Old two pan balance
Old two pan balance (Photo credit: Wikipedia)

One of the strengths of the Scholarly Kitchen is that it’s a group blog, written by individuals who don’t always agree. This is something that is often misunderstood, and you’ll regularly find us arguing with one another in the comments section of each article. Usually we leave it at that, but something in a recent article by Kent Anderson struck me as both inaccurate and unfair, and that problematic remark ended up obscuring an important point.

In Review: “Free Ride: How Digital Parasites are Destroying the Culture Business, and How the Culture Business Can Fight Back”, Kent made the following point:

. . . companies like BioMed Central, PLoS, and PeerJ have emerged in our space. . . . Most significantly, none of these three entities currently cycles financial benefits back into the academic or research community.

And again:

BioMed Central, PLoS, F1000, Bentham, and dozens of others — return no revenues to the scholarly table

First, in the case of BioMed Central, the statements are inaccurate. While for-profit and owned by Springer, BMC does indeed publish journals that are owned by research societies and research institutions. Some of the profits from those journals are returned to the research community through those groups, as is the case for any other society journal publisher.

The statements are unfair in the case of PLoS. The Public Library of Science is a not-for-profit organization. Like all not-for-profits, PLoS must actually turn a profit to ensure its survival. Not-for-profits need excess funds in order to grow, to develop new offerings and technologies and to put aside in case of major upheaval or market shifts. PLoS has only recently had its first profitable year in 2010 (2011 numbers have not yet been released).

It’s unfair to demand that PLoS return excess funds to the community when it has only begun to generate those excess funds. Given the level of development and experimentation at PLoS, it’s unlikely that there’s much left over from those 2010 profits. They have likely been reinvested in PLoS itself, serving the community indirectly through the creation of new communication tools, things like article level metrics, open APIs, and whatever else PLoS has up its sleeve.

You can’t attack a company for doing something it hasn’t yet done. If at some point in the future, as PLoS continues on its upward trajectory, it starts hording cash beyond what it needs to continue to develop, grow, and survive, then one could make this sort of accusation. But that hasn’t happened yet, and given the nature of PLoS, I suspect that if profits reach that level, they will be returned to the community through philanthropic acts, donations, lowered APCs, etc. Keep in mind also that PLoS is facing a new wave of competition from eLife, PeerJ, F1000 Research, and others moving directly into their market space. Having a war chest of funds may be essential to their survival as market conditions may radically change.

I think Kent did have a reasonable point he was trying to make with the statement — there are publishers who do return the profits they make to the research community (not-for-profits, university presses, society publishers) and these should be contrasted with those who remove funds from the community, putting them instead into the pockets of shareholders and investors. The line between them can’t be drawn by looking at their open access (OA) policies. There are some OA publishers trying to siphon money out of the research community and into the bank accounts of venture capitalists and stockholders.

The poor choices of examples used and the combative tone ended up turning what could have been a provocative discussion in the comments into a bog-standard unproductive flamewar.

And that’s a shame, because there’s an important dilemma facing the scholarly publishing world that could use further exploration. We want things to progress, we want new tools to be developed, new discovery mechanisms, new ways of using and re-using the literature. But these things don’t come cheaply. Big changes require significant investment. In order to motivate that level of investment, we have to offer rewards to those taking the risk.

The question, then, is how to balance the rewards offered with a desire to limit profiteering. If we prevent the Elseviers of the world from reaping their profits, then they don’t spend the money necessary to create SciVerse, nor do they do all the interesting things they’re doing with APIs and encouraging app creation. If publishers can’t profit from journals, then they can’t pay for the promising new semantic technologies under development. A world without profit is seemingly a world with a lot less innovation.

The subject came up in a recent comments thread following Judy Luther’s piece about altmetrics. We see a horde of new startups, many privately-owned and for-profit, working to develop new metrics, ways to improve upon and replace the impact factor. Innovation will come faster if those companies are offered the carrot of profits as a reward for their hard work. At the same time, the scholarly community cannot afford to allow itself to become solely dependent upon a for-profit service that can change the rules and change the levels of transparency and access at the whims of investors.

There’s no easy answer here.

Another good example can be found in recent OA mandates. The RCUK mandate requires that papers be released under a Creative Commons CC-BY license. This means the articles can be re-used for commercial purposes without any compensation to the author or the publisher. The idea behind it is to spur innovation, to drive the creation of new businesses based on re-use of the literature. The problem is that it also removes any motivation on the part of the publisher for improving the articles, updating them as new technologies come along that make them easier and more informative for that re-use. Why bother building open APIs, creating systems to easily allow mass downloads, when there’s no return offered on your investment? Why strive to improve an article’s metadata when you’ve already been paid all you’re ever going to be paid for the article in question?

In this case, the scholarly community might be better served by thinking in terms of compulsory licensing. Mandates could require that non-commercial re-use should remain free, but for those intent on using journal content as raw material to make a fortune, a fair and affordable price is set for that raw material. Think of the small fees a radio station pays each time it broadcasts a particular song. This compromise would still enable businesses to be created from content re-use while at the same time motivate publishers to drive that re-use forward.

And that brings us full circle back to the important message of the book Kent was reviewing in the article that caused such a commotion. The growing business strategy of our era is to drive the cost of everyone else’s product to zero in order to make more money from your own product. If you can convince the world that the raw material you need to get rich should be free, then it’s easier (and cheaper) for you to get rich. It’s a great business model if you’re on the side where you’re getting things for free. It’s not so great if your livelihood has been devalued by others trying to cash in.

And most importantly, this imbalance stifles innovation and creation. If we want great new music to enjoy, we have to pay musicians. If we want new tools for scholarly research and we want the literature to evolve in exciting new ways, we have to offer rewards for those willing to invest the effort and money needed to drive things forward. There’s a fair balance in there somewhere between “free” and “way too much.”

Enhanced by Zemanta
David Crotty

David Crotty

David Crotty is the Editorial Director, Journals Policy for Oxford University Press. He serves on the Board of Directors for the STM Association, the Society for Scholarly Publishing and CHOR, Inc. David received his PhD in Genetics from Columbia University and did developmental neuroscience research at Caltech before moving from the bench to publishing.

View All Posts by David Crotty

Discussion

40 Thoughts on "Driving Innovation: Finding the Balance Between Fair Reward and Profiteering"

David, thanks for this. However I do want to disagree (surprise, surprise) with your characterisation of a CC-BY marketplace. If we take as given that the government agenda on this is to maximise all possible exploitation to generate impact then the publisher value offering has to address that specific demand. If we assume an APC (or submission fee) system then yes you don’t make additional money directly by enhancing existing articles but every one of those existing articles is a potential proof point (or advertising if you prefer) for whatever new enhancement service you want to sell to the next customer.

So I think there is a strong incentive if you think look at this as businesses competing to ensure the widest possible dissemination of the work they are paid to publish. And what is better, if its released CC-BY and there is a demand for a new kind of enhancement then if the original publishers refuse/can’t be bothered/don’t get enough return to do that enhancement then a third party has the rights to do that, either commercially or non-commercially so we safeguard the ability to keep that enhancement going into the future.

Your model of compulsory licensing could actually make it appealing to publishers to _not_ enhance a work because then they could charge a compulsory license fee to any commercial downstream party that wanted to do so.

Full Disclosure: I’m an employee of PLOS which is a not-for-profit that makes the majority of its income from author side charges for publication of articles and releases those articles under a CC-BY licence.

Hi Cameron,
I would argue that for the author, the extent of dissemination is pretty far down the list as far as choosing a venue for publication. Under the current system (right or wrong), the Impact Factor is the dominant player. Authors are strongly rewarded for publishing their papers in the journal with the highest Impact Factor, not the journal with the highest circulation. I’d also put speed of publication well ahead of dissemination. And, in a world of OA mandates that don’t provide the researcher with direct funds for APC’s, price is also going to be a major factor. I’m not sure how much publishers are going to be willing to invest in something that’s of such low priority as far as bringing in APC’s.

Your model of compulsory licensing could actually make it appealing to publishers to _not_ enhance a work because then they could charge a compulsory license fee to any commercial downstream party that wanted to do so.

But right now, that publisher can completely prevent anyone else from doing those enhancements. So we need to think of a fair mechanism to at least open things up. A compulsory license means the publisher can’t say no to any re-use, and eliminates uncertainty for the re-user. They know they’ll have access to the material and they know exactly what it will cost (and that the cost will be fair). If the publisher sees other publishers receiving a steady stream of income from re-users but is left out of the process because their content isn’t yet up to snuff, that seems to me to offer a motivation for getting the content into better shape.

I grant you, I’m thinking a little further down the line, when funders are explicitly focussed on downstream impact and demanding evidence that outputs are communicated in such a way as to maximise the potential for impact. That is coming and once it does the price-dissemination balance for whoever is paying the publication charge is going to a very serious consideration. Mandates are blunt instruments at the best of times but we’ve a way to go before the right kind of market is in place, and I can see an argument for transitional arrangements. But the impact factor itself is steadily being eroded as the important measure, at least in Europe and North America, prestige remains important and will continue to do so for a while but I’m seeing an increasing focus at the article level from a wide range of sources.

I can see your compulsory licensing argument as a step along the way but not an end game. It’s a way of forcing the publisher to give up some of the value they’ve captured but it still creates friction to innovation. Equally I find Kent’s argument that CC-BY removes all incentives to discover audiences peculiar. If their is value in a specific utility for a specific audience and that value can be delivered then there is a market. Indeed one of the key points of using CC-BY is to create those kinds of secondary markets (which NC does prevent and SA arguably could) and allow anyone to exploit them. It’s just that someone has to actually create real value and utility to realise that market, rather than capture all the potential value up front and sit on it. A business that sits around complaining about lack of incentives while preventing other businesses from creating value in the spaces that are left isn’t really contributing to the community either in my view.

Cameron, you’re mistaking copyright for a lock-up. Publishers give permissions to others all the time. We just expect them to pay a fair price if they want to use these materials for commercial purposes. This isn’t “sitting on them.” This is how economies work.

Thanks Cameron. I’m also thinking in terms of back-content. To be really effective, we want new tools that can analyze the entirety of the scholarly literature. Even if you have mandates that require everything going forward be freely available for all re-use, you’re still left with nearly a hundred years worth of material that’s under copyright (and that’s not even getting into material that’s not in journals, things like scholarly monographs and other books). Even if you can gain new authors (and new APC’s) by making new articles freely available, I’m not sure you’d get the same sort of gains from doing so with a 50 year old article. Why bother?

I’m trying to think of a practical way to incentivize the copyright holders to make that material available and to invest the money necessary so it can be re-used. At the same time, we want a system that’s affordable for those looking to create a new business from the old material.

Ultimately, in a real marketplace, this issue would resolve in pricing. If a CC-BY license actually adds value to what a publisher does (a dubious assertion), it simultaneously diminishes the long-term market value of the work because the work is never saleable again, and cannot be protected (it goes from “rival” to “non-rival”). The obvious answer is to charge more at the only point possible — through the APCs. The problem with this rational economic answer is that the marketplace for scholarly publication imagined by mandates keeps prices artificially low and makes value-based pricing nearly impossible. This would weaken scholarly publishing tremendously, make funders much more powerful than they already are, and drive many players from the market. The CC-BY license has to be viewed as part of a larger story about whether the future is one in which scholarly communications thrive, or one in which they slowly drop expensive quality aspects, lose ambition, stop experimenting, and shrink into a repository-like state.

As to whether the CC-BY license adds value, I am very suspect of these claims. Dissemination as envisioned by the CC-BY world becomes a fairly passive act, and incentives to make it an active, market-oriented activity vanish. Posting content is not the same as publishing content, and publishing content without addressing a clearly definable and interested audience is also not likely to generate sufficient awareness to make free availability much of a factor. Availability and desirability are two different things. Relevance and utility are huge factors. If the APCs are stymied by dependence on grant-funder and government adjustments for budgetary ebbs and flows, and if there is no other market upside to publishing content, why would a rational business invest the significant marketing, market research, and brand dollars necessary to develop, address, and inform a specific audience? This is one of the larger risks being courted by a CC-BY world. In the copyright/subscription world, there are real incentives to get published content into the right hands. In the CC-BY world, those incentives disappear.

Kent, you are ignoring the role that these same funders who mandate CC-BY have an incentive to make the literature vital, and have a long history of doing so through, for example, PubMed Central and many other initiatives of the National Library of Medicine. And dissemination of CC-BY papers need not be passive, as PMC demonstrates.

You also act as if CC-BY precludes others from adding value to papers, when it explicitly does not. If you develop tools that make individual CC-BY papers, or collections of them, more useful, you are free to charge people to use them. Nothing prevents you from curating the literature for a particular audience and selling your curation to interested parties. Indeed, by removing access to the underlying content, which publishers did not produce in the first place, it gives them incentive to actually innovate and develop services that people actually want to pay for.

However, under CC-BY, you as author do give up the right to object to a bad translation of your work (unless you can make a case that it was intentionally distorted in the process), and if a first translation that appears is poor, the incentive for others to produce a better translation may be lacking. You also give up the right to object to inclusion of your article in a collection whose constitution, for whatever reason, may be objectionable to you. Maybe scientists don’t worry about such things, but I can assure you that humanists and social scientists do.

Authors give up *all* rights if they transfer their © to a publisher. Inclusive of the right to object to a poor translation, inclusion in an objectionalbe collection, etc. They can object, but the © holder has no obligation to heed their objections. Perhaps humanists and social scientists never transfer their © to a publisher. They would certainly be well-advised never to do so if they wish to have the right to object to any use of their articles they don’t approve of.

It is standard practice among university presses, at least, for authors to be given the opportunity to check translations and to veto any inclusion of their articles or book chapters in anthologies.

So let me get this straight — if a company takes content they didn’t pay for, does something they claim is an “improvement,” and then charges for it, you’re OK with that?

I have great news: that’s what you claim subscription publishers are doing. Sounds like you approve!

This is one of the most telling comments of all time. Thank you.

So you’re saying you’re OK with making the papers themselves immediately freely available? Cool.

Actually, I’m happy to see the contract research reports taxpayers paid for made freely available, per the mandates and requirements of government grants. That means having them posted on government sites, which is an approach David Wojick endorsed a while ago. That’s really all taxpayers have paid for, and the DOE makes it work just fine. Why the NIH can’t make it work should exorcise the community a bit more, I think.

So, it’s solved. We agree. Taxpayers should have access to what they paid for (contract research reports), and if a company takes those and improves them, they can charge for access to those. Glad we can put this behind us.

If you actually looked at DOE research reports, instead of just cynically trotting them out your “solution” to access to the research literature, you’d know they consist of two things:

1) perfunctory summaries of progress on projects that contain little or no useful information

2) the full-text of publications arising from DOE sponsored research – something the DOE can do because US copyright law precludes federal employees from taking (and therefore assigning) copyright in their work

As it stands now, for the vast majority of NIH grants, the “research reports” you cite the same as category 1 – progress updates that contain for grants officials, that contain only a fraction of the information contained in papers.

Following the lead of the DOE, the NIH has tried to make the actual papers available, something you and your allies have fervently opposed.

Now, if you are suggesting that rather than put their time and effort into writing papers to be published in journals, NIH grantees instead devote their energies to writing research reports that contain a complete accounting of their ideas, methods, data, results and conclusions, and making those freely available, then we would actually have a solution.

So you are suggesting that we write everything out in great detail twice? It sounds like a massive waste of scholar’s time and will only make the literature harder to navigate with duplication of some content, where it isn’t obvious what corresponds to what. The logical extension of that suggestion is that the funders have their own journals and the traditional journals are bypassed.

I’m not suggesting it. Currently, mandates from government and other funders require a report be filed. Then, the results can be submitted for publication. That’s how the world works. And if funders create their own journals, who is going to believe them? Where is the independent third-party validation? Where is the filtration and ranking?

Sure, but the reports are usually summary reports and not terribly useful for anything. If they were to fulfill the goal of public access to research results they would have to be beefed up to the point of duplicate publication.

Sure Kent, exactly. As long as you don’t demand the copyright to the original material so you end up with complete control of it. If you can add value to it by anointing it as worthy of being in your journal, copy editing dissemination, archiving, assigning a DOI, etc. and by doing so motivate people to pay you for your effort, more power to you! That way you are being paid for the actual value you add as a publisher. As a side note, perhaps the charge for anointing it as worthy of being in your journal should go to the author since it is largely for their benefit.

In fairness and deference to David Crotty’s well written balanced post, what I just said is a gross simplification, but so is your post above.

Isn’t it hiring and promotion committees that benefit from the ranking system that having journals of varying levels of prestige provides? After all, it saves them the enormous amount of time it would take to read the actual papers. Perhaps it is they who should pay the charge, rather than authors.

Thanks for a thoughtful and non-inflammatory post.

I think we all share the same broad goal, which is to make the research literature as useful as possible to as many people as possible at a reasonable cost. I start with the assumption that the default state for any product of publicly funded (and here I include private funders with a public mission, like HHMI, Wellcome and other non-profits) is that it be made freely available to anyone without restriction. Restrictions on how scientific information can be used and who can use it inherently reduce its utility unless – and this is of course the crucial part – they are necessary to achieve the wider goal of maximizing the value of the literature.

So the question to me is is the value of the research literature maximized by taking something that we can easily make freely available to anyone to access, use and reuse at a relatively small cost (PubMed Central operates at a tiny fraction of the cost of subscriptions to scientific journals) and giving it to publishers who provide a series of services back to the community (reviews, editorial assessment, classification, etc… – much of which is done for free by scientists) at a very high cost (~$10b/year) with the end product available to a narrow audience as a result. To me it seems intrinsically obvious that we are better off if we (the research community) renegotiate this deal, paying publishers directly for these services without giving them control over the product.

I understand your and (to some extent) Kent’s point, that publishers need an incentive to provide these servies and to innovate in the process. This is why I have never argued that money should be removed from the system. But I fail to see how granting publishers copyright control over the literature accomplishes this. It’s been my experience that the balkanized control of the literature has been a major disincentive to innovation in its use. It took a long time, for example, for many journals to digitize their archival content, not because the technology wasn’t there or because there weren’t parties interested in doing so, but because they hadn’t yet figured out a way to sell it. From my point of view there has been remarkably little innovation in ways of accessing and using the literature because nobody is in a position to build new tools that can be applied to the literature in its entirety. I acknowledge that this is an unproven conjecture, but I believe we have cost ourselves over a decade’s worth of innovation in the ways the scientific literature are used because neither I as a scientist, nor any entrepreneur with an interesting idea. Instead we have Elsevier developing tools for Elsevier papers, Nature for Nature papers and so on. And the results to a practicing scientist like me is confusion and annoyance, as I have to care who published a given paper – something in general I don’t care about. And, frankly, I’ve not been all that impressed with the innovations publishers have made since the dawn of the Internet. I can’t think of one I couldn’t live without, and don’t think any come anywhere close to justifying the massive restriction on access and use that accompany them.

We’ve made the analogy before, but I still think its useful. Imagine where we would be if publishers had not made what, from their point of view, was a colossal error when they stopped publishing (and thereby having copyright on) DNA sequences. If the only way I could access an individual DNA sequence was through the website of the publisher who happened to have the paper that described it, and if we had to pay a fee for each one or have very expensive subscriptions, then none of tools that we rely on for analyzing sequence data would ever have existed because nobody would have bothered to develop them.

This analogy also highlights another important aspect of the “incentives” discussion – which is that scientists have a intrinsic incentive (building careers and reputations) to innovate in how the literature is used – this is what drove innovation in sequence analysis. I know a lot of scientists who are bursting with ideas about how to use, reuse and enhance the scientific literature – but they are effectively precluded from doing so by their inability to download the entire corpus and work with it in their labs. There is no technological reason they can’t do so – but they are effectively (and in many cases assertively) prevented from doing so by publishers who own the literature. This is, incidentally, what led us to launch PLoS in the first place – our inability as practicing scientists to use the literature as a computational resource in our research. I know from my experience in sequence analysis and other fields that, given unrestricted access to this information, scientists will do amazing things with it. And I see no reason to prevent this in the name of hypothetical innovation from publishers.

Anyway, I greatly appreciate your thoughtful post, and I think this is an interesting topic to discuss further.

This is a helpful post. It’s nice to see someone not accusing for-profit publishers of “killing” or stealing. However, can I assume that the qualifier “the default state for any product of publicly funded [research],” is a major and decisive one? By this reckoning, the argument stated here applies very, very narrowly. And, moreover, it depends upon the largesse and non-political discrimination of “public” funding arms, yes? Isn’t that a 600 pound elephant in the room?

Science depends on the largesse of public funding arms. Do you think they would spend upwards of hundreds of billions of dollars on research, but suddenly stop funding the dissemination of the results?

That’s not the question. The question is whether some quirky idea from an untenured professor at a small university in the mid-West will get the same funding as Cal Berkeley. Or whether, to the contrary, professors at Cal Berkeley control the “results” of public funding more than deservedly. If science depends on politically charged public funding arms, I think it’s fair to say we’re all in trouble!

Hi Michael,

Wow, there’s a lot to unpack in your comment. Let me address a few points, I don’t think we’re particularly far apart on where we stand.

First, I think that non-commercial re-use is essential. I perhaps should have stated this more clearly in the article. That seems to me a big part of what PLoS (and others) are trying to do. Open up the literature for more use in research, so we can mine it and drive new knowledge from work done in the past. That is an absolute, key goal.

The questions arise though, when we get into commercial re-uses. If someone is going to make money from that re-use, should they pay to support the raw material they’re going to exploit? If we restrict things to non-commercial use alone, I think that slows progress. I worry that we’d end up sending the brightest minds off elsewhere to do projects where they have some hope of making money. Given that, can we create incentives for publishers to make that raw material as re-usable as possible?

One example: OUP recently spent an enormous amount of time and money building “The Oxford Index”, a comprehensive, accurate and organized metadata catalog of all OUP content (journals, books, reference books, etc. dating back 500 years). It’s a really helpful (and freely available) resource for aiding discovery tools. The index should increase usage, and that will increase revenue from book sales, subscriptions, etc. If there were no way for OUP to commercially exploit that back content, to earn back the money invested in creating the index, would they have created it?

The back content is a particularly thorny issue here. What PLoS and mandates are doing may solve the issue going forward, but we want those tools to include the last 80 years or so of material that’s still under copyright. We also want them to include new content that may not be covered by mandates from particular funders. Can we create a system that incentivizes making that material available for re-use but one that keeps the price fair and affordable? And if by doing so, we make that same work freely available for non-commercial re-use, that’s a big win as well.

Mike – I’d take issue with your assertion that publishers delayed digitizing their archives because they couldn’t find a way to sell them.

For non-profits at least, the reason was simply the costs involved. These were initially astronomical; however, around the time Google got involved they dropped by orders of magnitude. At that point, every non-profit I can think of digitized their archive, and most who’d responded positively to the initial PLoS petition in 2001 simply included this back content with the existing freely available archive released after 6 months.

We approached a large number of non-profit publishers in the late 90’s asking if they would let us digitize their back content and put it into the public domain if we paid for it, and they all declined citing potential future lost revenue.

Universities have typically shown a reluctance to invest enough in their own presses to spur innovation, yet when commercial publishers made the necessary investments, universities complained about the prices they charged. Somehow, it seems, the higher education community just wants innovation without ever being willing to pay for it.

There are undoubtedly good intentions behind this piece, but I am unsure of the logic in a number of places.

1) “return[ing] the profits they make to the research community” is a line worth drawing. As the public money that go into publishing could have gone to the research community, we need to look at the total calculation. If the public spends $1000 on an outdated Toll-Access publisher that returns $50 to the community, that is plainly a worse deal than spending $100 for the same service on a modern OpenAccess publisher that returns $0 to the community, right?

2) “We want things to progress, we want new tools to be developed, new discovery mechanisms, new ways of using and re-using the literature. But these things don’t come cheaply. Big changes require significant investment”. For litterature that is free to use and re-use, new ways of doing so obviously don’t require “significant investment”. If “significant investment” means money on the scale of Elsevier or other large Toll-Access publisher profits then I believe that ArXiv didn’t require “significant investment”. CiteULike didn’t, Mendeley didn’t. Right?

3) “If publishers can’t profit from journals, then they can’t pay for the promising new semantic technologies under development. A world without profit is seemingly a world with a lot less innovation.” This is hard to follow. If (some) publishers can’t profit, it doesn’t follow that a “world without profit” happens, right? It also seems to imply that publishers are a particularly good place for innovation in eg. “promising new semantic technologies”. There is no reason why this should be so. Google or search engines are probably much better placed for that, right?

4) “The problem is that it also removes any motivation on the part of the publisher for improving the articles, updating them”. Why is that a problem? If others have free access to update and improve them, why is it a problem if a specific publisher lacks motivation?

Hi Anders,

Let me see if I can add some clarity to my logic:

1) As you note, it’s a question of finding the right balance. For the researcher, your goal is to minimize costs, maximize the service levels and maximize the amount of funds that remain in the research community. Publishing is a service industry and it’s up to the researchers to make the right choices that best serve their needs.

I’m not exactly sure how to answer your specific question though. It doesn’t seem like an apples to apples question. Is the public spending nothing for authors to publish their papers in the toll journal and then $1,000 for others to access the papers? Is the public paying $100 for each author’s paper in the OA journal and nothing for everyone to access them? I suppose the answer would then depend on the total number of papers. For 10 papers, the public spends the same $1,000 on both but the research community gets $50 from the toll choice.

A better question is to think in terms of equals. Two fully OA journals, each charges $100 for an APC. Journal 1 is for-profit and owned by venture capitalists. Journal 2 is owned by the research community. There’s a $30 surplus left over after costs on that $100 for both journals. Journal 1 puts that money into the pockets of the venture capitalists. Journal 2 puts that money back into the community through funding research, scholarships and advocacy. It’s pretty clear which is a better deal for the community.

2) It really depends on the individual tool. Why limit progress to just things that are inexpensive? Won’t that slow the pace of things? How much did the human genome project cost? Was that worth the big investment? Would the improvements in sequencing technology ever have happened without that investment (and without the private, for-profit companies that were involved)?

As for your examples, arXiv costs some $400,000 just to keep running each year. That’s a significant amount of money in my book. The privately-owned, for-profit Mendeley cost an enormous level of investment from its ownership, mostly former record company executives from what I’ve been told. They’ve made great progress, but it certainly hasn’t come cheap.

3) Sorry if having those two sentences together implied that only publishers can bring progress. Progress is welcome from all areas. That said, publishers are a particularly likely arena for progress in publishing. Publishers already have the infrastructure in place, as well as a depth of knowledge and experience that can help move things forward. Before you scoff, please remember that PLoS is a publisher. PeerJ is a publisher. The Society for Neuroscience is a publisher. Cold Spring Harbor Laboratory is a publisher. The University of California is a publisher.

I’m hesitant to endorse Google as the model organization for progress for the research community. Google is publicly traded, and their responsibility is to generate a return for stockholders, not to improve things for researchers. The money to be made off of selling Google Ads to the research community is likely pretty small compared to other target markets, so I don’t expect huge efforts from Google. And even so, what’s the difference really, between being under the thumb of Elsevier’s investors and being under the thumb of Google’s investors? Isn’t it better for the research world to own the companies it relies on and to control it’s own destiny?

4) It’s a problem because it slows progress. Why not offer rewards to those willing to move things forward? Again, why limit progress to only those who can work inexpensively and expect no return on their efforts? Is this the best way forward or is this a slow path to stagnation?

It’s also a problem on a very practical level. If we want to expand access to the literature, and offer everyone that free access you mention, then we’re going to have to deal with the last 80 years or so of material that’s under copyright. Won’t new tools be better if they can use the entirety of the scientific literature, rather than having to ignore most of the 20th century and part of the 21st? Can we offer an incentive to copyright holders to open up that material, freely to those doing non-commercial research, but with a decent (but affordable and fair) level of return for those using it as raw material to turn a profit? Seems like a win-win to me.

1) I meant total cost or total cost per paper. Not all publishers have the same costs, so for the public to pay an expensive publisher lots is dumb, if the extra cost outweighs any research community kick-back scheme. Right?

So when you call for an “apples-to-apples” question, that is exactly where I can’t follow you. Do you claim that this is the relevant issue to look at – because you think almost all publishers costs the public just about the same in total per paper, and that only the amount of return contribution varies?

2) I guess the burden of proof is on you to show that “significant investment” of public money is needed for new ways of reusing litterature that is free to use and re-use.

And you seem to mix up development and ongoing cost. However, neither development nor ongoing costs of ArXiv is a “significant investment” on the scale of Elsevier or other large Toll-Access publisher profits. Right? The distinction of developement and ongoing is very relevant however. If good ideas are cheaply developed, the public can pick those and finance them going forward. Just like ArXiv is financed ongoing.

3) Again, as in 1) if Google can develop “promising new semantic technologies” cheaper than commercial Toll-Access publishers like Elsevier, then I don’t see why we should prefer Elsevier. Do you? Especially since Google has a history of developing services (Google Scholar, Google Books) without requiring an arteficial monopoly on the material being analysed.

4) “Why not offer rewards [to publishers]”. Or rather, why? I think it is fair that you have to argue that the public money is better spent on those rewards than on actual science or on someone else than publishers.

And on the practical level you seem advocate holding the future hostage to the past. How about we implement our open access mandates now, and then see what innovation rises from that free access to improve and update articles. That sounds like a better way to spend public (and private funder) money to me. How about you?

Hi Anders–

1) I think my answer and your response both speak to the futility of asking your original question. There’s so much complexity, and so many different players with different agendas involved that asking over-simplified questions isn’t appropriate. It leads to magical thinking, rather than practical, real world solutions.

But as noted, each party involved should do what it can to maximize the return on its investment. For a funding agency, that may mean demanding that funds be spent strictly on research, rather than on publication of research (either through subscription fees, page charges or APC’s). For a postdoc, maximum return may be based on the things that will lead to a job, like the Impact Factor of the journal, rather than the dissemination method. For those leading a research society, a balance must be sought between maximizing funding for the societies activities and doing what’s in the best interests of the members and the field.

2) Hmm. Elsevier doesn’t publicly release their budgets, though I’d love to know what they are spending on SciVerse. Some places where numbers are available:

I think we can agree that PLoS has been innovative in creating new ways to use and re-use free literature. They were started with a $13 million grant from the Moore Foundation, which strikes me as a significant investment. In 2010, PLoS operated in the black for the first time. We have regularly been told by PLoS representatives that the excess funds from that year have been reinvested into new developments at PLoS:
http://scholarlykitchen.sspnet.org/2012/07/26/review-and-discussion-free-ride-how-digital-parasites-are-destroying-the-culture-business-and-how-the-culture-business-can-fight-back/#comment-51539
Michael Eisen: “PLoS is investing much of its surplus into new tools to make the publishing and post-publication process better and more efficient.”

As another example, CrossRef has provided (and continues to provide) tremendously useful tools for improving usage of the research literature. Their launch cost some $4.2 million (http://www.crossref.org/08downloads/CrossRef10Years.pdf).

Both of these groups are important sources of progress for the research community. But both have incurred significant costs in producing the useful tools they’ve created.

As for ongoing costs, those are extremely important. There’s very often grant money offered to create a new tool or a new community resource. Once up and running, it’s often very difficult to find funding to keep that resource alive. A resource that ceases to exist ceases to provide value to the community. One must look beyond mere startup costs, and as a resource catches on, costs often scale up enormously.

3) I’m not suggesting that you should prefer Elsevier or Google. Both are Wall Street-driven for-profit megacorporations. I’m suggesting you should prefer (when given the choice) technologies and companies that are owned and run by the research community itself, with goals of benefiting research and researchers, rather than making hordes of money for shareholders.

As far as Google not wanting artificial monopolies, you are clearly unfamiliar with what they were trying to do to orphan works, establishing a private, artificial monopoly via their Google Book scanning program and cornering the market on being the sole provider of access to an enormous amount of the world’s literature.

4) Why not publishers? Why not offer those rewards to anyone who can improve things? Are you so biased against publishers (publishers like PLoS, BMC, PeerJ, eLife, university presses, research societies, etc.) that you’d rather cut them out of the picture altogether and only instead offer rewards to private entrepreneurs? Why artificially limit progress?

As for holding things hostage to the past, I don’t see why it’s an either/or question. Yes, implement OA mandates now and going forward. But I think it’s useful to try to broaden access to works already published. The last century of science is fairly important as far as I’m concerned. Why not work to improve access to both? You can’t magically wish away the copyright status afforded to works over the last century or so. You either have to deal with it, or ignore that part of the literature. The former seems a lot more practical than the latter.

1) I think anyone with just a minor bit of economical sense would answer my original question clearly with “Yes, look at total costs. No, kickback schemes are no excuse for wasting money on inefficient publishers”.

Do you have a different answer?

2a) Strictly speaking, even if those examples were valid, then you’re not showing that ““significant investment” of public money *is needed* for new ways of reusing litterature that is free to use and re-use”. You are giving examples that varying amounts of money CAN be spent. On this we agree.

2b) You never answer my “neither development nor ongoing costs of ArXiv is a “significant investment” on the scale of Elsevier or other large Toll-Access publisher profits. Right?” Does that mean that you agree with me that we clearly overspend on large Toll-Access publishers?

2c) I didn’t say that ongoing costs are not important. I said that you seem to mix up development and ongoing cost. Do you think there is no relevant distinction?

3a) I disagree. I think we should look at which options allign with our needs and goals. Eg. as a citizen and a funder of science (through taxes) I would like to maximize the utility of the science I fund. In that respect Toll-Access publishers that artificially restrict the dissemination of science are crap pretty much however they are owned and run.

3b) I think we disagree on the mechanics of the Google Books settlement. However, I think we both agree that Google has made very successfull products that don’t requre a monopoly on the input – Google Search, Google News, Google Scholar, right? Therefore I don’t see why we should waste money on publishers if they claim that they need monopolies to build services on top of scientific litterature. Do you?

4) The whole point is that if you really want to offer “rewards to anyone who can improve things” you can’t also advocate overspending specifically on publishers, like you did originally. To actually be able to offer “rewards to anyone”, then you need to spend the minimum possible on publishers, and then offer – open to “anyone” – the left over money for innovation.

Purposefully overspending specifically on outdated publishers doesn’t fit my definition of offering “rewards to anyone who can improve things”. Does it really fit yours?

Anders, I think because of your anti-publisher bias, you’re reading a lot more into this column (and my answers) than is really there. At no point do I suggest that we should overspend at all, nor do I advocate overspending specifically on publishers. I’ve repeatedly stated that each interested party has a duty to maximize the return on their efforts and investments. I have stated that publishers do come to the party with a wealth of experience and an existing infrastructure that can help provide a better economic value than those starting from scratch, but that shouldn’t prevent others from participating (the altmetrics companies I specifically mention in the post aren’t “publishers”).

1) I didn’t answer your question because it’s pointless. In a magical world, if pony rides were free, would this be better than a world in which pony rides cost money? The answer was obvious, but the question doesn’t really help us formulate a business strategy for transportation.

2) Do you think that I think there is no difference between startup costs and ongoing costs? Do you think that both are important and essential in providing a useful service or running a business? Can one be completely ignored in favor of the other or must both be considered?

3a) I’m suggesting that big, money-hungry, Wall Street-driven corporations may not align well with your goals.

3b) I’ll stick with the Department of Justice and the court’s interpretation of the Google Books settlement. Services that avoid monopolies are better for both consumers and content creators.

4) Let’s make this clear then: get the best value you can for your money. Spend where it’s wise to spend, wherever that may be. The more you control your own destiny, the better off you are, rather than relying on commercial interests that may or may not align with your interests depending on the whims of stockholders or Wall Street. Fair enough?

David, I am a huge fan of PLOS and I think I am going to like eLIFE quite a lot also. So I don’t know why you claim I have an anti-publisher bias? If you wanted to make a true statement, maybe you could claim that I have a bias against inefficient, outdated, Toll-Access publishers. That would be correct.

If you are not suggesting to overspend, why don’t you just answer my 2b) above, about “”significant investment” on the scale of Elsevier or other large Toll-Access publisher profits”? It is a relevant question since in you original post caution “If we prevent the Elseviers of the world from reaping their profit […]”.

And in the same paragraph you have ““If publishers can’t profit from journals, then they can’t pay for the promising new semantic technologies under development. A world without profit is seemingly a world with a lot less innovation.””

So as far as I can tell, you actually do “advocate overspending specifically on publishers”.

And while you claim above that you’ve “repeatedly stated that each interested party has a duty to maximize the return on their efforts and investments”, your original post with it’s cautions agains moving profits from eg. Elsevier, seems to me to go completely in the opposite direction from recognizing that taxpayers and funders have a legitimate interest in moving money away from outdated, inefficient, toll-access publiser and using the funds on actual science or on modern open-access publising that allow anyone to innovate on top of the results of the research that funders have paid for.

I’m sorry Anders, you still don’t seem to be getting the gist of this blog posting. Let me try to summarize again for you:

Progress can be sped up by offering rewards for achievement. If we want the best minds to work on solving our problems, if we want significant investment made to improve things, then likely that’s going to cost money. If we want the bright minds figuring out valuable altmetrics to spend their time doing so, rather than going to crunch algorithms for investment bankers, we need to pay them. If we want the really interesting tools Mendeley offers us, then we’re going to have to pay back the $2 million that venture capitalists initially invested many times over (not to mention the subsequent rounds of fund raising).

As you point out, there is an alternative to this. Instead, one could wait and see what develops organically, what comes from academia and the non-profit world. In many cases, this may suffice. But it also means that progress may be slower–there may not be as strong a motivation if there’s no financial reward offered. And depending on the issue, that slower pace may be acceptable. In planning any major project, we are faced with choices:

We can have the work done quickly.
We can have the work done inexpensively.
We can have the work done at a high level of quality.

The standard cliche is that you can have any two of the above, but never all three. Each situation requires an individual judgment about what the priorities are and what the right level of investment is.

I do not, in any way, for any industry, advocate “overspending”. I’m pointing out that profits are part of the picture for progress. If the progress offered is not as valuable to you as the price it costs, then do not spend the money. I can’t tell you what’s fair and what’s “overspending” for you. We each are in very different positions, and my needs are different from yours. Each player in the game must make that call for him/herself. I don’t purchase or use any Elsevier journals. The content and services they offer are not of value to me. You must decide for yourself the value they offer. If the balance falls too far on one side, then spend your money elsewhere.

To get back to your question 2b, I can’t answer it as I don’t know how much Elsevier spends on R&D. I’m also not sure that percentage of revenue spent on R&D it’s a meaningful measure of anything. Apple spends 2.2% of revenue on R&D, but given the value and use I get out of my MacBook and iPhone, I don’t consider those “overspends” (though conversely, the much less expensive Dell laptop and Blackberry I own are definite overspends).

In 2010 (still no 2011 numbers which are expected to be much higher), PLoS had a profit margin of 22%, significantly higher than either of the academic community-owned presses where I’ve worked (CSHL Press and Oxford University Press). Does that mean the community is “overspending” on PLoS? Those other university presses return 100% of their surplus to the academic community to fund research and other activities, where PLoS spends 100% of its profits on further developing PLoS. Higher profit margin, no return to the community–is that an “overspend”? If the only thing that mattered was getting the lowest price and minimizing profit, the answer would be yes. But in reality, I’d say it is not an “overspend” because I think there’s great value in the new methods and technologies PLoS is developing–I want them to have the money available to continue to hire the best people and do interesting work, that in the long run, that will pay off for the academic community.

Some times it is more rewarding to spend more. As I said in the original posting, “there’s no easy answer.”

And I say you have an anti-publisher bias because you seem very upset with Elsevier’s copyright policies, profits and margins, yet perfectly willing to support companies like Google who have similar margins, who make 7.5X as much in profit, whose entire business is based on keeping knowledge (algorithms) secret, who have tried to create a monopoly on access to the world’s literature via their proposed book settlement, and who are engaged in brutally clamping down on innovation through restricting the reuse of intellectual property through patent lawsuits. Seems somewhat contradictory, doesn’t it?

3b revisited) Your faith in Google seems to require a willful ignoring of the reliance of their business on patents, particularly software patents. Why is this sort of monopoly control of intellectual property acceptable? And given Google’s recent purchase of Frommers, shouldn’t we refer to them as a publisher?

Comments are closed.