News reached me today on Twitter that Peter Binfield, Publisher of PLoS ONE, is leaving PLoS to found a new company, Peerj.com.
Peerj will be an open access publishing company with a radical new premise: publishing in a scholarly journal can cost a heck of a lot less than the extortionate fees charged by PLoS. If you go to the Web site, you will see, among other things, this statement:
If we can set a goal to sequence the human genome for $100, then why can’t we do the same for academic publishing?
Of course, setting a goal and achieving a goal are not one and the same, but you get the idea.
So now we see the Achilles Heel of the author-pays business model for open access laid bare, and every shrewd archer of the OA movement will be pointing an arrow at it. Whether or not toll-access publishing is too expensive is one thing, something we can argue about; but author-pays OA is definitely too expensive. After all, an author-pays service is essentially a hosting service with a PR department and some Web 2.0 tools. I think they’re great, but let’s not pretend that they are something more than this.
I mentioned that I became aware of Binfield’s move on Twitter because it says something about the kind of company that Peerj plans to be. Peerj will play in the pool, deep or shallow, of the social Web. Among its plans are to raise money on Kickstarter, a new source of start-up financing. This will be a company that pushes the metaphor of information technology as far as it can, automating whatever can be automated, and much more besides. This means there will be a small number of talented software developers and not much more (but don’t forget the evangelist for social media). That means a low cost basis, an essential part of the strategy to drive down costs. Low costs translate into lower author fees. We have seen this strategy before. (See Wal-Mart.)
So the OA price wars have begun. What will this mean to the new eLife, the OA ventures of Wiley and Sage, and so many others? And let’s not forget the massively overpriced PLoS ONE itself. At some point, even Peerj’s proposition of lifetime publishing services for $100 may come to seem expensive.
39 Thoughts on "The Open Access Price Wars Have Begun"
This reminds me of the ongoing debate about how ebooks are being overpriced when they retail for over $10 because, of course, it is so cheap to publish books online. So we now get competition even among self-publishers to price as low as even 99 cents. There is much irony in the rhetoric of OA advocates coming back to bite them in the butt. We’ll see how many OA journals can afford to carry on while charging authors very low fees because online publishing is so inexpensive. Of course some will, by hiding all sorts of subsidies they receive from sources like universities.
The expectation that OA publishing will enter an all-out price war ignores the well-known maxim “you get what you pay for”. Anybody who thinks that PLoS One is expensive ($ 1,350 publication fee per article) might want to take a look at the new Cell Press journal Cell Reports, which charges a publication fee of $ 5,000 per article, and has apparently rejected ~94% of its submissions to date (http://news.cell.com/cellreports/cell-reports/being-selective). The modern era in science is dominated by two trends: (1) an incessant flood of scientific publications; and (2) increasing competition for research funds and academic jobs. This combination makes journal branding more and more important for authors who must show that their publications stand out from the crowd in order to succeed in a very difficult competitive environment. PeerJ may well succeed financially and may provide a useful service, but do not be surprised if authors continue to line up to pay much higher prices at more selective journals – they have very good reasons to do so.
Mike, you’ve really succinctly analyzed this market and its drivers. In a satirical piece I wrote for April Fool’s 2010, I introduced a new low-priced, no-frill OA publishing service for 99 cents. Strip publishing down to its basics, and all one requires is a place to put a public file. For most academics, however, this will not be enough. As the barriers to being “published” continue to fall, scholars will need to distinguish themselves in the marketplace and branding will become even more important.
I am skeptical that “branding” will be the main way that scholars distinguish themselves. I hope we are better than to judge people by the label of the clothes they wear, or the label of the journal their work appears in. In the end, scholars distinguish themselves by scholarship; venue is a very secondary consideration.
… except that in the context of scholarly publishing (which is the context for Mike’s and Phil’s comments) “branding” means something very different from “the label of the clothes they wear.” It means having the quality of one’s work recognized and given the stamp of approval by a trusted scholarly authority with an established track record for discriminating between high- and low-quality publications. Such “branding” has always been important to scholars and provides a not-bad way for students to make rough-cut discriminations between more and less worthwhile sources.
Not really, no. I have certainly never credited an article with more or less authority on the basis of whether it was published by Elsevier, Springer or PLoS. I might be more inclined to let myself be influenced by who peer-reviewed it, but the same pool of people are available to peer-review a given article whatever journal it’s in. Bottom line, we need to be judging research by the quality of the research itself, no extrinsic factors.
I have certainly never credited an article with more or less authority on the basis of whether it was published by Elsevier, Springer or PLoS.
The brand in question isn’t the publisher, but the journal itself. You may never have credited an article with more or less authority on the basis of the journal in which it’s published, but I can assure you that many of your academic colleagues (particularly in tenure-track positions) pay quite a bit of attention to that kind of branding, and that students are encouraged to pay attention to it as well, for the reasons I cited. None of this is to say that publication in a highly-respected journal is always an ironclad guarantee of quality — only that it is one useful metric, and that we’re talking about something very different from designer jeans.
Yes, scholars distinguish themselves by scholarship, but what if the amount of scholarship out there is beyond the ability of other scholars to read and assimilate? One of the more distinguished colleagues at my institute once remarked to me that if he would actually try to read all the papers published in his field, who would not have time to eat, sleep or go to the toilet, let alone run his own lab’. In this situation he reads all the papers from his field published in a select list of journals that he deems worthy of his attention, plus a highly filtered selection of the rest. In an ideal world, publication venue would be an irrelevant consideration. Unfortunately in the real world of the working scientist, publication venue is a critically important consideration.
In this situation he reads all the papers from his field published in a select list of journals that he deems worthy of his attention, plus a highly filtered selection of the rest.
I have to say I think your distinguished colleague is using an outdated strategy. There was a time no doubt when filtering by journal was the best or even the only practical way to choose a subset of the literature to read. But those days are behind us and receding fast. Anyone who’s moderately well connected in their discipline now, via email lists, blogs, Twitter, Google+ or other means, has a much more effective distributed network for guiding him to relevant new work.
I know I have made this point before but which journal a paper is in a way, way down the list of things that I look for in deciding what to read, and for the majority of papers I’ve read I wouldn’t even be able to tell you what journal they were in. These days, research comes in a cloud with the “branding” trailing along behind. No doubt some here have different practices; all I can tell you is that I am very far from alone.
Then why did I lose books by junior scholars to presses like Cambridge when the quality of the list I built at Penn State was fully comparable? The perceived “prestige” of the imprint in scholarly book publishing still has a major influence on where authors publish.
I see that you work in paleontology. How many papers you have to filter in your field per year? My ‘outdated’ colleague works in a field that generates ~1,000 papers indexed in PubMed PER MONTH…
I rest my case.
Now the only way I can see them making this work, is to further monetise the byproducts of the publishing process. That means datamining and selling the insights, probably with an advertising model built in as well. They’ll datamine the content and they’ll datamine the interactions of the users of the service I think. They may well do what facebook does and look to mine all the (scholarly) interactions that occur when a user of the service moves across other sites.
On the basis of their karma page (http://peerj.com/karma/), which encourages people to earn “peer points”, I expect they plan to sell authors additional services: e.g. express turnaround times, HTML article version, promotion in subject-specific bundles, hosting of supplementary materials, etc. For your $99 membership you’ll get something very basic; spend more on an individual article (cash or peer points, earned through reviewing and promoting articles) and the article will come with bells and whistles. So I guess they would be monetizing the individual components of the publishing process. Idle speculation on my part, of course.
Joseph, you say: “After all, an author-pays service is essentially a hosting service with a PR department and some Web 2.0 tools. I think they’re great, but let’s not pretend that they are something more than this.”
If you add ‘peer-review facilitation and organisation’, I would agree with you.
But the question I have is: how exactly does that differ from a traditional toll-access publisher?
There’s some much WTF in this post, I don’t know where to start.
“Whether or not toll-access publishing is too expensive is one thing, something we can argue about; but author-pays OA is definitely too expensive”.
As in, prices which give >30% benefit are reasonable, but prices which all “reasonable” publishers said were insufficient to break even are “extorsion”?
And what’s this with the Achilles’ heel?
Scoop #1: competition is good, monopoly is bad. When the author pays, there’s competition because I have a choice of journals which are to some extent fungible. When the library pays, there’s a monopoly because once an article is published it is non fungible, and we need access.
Scoop #2: scientific publishing is to benefit scientists and funders (ultimately the public), not to benefit publishers. Thus driving costs down is not a problem, it’s an advantage.
What you see as Achilles’ heel, most of us see as a big huge advantage.
Final WTF: are you insinuating that it’s a problem or a default to be using twitter and other social web tools? Are you doing this on a blog?
Joe, while your post is thought-provoking, I take issue by calling PLoS ONE “massively overpriced” and PLoS charging “extortionate fees.” PLoS holds no monopoly in publishing, generally, nor in open access publishing, specifically. Competition allows for choice in the marketplace, and this is something most of us would consider to be a good thing. If authors (or their funders / libraries) are willing to pay PLoS fees, then they are no more overpriced than any other service out there.
I refuse to believe the authors who publish in my journal are, to use a Dillbert term, clueless weasels who don’t realize they can be published cheaper elsewhere. They come to us because we give them solid peer review and editing which improves their manuscript, an audience of regular and loyal readers, and reputable branding (and Impact Factor) for their tenure committees. It’s a free market, and you indeed get what you pay for.
What comes to mind for me immediately is the difference between cost-based pricing and value-based pricing. Reader-pays publications have a much easier time selling on a value-based model — it’s not perfect, there can be a lot of pushback if the seller prices too high, but ultimately demand-based, value-based pricing can be higher. The value-based model works because each piece of content or branded journal or book by a certain author is unique. You can’t easily exchange Harry Potter for Twilight or Malcolm Gladwell.
Author services like OA journals are much more susceptible to being priced at or near cost, because ultimately the service is a commodity. As noted above, even selling premium services is still selling at the margins of a commodity, so you need volume and high throughput for it to work, and pricing will still feel downward pressure as automation squeezes everything toward a commodity.
This was bound to happen. When you create a model that sells only services and gives content away for free, you only have services to price. And they tend to be drawn into cost-based competition.
The other interesting thing here is the introduction of what is essentially an OA subscription model for individuals — that is, pay $99 for life (lifetime subscription). If left there, the model would fail. But if this is just a low-priced entry point for what is essentially a freemium model, then the premium part could work as well or better than other OA business models. The challenge will be volume-based, which is also frequency-based (how often can one subscriber publish a paper?).
This should prove a fascinating experiment. How exactly does the $99 fee work? Must all authors on a paper pay to join, or only one?
Regardless, if it results in a move to a market where price is the determining factor for where one publishes, then doesn’t this favor the bigger publishers? Those with already existing infrastructure don’t have building costs, but more importantly, benefit from economies of scale, which means they are likely to be able to provide the services here at a lower cost than a smaller independent publisher. Does this race to the bottom (line) mean an end to university presses, independent publishers, and society publishers in favor of the big corporations?
You’re right, this is a “scale and consolidation” model. The issue for established, scaled, and consolidated publishing businesses is that they’ve built, scaled, and consolidated something other than what Peerj might do.
How different is it though? If Peerj offers something compelling that authors pay for in droves, can a bigger company offer those same services at a lower price? Is there that big of a difference between the copyediting or peer review process that either can offer? Would an Amazon strategy work here for a big commercial publsiher, selling at cost/a loss long enough to drive the competition out of business?
I don’t have a problem with open access journals becoming as cheap as possible. I do have a problem with the troubles that will arise when the least-successful journals go belly-up. What will happen to the articles published in those journals when they go bankrupt? Sure, many journals are preserved in LOCKSS and other archives so their content won’t vanish completely (according to their assurances). But who will be responsible for continuing to provide failed journals’ content to the public on easy-to-navigate websites? Who will be responsible for making sure libraries link to the preserved content correctly? If failed journals’ content becomes less accessible (which seems likely) then the research community and the public loses. The research community loses accessibility to the information in the articles published in those failed journals. The public loses its return on taxpayer-funded research investment, as published results are not disseminated as effectively as they ought to. Natural selection among open access journals must result in some journals failing, and those failures will hurt science.
I’m not too worried about that. One of the important (but little-heralded) advantages of OA is that anyone can archive — and people do. For example, PLoS articles are available in torrents as well as from the PLoS websites and BMC; famously, Ross Mounce habitually carries around a complete PLoS archive on a USB thumb drive. Basically, once something is released into the wild unencumbered by restrictions, it will replicate.
Beyond access, there are all sorts of questions about maintaining and updating the older literature. If it’s being presented and sold by an existing company, there’s motivation for bringing things up to date, just as many existing journals have scanned and made older issues electronically available.
But if that old content is from a defunct publisher, or for some reason must only be available for free, then what happens when standards change? As we move into a world of semantic technologies and machine reading, will the current version of an older article suffice? Or will those old articles need to be moved to new platforms, converted to new file formats, have new types of metadata added, etc. If so, who pays for that if the end product is not allowed to be sold? Who can provide the rights to make these sort of changes to material from a defunct publisher for articles still under copyright with no obvious copyright holder identifiable?
Lots of good questions here. Only the last is easily answerable: “Who can provide the rights to make these sort of changes to material from a defunct publisher for articles still under copyright with no obvious copyright holder identifiable?” So long as the content is under a suitably permissive licence (i.e. anything conforming to the Budapest OA definition), then you don’t need the copyright holder’s permission.
But that does nothing for the last 100 years or so worth of literature which is all under copyright, so it’s not a really useful answer at all.
It does make for an interesting question for the OA publisher’s business model–if you are paid when you publish the article, is there an economic incentive to maintain/improve the article later down the line? If you’re selling access to that article, the answer is yes, but if you only get paid for publishing new articles, then devoting funds to the back catalog becomes a more difficult decision.
You are right that retro-fitting helpful licences to older literature is a separate (and difficult) problem — I was considering only new articles, where the production process is paid for by an APC. In general (unless you know different?) digitisation of old articles is usually paid for by the journal, in the expectation that they will recover those costs by selling access. (For what it’s worth, although I think paywalls in this case are still harmful, I do concede that they are morally perfectly reasonable.) Perhaps going forward we will see offers from public bodies to buy batches of older articles out of copyright?
This is not an open access issue but an electronic versus paper publication issue. Subscription publishes have, and in all likelihood will continue to occasionally go belly up as well. As far as taxpayer funded research investment, it my not be the primary purpose but it is an additional advantage of the NIH mandate for archiving the manuscripts they fund in PubMed Central.
Not much new here, and it is difficult to see how this “publish for cheap” approach is going to attract any good advanced science for publication in their journals.
To compare apples to apples, Hindawi open access journals had their prices much lower than open access PLoS Biology ($1,000 vs $3,000) for long time. But most scientists would prefer to publish in PLoS Biology because of this journal’s “brand name” as established by their high level of selection, reputation, impact factor, etc…
Since publication is the ultimate indicator of their work, when choosing where to publish their research, scientists care much more about the prestige and reputation of the journal than about article charges or business model employed (open access vs. standard access). So, a publication in Nature or Science or Cell remains a wet dream of every scientist in biology or medicine, doesn’t matter what are publication charges in these journals.
In terms of economics, the cost of a publishable project in a university research lab typically comes to $100,000s of the grant money and, more importantly, a few years of the scientist’s life. No scientist is going to risk the value obtained by their hard work and publish a good result in a less prestigious journal to save a $100 or $1,000 (of the grant money).
The current edition of the New Yorker has a very good, long-form article (by subscription) on Clayton Christensen that, I think, may pertain to this discussion–and add some counter perspective to Joe’s post.
The New Yorker website also had a live chat today with Larissa MacFarquhar, the author of the article. Money Q&A:
Q: Which businesses today do you think are most primed for disruption?
A: (Larissa MacFarquhar): I wouldn’t presume to predict specific failures, but I believe that Christensen would say there are certain indicators to watch out for. A company that has abandoned its high-volume, low-margin products in order to focus on manufacturing higher-margin, more expensive products is poised to be disrupted by start-ups for whom the lower margins are attractive in a way that they no longer are to the big company. Focusing on products rather than companies, Christensen’s attention tends to be drawn to products that seem too expensive and too fancy for most consumers–that are only of interest to customers at the very top of the market. Then again, it was this focus that led him to predict that the iPhone would not succeed: he saw it as an overly-fancy phone, not, as he came to see it later, as a disrupter of the laptop.
It’s not a “price war”, it’s a “disaggregation war”.