more of TFA, the "episciences" thing being pushed by Gowers is an overlay of the arxiv... the problem is that "peer review" looks much more like Facebook than a journal. it doesn't work unless everyone is participating... and special software to standardize a "workflow" is just an obstacle)
posted by ennui.bz at 7:53 AM on March 29, 2013


many private academic journals expect "camera-ready" files meaning they are not providing this kind of work to begin with

If we want to be fair, this is probably the only real service they are providing appart from keeping the servers running and producing paper copies. From my own brief experience of putting together the proceedings for a handfull of smallish conferences in the sciences, I can tell you there is a 5% of authors that will consistently fuck up any kind of author instructions or template. The best ones will provide a file that looks passable or even correct on visual inspection of the printout, but is otherwise almost entirely useless.
posted by Dr Dracator at 7:54 AM on March 29, 2013


Are there still math journals who protest papers appearing on arXiv, ennui.bz? I thought they'd all stopped that? If some still do, they need to face a boycott.

I denounce myself: I am a young academic who just published in an Elsevier journal... the problem with a boycott is that it hurts people like me: publish or perish is a pretty strong force boot to the face.
posted by ennui.bz at 7:54 AM on March 29, 2013


But Elsevier doesn't require you to remove your submission from arxiv; at least that's their official stance. Can you give some more details?
posted by Mapes at 8:15 AM on March 29, 2013 [1 favorite]


But Elsevier doesn't require you to remove your submission from arxiv; at least that's their official stance. Can you give some more details?

hmmm... looks like I might not have to take it down from the arxiv after all... but I swear one of the web pages I had to click through proscribed preprint servers, though it looks like from reading the formal publishing agreement again (which I hadn't read carefully) that I don't...

so there's at least that.
posted by ennui.bz at 8:42 AM on March 29, 2013 [1 favorite]


I have to say I really enjoy being priced out of reading the science I was trained to do.

It saves me so much time!

I hope things don't change because then I would have no excuse for not being current.
posted by srboisvert at 8:43 AM on March 29, 2013


I denounce myself: I am a young academic who just published in an Elsevier journal... the problem with a boycott is that it hurts people like me: publish or perish is a pretty strong force boot to the face.

Seems to me academic journals are in an advantageous position. Academia, to some extent, is in the business of producing published papers. A product which is needed to gain credibility, advance a career, and ultimately solidify a position in a bourgeois social class of credentialed experts. And the quantity of PhD candidates and new academics globally must be growing rapidly. It's not surprising that publishers may be taking advantage of the situation, and as always taxpayers are probably the least protected. I imagine the peer review process is not free of nefarious motivations either.
posted by Golden Eternity at 8:47 AM on March 29, 2013


Seems to me academic journals are in an advantageous position.

They're also in a vulnerable position. Their most valuable asset is the continued cooperation of the senior academics who sit on journal boards and conference program committees. If these people revolt, as happened with Topology, they can take the journal's reputation with them and leave the publisher with nothing.

They are conservative. Not a lot of people are willing to stick their necks out for an unsure proposition. But once a few more of these rogue journals break their bonds and prove the viability of an open-access model, it might lead to a snowball effect. Fingers crossed.
posted by qxntpqbbbqxl at 9:04 AM on March 29, 2013 [4 favorites]


As to the online savings, my point was not so much that there is a cost to paper publishing (which I'm sure there is) but that the most important part of the entire process is already done for free.

Absolutely, and that is probably the most damning part of the whole process. It's silly to claim that journals do not add value -- a good web interface makes it much easier to find articles, for example, and that is not cheap -- but, since the value to which they are adding is free (or a cost center), it's a little disingenuous to claim that the costs are necessary (especially given the profits listed in the article -- even the low end are kind of high for a secondary service).

However, I do see a lot of "online is cheaper" rhetoric, which, as far as I can tell, is not true, especially over time. After all, the immense increase in journal prices (pretty much doubling every 7 years, substantially ahead of inflation) since 1970 pretty much paid for the publishers' electronic infrastructure.
posted by GenjiandProust at 9:10 AM on March 29, 2013


Jetlagaddict, my instuition is collating submissions for the REF (UK academic government funding related thing) through our research management system, which outputs to our repository. So if the academics want to be included in the REF submission (and most of them have to be), then stuff goes through to the repository automatically. (The repository also powers their publication lists on their faculty pages, which seems to also encourage them)

Of course, not that stops them doing thing like attempting to photocopy their book chapters or downloading a copy from our online subscription to upload but I have a long suffering colleague who checks for these things before they go through to the repository.
posted by halcyonday at 9:18 AM on March 29, 2013


Publishers certainly add value. But do they add enough value to justify their cost? A lot of people are saying no.

Some people here might be interested in reading this article by a colleague and friend of mine, Jason Priem, who also has an article in this special issue of Nature that we're discussing: Decoupling the scholarly journal. It does a nice job of explicating the functions of the journal, and offers opinions about what the new system might look like. Note: I helped edit this paper before it was published.

I also find it interesting that Nature - the journal this was published in - is closed access - but they do allow authors to choose an "open access" option. And it only costs $5000 - that is double what most other journals (that I am aware of, in my field) charge for the OA option. It's unclear to me whether the Comments pieces are always open to the public, or if just these ones are because they are about the future of publishing, or if they are available now because they are from the most current issue, but normally Nature does not allow just anyone to read the articles they publish. Interesting twist there.

Also pertinent: everyone on the editorial board of the Journal of Library Administration resigned today because Taylor and Francis, the journal's publisher, uses licensing terms that the board found too restrictive. The Board, they wrote, came to the conclusion that it is not possible to produce a quality journal under the current licensing terms offered by Taylor & Francis and chose to collectively resign. It will be interesting to see if others follow suit.
posted by k8lin at 9:56 AM on March 29, 2013 [1 favorite]


You are not paying for your article to be published, you're paying for the peer review. Which of necessity has to be done by a financially independent party.

I know it's frustrating for people without journal access but the alternative is NOT for research institutions to do their own peer review or to just publish unreviewed work. For starters gray literature is typically harder to access than white papers.
posted by fshgrl at 11:19 AM on March 29, 2013


USGS has a peer review process people here might be interested in checking out btw. It's pretty time and resources intensive tho.
posted by fshgrl at 11:22 AM on March 29, 2013


Wrong, you are not paying for your article to be refereed, fshgrl. Academic referees are not paid by the publisher. Editors who select and cajole referees are not paid by the publisher either. Yes, important stuff happens, but university employees do it all.

Author/institution pays style open access publishers improve over Elsevier, etc., but ultimately they represent merely a transitional phase to overlay journals run by governmental science foundations or donated university resources.
posted by jeffburdges at 11:33 AM on March 29, 2013 [3 favorites]


Ok, you are paying for someone to coordinate the peer review. Which is quite a lot of work, if you've ever done it.
posted by fshgrl at 11:41 AM on March 29, 2013


The person coordinating the peer review, usually the editor, is also an unpaid academic. It is definitely a nice thing to have on your CV, so it's not like it's entirely without benefit, but the point we're trying to make here is that the bulk of the work is volunteer.
posted by tractorfeed at 11:43 AM on March 29, 2013 [4 favorites]


Which academics in the process are paid varies by field and even by journal, I believe.

But it is certainly true that in most cases, most of this work is not paid for by the publishers.
posted by LobsterMitten at 11:45 AM on March 29, 2013 [1 favorite]


We can't actually say that all referees/readers aren't paid by the publisher. It varies by discipline. They definitely are in most of the humanities and social sciences. It also isn't true that publishers add no editorial value. In STEM publishing that may be generally true, but it isn't true at all in the humanities where university press' lists very much reflect the the interest of both acquisitions editors and editorial boards. It's also useful to remember there are two kinds of editorial value, one representing the interests and influences of the discipline and exercised through content acquisition, and the kind represented by those who work with the accepted author to improve the clarity and accuracy of the text. Good authors in any discipline are going to want that second value added to their work.

And yes, as fshgrl notes, it's the management of the process that the publishers currently offer, and it's no small feat. It should be noted though that there are many commercial and even some early open source software packages that have made that piece of the process a heck of a lot more efficient and less reliant on staff at a publisher.
posted by Toekneesan at 11:45 AM on March 29, 2013 [1 favorite]


In linguistics and philosophy and logic (the fields I'm most familiar with), reviewers and editors are not paid. I was unaware that reviewers got paid in other fields. I know a few people who are editors of academic journals and they are not paid, and they spend a huge amount of time working on the journal. I'd be curious to know which fields pay reviewers and how this affects the reviewing process.
posted by tractorfeed at 11:48 AM on March 29, 2013


You're paying the people who contact reviewers, answer queries, anonymize the papers etc. Support staff. Who need desks, computers, internet, heat, stamps, vacation pay etc....

I volunteer a lot of my time to stuff like this and to teaching (well my employer does). Even if you are running 100% volunteers on the professional side you still need some cold, hard cash.
posted by fshgrl at 11:55 AM on March 29, 2013


Look, no one in this thread is claiming that there are no costs associated with producing an issue of an academic journal. Obviously physical infrastructure is one of these areas; I don't see where anyone has claimed otherwise. In my experience in academia, the people who "contact reviewers, answer queries, anonymize the papers etc." are editors (or graduate students of same for some tasks), and are not paid. This might vary by field, but I know a number of academics in a large variety of fields and I have never heard of reviewers and/or editors being paid (other than in this thread). I have done all of these tasks myself (although for conferences, not journals) and I know that it is a substantial amount of work, and typically thankless.

Anyway, it's clear that cold hard cash is needed to produce a journal; this has not been disputed. However it does seem that the "value-added" which journals charge a princely sum for is the peer reviewing, which oftentimes is done for free. What could be an interesting topic of discussion is what are some ways to move forward which might free up this value without punishing younger academics who rely on the extant infrastructure to establish themselves in their field. It's fairly clear that if private publishers have the cold hard cash to pay for champagne-style frivolities in order to keep their pole position on publications then it must be very lucrative indeed. I think most academics would rather live and work in a world where the proceeds of their labor did not get funneled into the pockets of a corporation while their departments have to make drastic cuts, teaching suffers, facilities suffer, class sizes increase, etc. etc. along with all the other ills that currently plague higher education.
posted by tractorfeed at 1:50 PM on March 29, 2013 [4 favorites]


A few people are saying that the "online is cheaper / saves money" idea is false. Can someone explain that? Ignoring the fixed costs of organizing peer review, typesetting, etc., which will happen in either case, how is it not obviously cheaper to distribute an article electronically rather than physically?

In 2012, academic institutions made a total of about 13 million downloads from arXiv, which had expenses under $800,000. That's about six cents per article access. Is it possible to physically create and deliver limited-run print content to people for six cents per use? Especially when the majority of physical copies of any single paper will likely be read... what, once? Probably never for a large percentage of physical copies?

Obviously there are expenses for hosting content online, but I cannot imagine how they could even begin to compare to physically distributing the same content to everyone who reads it.
posted by whatnotever at 2:12 PM on March 29, 2013


A few people are saying that the "online is cheaper / saves money" idea is false. Can someone explain that? Ignoring the fixed costs of organizing peer review, typesetting, etc., which will happen in either case, how is it not obviously cheaper to distribute an article electronically rather than physically?

Because printing and distributing a journal, especially for large publishers that are already printing and distributing stuff, is not all that expensive. What you save in less (or even no) warehouse space, and workers packing and shipping, you make up in fewer but much more highly-paid jobs in tech support, interface design, etc. Additionally, you don't need to replace a press or a warehouse every few years, worry to the same degree about security, dealing with access issues, etc. I mean, there are probably some savings in the long run, but not as much as people seem to think.
posted by GenjiandProust at 2:33 PM on March 29, 2013 [1 favorite]


Academic publishing and scholarly dissemination are often based in a gift economy working in a reputation marketplace, but it does take management and it should sometimes produce things, like print. In other words it has costs and always will. That said, as almost everyone in this thread has identified, the price we're paying is way out of whack with its expenses. If we want to see this corrected, I think we need to DIY. Universities should work in partnership with academic societies and editorial boards to change the fundamentals of how scholarship is produced and disseminated. We need to consider how publishing benefits scholarship and tailor publishing programs to the needs of each discipline. We need to better leverage available technologies to ensure the widest and wisest dissemination strategies, and we need to remove commercial incentives from the process as much as possible. A big part of that will require reform in how we look at tenure and promotion particularly among faculty, professional societies and organizations, and institutional administrators. Expanding resources these institutions already have, between their libraries and their presses could go a long way to building the capacity for moving most of this content out of the for-profit realm, and back to the control of the academics who produce it and their associated institutions.

It is questionable to compare downloads to journal subscriptions. They don't indicate the same thing, not even close. And as for the savings that come with online dissemination, yes they are there, but they aren't as big as you might expect. Unit cost and distribution costs on a physical journal isn't even 10% of the cost. The cost is all in the editorial management and manuscript preparation process. For most scholarship that will continue to be significant. And to offer scholarship online exclusively ignores archiving and digital divide issues. It maybe be acceptable for a handful of disciplines, but it shouldn't be the only way scholarship is disseminated. I work for a publisher that publishes art history scholarship, and because color must be consistent in the presentation of the scholarship, print is much more dependable than most monitors at color fidelity. That may change someday, but for the time being, art history will need to occur primarily in print.

I agree that we can make this cheaper, but it might be worth looking at focusing on the societies and faculties rather than obsessing over the publishers. Create incentives for publishing open access and for moving away from commercial models and it will occur, but as long as departments value journals published by societies who choose commercial publishers, there will be a premium to pay which will be split between the society and the for-profit publisher.
posted by Toekneesan at 2:43 PM on March 29, 2013 [3 favorites]


Afaik, referees are never ever contacted, or even though about, by anyone receiving a publisher paycheck, fshgrl. All correspondence with referees is handled by academic editors who the publishers never pay.

There is just no reason for referees to ask publisher personnel any questions because the publisher's personnel actually don't know anything about the journal's content criteria. All they know about is layout style rules, sometime they don't even know about that.

Virtually all recent articles we actually read come from arXiv, meaning the publisher's staff never even saw it, much less touched it. Yes, arXiv versions incorporate the referee's corrections, but the publisher never saw the paper until well after that.
posted by jeffburdges at 2:47 PM on March 29, 2013


Because printing and distributing a journal, especially for large publishers that are already printing and distributing stuff, is not all that expensive.

That's what people have said, but I'd like to understand how the numbers actually compare. If arXiv costs six cents per download, how much does it cost per-use for a physically distributed copy?

It is questionable to compare downloads to journal subscriptions. They don't indicate the same thing, not even close.

Of course, and I wasn't comparing those. I'm hoping to compare cost per-download to cost per-use (i.e., reading a physical copy of one paper). I'm really just curious how the numbers work out, because it seems obvious to me that online is cheaper, while others are claiming that it isn't.

I mentioned the cost-per-use of a paper on arXiv, so let me try to estimate the cost-per-use of a printed journal.

This source cites an unnamed publisher with annual revenue of $16 million. Let's say 10% of that goes to printing, warehousing, and distribution (in the example, it's actually 20%), so $1.6 million. From Elsevier's library price list, the average price per issue over all of their journals is about $200. So perhaps we can say that the unnamed publisher ships 16M/200 = 80,000 physical issues per year (when most likely they don't, as revenue will come from electronic subscriptions as well).

80,000 physical copies printed, warehoused, and shipped for $1.6 million is $20 per copy. Yes, that's definitely cheap! But every physical issue would have to be used (i.e., an article from that copy read by someone) 333 times for the cost of producing and shipping the physical copies to match the cost of providing downloads from arXiv. I doubt that the average number of uses of a physical issue comes anywhere near that. Please do let me know if my estimates are horribly off somewhere.

It's a minor point, and I think it is mostly orthogonal to the primary issue at hand (that the value provided by publishers could be obtained much more cheaply in other ways), but with people saying online is not cheaper, I was (and am) very curious to understand how that could be.
posted by whatnotever at 3:49 PM on March 29, 2013


Because a paper is downloaded, that doesn't mean it's read. Just as because a physical journal is sent to a library and or a member of the society that published it, that also doesn't mean it's used, or that we can know for sure if the articles in it are read once or twenty times. Comparing metrics is a tricky business. I'm not sure you can extrapolate useful information from comparing cost per download to cost per number of print copies. And knowing what is actually read is another issue that I'm not sure any current metric claims to know. More useful metrics include things like the number of times a work is cited. That is also unfortunately a metric that is more likely to be corrupted in an open access environment than in print, or at least we have begun to see that in some of the journals on Beall's list. Those sorts of impact factors (citations) may be more useful in doing a cost benefit analysis of the various mediums, but I certainly hope popularity and cost efficiency aren't the most influential factors measuring the success of a dissemination effort. The focus should remain on, whatever the medium, whether the publication is best serving its discipline in its current form using its current dissemination model. It's not as simple as digital, cheap, and open are always good. They are not good for everything, or everywhere. Print can penetrate and serve some communities far more effectively than digital, and it's simply hubris to have absolutely no print copies of research in some library somewhere.
posted by Toekneesan at 5:37 PM on March 29, 2013 [2 favorites]


The person coordinating the peer review, usually the editor, is also an unpaid academic. - tractorfeed

Afaik, referees are never ever contacted, or even though about, by anyone receiving a publisher paycheck, fshgrl. All correspondence with referees is handled by academic editors who the publishers never pay. - jeffburdges

This isn't true for high-end journals like Nature and Science. Their editors are full-time professionals, recruited among people with impressive academic records. To maintain good relations with their professor-level reviewers, they can't overload them with manuscripts that don't have a good chance of getting published. Since they receive enormous amounts of contributions and want the very best, the pre-screening is a big job. The editors are also the ones who finally decide what gets published, based on the referee reports. It isn't fair to say that they don't add any editorial value.
posted by springload at 7:18 PM on March 29, 2013


As someone who works sort of publishing-adjacent, I always think the arguments against publishers are kind of strange. Big publishers rake in big dollars - whether they should or should not is kind of irrelevant. I have been to those champers and canape parties and I didn't exactly see anyone turn any of it down. It comes down to this: academics and librarians hate how much these things cost because of course they are expensive, come in bundled packages and with restrictive licenses. But publishers aren't charities and never have been. And yet, it's the academic system that is entirely in control of whether it's worth buying what publishers offer. Stop tenure decisions around these publications and you kill them dead. That's entirely in the hands of the very people doing all the complaining and no, it isn't easy, but it's way easier to change your own behaviour than it has to be to change that of an entirely separate and powerful entity like big publishing. You can yell at the Elseviers all you want and they'll just laugh at you while rolling in piles of dollars full-on Scrooge McDuck style (I have not been to that party but I don't doubt it exists). Because what they know is that you're just mad you bought into their bullshit and the second that you stop it it, their blood runs cold because its over. But you can't expect them to be magnanimous about it though. Change from within and you win - and until then it's just a waste of energy.
posted by marylynn at 10:25 PM on March 29, 2013 [1 favorite]


Alright, fair enough, springload. Yes, pre-screening is a big job anywhere, but most journal editors do it unpaid. Nature and Science sound anomalously large. We might self-select slightly more in math and physics too, not sure.

There are editorial bords seeking to negotiate better deals with their publishers, marylynn, but anyone demonizing the publishers simply wants them gone precisely by way of reforming academia itself. All the bitching about publishers is aimed at influential academics who make hiring decisions and whose articles will improve non-commercial journals.
posted by jeffburdges at 2:16 AM on March 30, 2013 [1 favorite]


I thought this issue was the very reason the internet was created. Software engineers have been using Version Control and Request for Comments (RFC) mechanisms for decades. Some use TeamForge, GitHub, or bitbucket. (previously) and markdown.

With a solid identity mechanism, to ensure everyone actually are who they say they are, peer review and comments would be easy to track, and 'nearly free' depending upon the repository, compared to the big sci-paper publishers.
posted by Monkey0nCrack at 7:37 AM on March 31, 2013


But publishers aren't charities and never have been. And yet, it's the academic system that is entirely in control of whether it's worth buying what publishers offer. Stop tenure decisions around these publications and you kill them dead. -posted by marylynn

While this is true, it is the equivalent of telling sheeple not to bank at BOA or Chase, or use Wal-Mart or Target. This isn't going to happen - cheesy-poofs are just too expensive at the mom&pop, and my local credit union's online bill pay sucks and they don't have an iPhone app.
posted by Monkey0nCrack at 7:46 AM on March 31, 2013


There are plenty of academics who simply don't need journals owned by publishing houses, Monkey0nCrack, anyone with tenue and a reasonable track record for grants, or any really truly brilliant paper.

In mathematics, universities or mathematical societies own well over half the general journals. Yes, some society journals cost too much, but that money does some good, well math societies aren't afaik wasteful. Yes, some university journals cost too much, but if the math department doesn't get that money, or has the budget cut accordingly, well that's the department being stupid. Just catch the university president spending it on hookers and blow or something.

We roughly consider general journals as better than specialist journals because work there represents something of wider interest, usually meaning better mathematics. So, if your theorem is good enough, you should simply write the article well enough that a general journal takes it. If however your result is good but not quite good enough, then you might need to publish in a specialist journal. In math, too many top specialty journals fall under publisher ownership. If you're successful enough, you could simply restrict your selection in specialty journals, depending upon your general journal better articles to maintain your career. You'd encounter a problem if all your papers really belong in specialty journals though.

Credit unions are usually way better than banks, but obviously exceptions exist. Is there really only one local credit union though? Walmart and Target are really hard to avoid if you don't earn much money, but time preference is their real catch, not just price. I usually avoid buying from Walmart not by overpaying at poorly stocked family stores but by adjusting my time preference to order online directly form China, India, etc.
posted by jeffburdges at 9:41 AM on March 31, 2013


« Older They let us get probably a thousand times more rad...  |  Last month, Robbie Rogers came... Newer »


This thread has been archived and is closed to new comments