Source: Chemistry World
Source: Chemistry World

What do we mean when we say “open access” (OA), and what are we working towards when we promote it?

Definitions: Less Ambiguous

The question of what OA means seems now to be pretty much settled: as I listen to and participate in conversations within the OA community (by which I mean the organizations and individuals who advocate for OA and who set up and manage OA initiatives such as institutional repositories, publishing concerns, lobbying groups, etc.), there seems to be a hardening consensus around the proposition that OA is defined by the principles laid out in the Berlin Declaration on Open Access to Knowledge in the Sciences and Humanities of 2003 (which itself expands on the Budapest Open Access Initiative statement of 2002). There are some who maintain that the Berlin Declaration has been the official definition ever since it was published, but until fairly recently I’m not sure the community itself had reached consensus on that. It seems to have done so now; today I think few, if any, in the OA movement would disagree that OA means what the Berlin Declaration says it means.

To the degree that it has settled this question, then, that community has adopted a definition that includes two major provisions:

a) unrestricted public access to “a complete version of the work and all supplemental materials,”

b) without any meaningful copyright restrictions.

Consequent to the hardening consensus around the official definition of OA, an important distinction between “open access” and “public access” has emerged. In this context, “public access” is what you have when the content of a document is freely available to read, but remains constrained by traditional copyright restrictions. In the United States, when government granting agencies define access requirements, these are generally couched in terms of public access rather than open access (see, for example, the White House Office of Science and Technology Policy memo and the National Institutes of Health Public Access Policy), thus allowing authors to retain exclusive copyright prerogatives in their work. Among authors, for a variety of reasons, there remains significant ambivalence about abandoning those prerogatives.

Ultimate Goals: More Ambiguous

The question of what we are working towards when we undertake to bring about an OA future is less settled. Prominent figures in the OA movement have expressed a diversity of visions for the future, and here it’s important to distinguish between models and goals. There are different models of OA, and variations on each one. Under the “Gold” model, documents are made freely available from the moment of publication (preparation costs being covered by institutional subvention or author fees). Under the “Green” model, the document is published in a normal toll-access venue, typically a subscription journal, but some version of it (often the author’s final peer-reviewed manuscript) is placed in a public repository, often becoming publicly available after some embargo period designed to let the publisher try to sell access before access becomes free. Different models of OA are not necessarily in competition with each other; multiple models can coexist, and do.

What can’t coexist are a world without any toll access and a world that includes at least some toll access, and the question of which of these we should be working towards can be a touchy one. Not everyone who believes in OA is necessarily working towards a world of universal and immediate OA. Some prominent advocates see a legitimate place in the future for nonprofit toll-access scholarly publishing (no less an eminence than Robert Darnton has expressed support for this scenario). Some, especially OA publishers like Elsevier and Springer, see embargoes as an acceptable long-term solution to some of the challenges inherent in the Green model, while other OA proponents reject the idea of embargoes out of hand. Still others (such as Heather Joseph of SPARC) have expressed qualified support for embargoes as an element of compromise during the transition from toll access to OA, but one that should eventually be “remov(ed)… altogether.”

Questions for Authors

The fact that there is such a diversity of opinions and visions, even within the OA community, as to what that community’s ultimate goals should be, raises some questions for authors who have to make decisions about how to publish their work. These include:

  1. What do I think the future should look like? Do I want to continue supporting the prevailing scholarly-communication system (one that generally requires people to pay for access to what I publish), or do I want to help create a new one, in which access to what I publish is freely available for reading and reuse? If the latter, then which model of OA makes the most sense to me, and do I want to see that model become the universal program? Or would I prefer a blended solution, or perhaps some third option?
  2. When I am encouraged to support OA initiatives or join OA organizations, what are the ultimate goals of those initiatives and organizations? If you’re being invited to climb aboard a train, it’s reasonable to ask where the train is going. Interestingly (and perhaps a little disturbingly), I’ve found that it’s not always easy to get people or organizations to talk about their ultimate goals.
  3. What are likely to be the unintended consequences of any choice I make? Advocates for any program—whether for traditional publishing or for some version of an open access future—will invariably describe that program in terms of its desired outcomes and intended consequences. This is in the very nature of advocacy, and advocates will often object, sometimes angrily, to any attempt to discuss the possibility of undesired outcomes and the inevitability of unintended consequences. But for those who want to make rational decisions about where to invest their time or energy or money, discussion of those things is essential; what matters about a program or initiative is not just what it intends to do, but what it is actually likely to do.

The issues around access to scholarship are complex and difficult, and the right position on them is not always clear. For authors, especially—whose work is the lifeblood of the scholarly communication ecosystem—thinking carefully and critically about them is essential.

Rick Anderson

Rick Anderson

Rick Anderson is University Librarian at Brigham Young University. He has worked previously as a bibliographer for YBP, Inc., as Head Acquisitions Librarian for the University of North Carolina, Greensboro, as Director of Resource Acquisition at the University of Nevada, Reno, and as Associate Dean for Collections & Scholarly Communication at the University of Utah.

Discussion

51 Thoughts on "Open Access: Meaning(s) and Goal(s)"

Rick, my view is that researchers’ main goal is to do research. If they have something to publish, they want to get that done, and get on with their exciting research again. They are already finding the act of writing papers (e.g. following detailed, tedious author instructions for a particular journal, setting aside half to a whole day to get past a submission system) a burden. They are not really interested in copyright law, but just want someone to tell them a safe license that lets them disseminate their report to their peers as quickly and efficiently as possible.

I think it is the job of publishers to understand, and then to guide authors, or at least give them a reasonable default to choose, rather than give them a list to choose from (e.g. CCby, -ND, -NC, etc). At the very least the authors should know the unintended consequences of slapping on ND or NC.

Kaveh I can see that you have never negotiated a contract with an author! All my authors and there have been hundreds have questioned me about copyright and the implications of clauses regarding it. This was true for both domestic (US) and international authors, society and commercial publications, and OA and traditional means of distribution.

In short, if this topic was not a concern of authors there would be no color coded OA or an alphabet soup of various rights reserved or given by an author.

Thanks for this. Just one quibble. You write that open access is “without any meaningful copyright restrictions.” I assume you mean by this, without restrictions on re-use derivative works — which is true. But of course copyright does have one role to play in OA.

To quote the Berlin Declaration:

… a license to copy, use, distribute, transmit and display the work publicly and to make and distribute derivative works, in any digital medium for any responsible purpose, subject to proper attribution of authorship.

As the declaration goes on to say, “community standards, will continue to provide the mechanism for enforcement of proper attribution and responsible use of the published work, as they do now”. But the clever bit of legal jiu-jitsu that is the Creative Commons Attribution licence (CC By) does more: it effectively makes plagiarism illegal. That is, it licences free use of open-access papers only so long as the author is attributed — so that uses where the author is not attributed are not covered by the licence, and so constitute copyright violation.

So rather than writing “without any meaningful copyright restrictions” I would say something like “with copyright used only to enforce attribution”.

Attribution is not a property of copyright protection. Plagiarism isn’t about making unauthorized copies or creating derivative works; it’s about taking what someone else wrote and pretending you wrote it. As for copyright, If you look at the exclusive prerogatives afforded to authors under copyright law, none of them exists under CC-BY licensing.

(Just to be clear, that’s not a critique of CC-BY licensing; I’m glad that it exists as an option for authors. But it is a reason why authors who value their rights under copyright law are hesitant to adopt CC-BY.)

Insisting on attribution in a license isn’t the same thing as turning plagiarism into a breach of copyright law. Copyright law is completely unaffected by the terms of licenses.

OK, I will try one more time, and call it done.

Scenario A: someone uses my CC By work and attributes me. Their use is covered by the CC By licence, and all is well.

Scenario B: someone uses my CC By work without attributing me. Their use is not covered by the CC By licence (nor presumably by any other licensing agreement). Absent any functional licencing agreement, their use of my work is a violation of my copyright.

Now is it clear?

It is clear, but you’re still mistaken. You said that CC-BY “makes plagiarism into a copyright violation.” It doesn’t. Failure to attribute may be a breach of the license and thereby make an unattributed use technically illegal, but it’s not necessarily illegal because it breaches copyright. If the use was fair under copyright law, it remains fair under copyright law regardless of whether it’s within the bounds of CC-BY terms.

Are you merely pointing out that CC By works, like works licenced any other way — or indeed not licenced at all, but with all rights reserved — are subject to Fair Use?

That’s one element of what I’m pointing out, yes. But if you’d like to discuss this further, let’s take it out of the comments area — feel free to email me at rick.anderson@utah.edu.

I can’t find much evidence of the CC BY license being tested in court. Most of what’s listed on the Creative Commons wiki is either in progress or dealing with NC and ND clauses:
https://wiki.creativecommons.org/Case_Law

Has anyone sued successfully over attribution alone?

One thing to consider in these discussions is the motivation for a publisher to engage in lengthy and expensive legal actions on behalf of authors. Under traditional copyright, the publisher has a financial interest in enforcing the terms of the copyright and requiring licensing of material. Hence they will send in the lawyers if someone is in violation. With CC BY, that financial motivation goes away–the publisher is not going to earn a penny from anyone’s reuse, so why spend the money on lawyers?

As I recall, when Apple Academic Press reused CC BY articles without proper attribution, no legal action was taken against them:
http://rrresearch.fieldofscience.com/2013/08/a-clearer-perspective-on-cc-by-reuse.html

Do you know of any OA publisher that has taken legal action over attribution?

Michael I have had entire books and articles placed on the web by unscrupulous people. The cost of having these works removed are high. I do not think the OA movement was meant to make works freely available for the profit of someone who steals.

Thus to me, the ease with which someone can steal another’s work is the problem.

When I discuss copyright, I ask are you prepared to defend your copyright? I ask because the company will not do so and I make this point very clear. This always gives pause and a reconsideration of weather or not to hold copyright.

As a case in point. Say I put together a reader for college students. I simply take articles which have your statement on the cy page. I get lucky and the book sells lots of copies and I make a pile of money. On the book’s cy page there is the traditional cy statement. I can put it there because I am copyrighting the presentation. Say, one of your articles is in the book and in fact is the reason for its success. Should you get a piece of the pie? Are you prepared as the author to go after the company?

The OA publishers don’t care if the above happens. Their expenses and profit are covered by author charges.

Harvey, you can’t steal what is given away freely. By choosing a CC BY license, the author is making a deliberate choice that no, they do not wish to share in any monetary rewards anyone else derives from reuse of the material. So if that decision is made, then the rest of your questions become moot.

I think Harvey’s points become more relevant in situations where authors are coerced into accepting CC-BY, rather than accepting it freely. (Of course, there are shades of coercion at play here; having to accept CC-BY as a condition of getting research funds is somewhere in the middle between free adoption and true coercion.)

Very true. I think he also makes an important point (in line with a comment I left above) that if you choose to go this route and there is a violation of the CC BY license, you, as an author, are likely going to have to handle this on your own, rather than having the legal team of your publisher pursue it on your behalf.

Wow. That sounds rather familiar. Like when I am coerced into to signing over copyright in order to get my manuscript published in a subscription journal.

That’s exactly right. To the degree that you depend on publishing in a journal that requires you to sign over copyright, there is coercion involved in that transaction. Many authors see this as one of the downsides of traditional publishing arrangements.

This requirement that the CC-BY license be used for a publication to be considered fully OA has troubled me for some time. As managing editor of a society-owned and -published OA journal, we have chosen to use the CC-BY-NC-SA license, which promotes full use and access but still offers the author some degree of protection from other entities profiting from their work (often without even notifying the author of the use). It seems to me that the NC-SA license maintains the spirit (if not the letter) of what is intended by OA.

Even under the ancien regime I cannot recall any publisher actually going to law over the violation of copyright in a journal article. If an author complained as did happen from time to time a stern letter might be sent but actually a little violation like this was not of any financial concern to the publisher.

Anthony

It has happened a number of times in my career, but is kept quiet. Just because we don’t hear about it doesn’t mean it isn’t happening. It’s often in everyone’s best interest to keep these matters quiet, the rights violator being among them.

Of course we all want a talented support team that allows us to do the creditable work without getting bogged down with the creation of a publishable product. Logically, this enables one to produce more creditable work and thereby propel one’s career at the highest possible rate. Wide access to that work assures the greatest possible credit. Maximizing productivity and credit is the ticket sought.
The problem with this desire is that a talented support team that handles all of the details will cost money. The farther that work is from a DIY effort, the more expensive it is.
So, how is this managed so that costs are minimized and who pays? What organizations are best suited to pursue these goals? Organizations that seek profit and organizations that seek surpluses to fund their other operations seem to be in conflict with minimizing costs no matter who pays or how they pay. The inflated cost of publishing has to be dealt with first – all else is tied to it.

You write “What can’t coexist are a world without any toll access and a world that includes at least some toll access”. I’m not so sure. I’m also not sure that either Green nor Gold will end up being the business models that will get us 100% open access in a sustainable manner. The best estimates I’ve seen is that around 45% of articles are available freely under a green model and that 5-8% are available under a gold model – this 20 years after Budapest. Seems that there’s a lot of friction to overcome yet and I would have thought if Green or Gold were the killer model, they’d be yielding results faster. Which leads me to wonder if we’ve all been looking at OA as a black-white thing – the content is either free or it’s toll-access. But what if it’s both, what if the solution is a shade of grey? What if publishers took a leaf out of the low-cost airlines book and broke up the value proposition into different pieces and charged accordingly? Or, if you don’t like low-cost, why not look at the many other digital media businesses that allow some level of free access and charge for premium services? In other words, a Freemium business model. In this model, the content would be free to all but only subscribers would have access to value-added services. So, yes, toll access and no toll access coexisting. I have real problems with Green and Gold because neither model encourages any investment in audience-building (you can be free and have no audience, you can’t be Freemium without a large free audience); and Green and Gold exclude the reader from having a say in what the final service looks like (with Freemium, the reader is at the heart of the service). At OECD, we’ve been a Freemium Open Access publisher for a number of years and it’s proving to be an interesting business model – for books as much as for journals. It means that 100% of our content is freely available to all via a basic read-only service (and we encourage them to share and embed these files on their blogs/websites too) and 100% is also behind a service paywall.

You write “What can’t coexist are a world without any toll access and a world that includes at least some toll access”. I’m not so sure.

Toby, the freemium model that you then go on to describe is one that, I’m pretty sure, OA advocates would characterize as representing toll access (since it doesn’t provide free and unrestricted access to “a complete version of the work and all supplemental materials”). But I’d be interested in hearing from OA advocates who disagree.

I’m also not sure that either Green nor Gold will end up being the business models that will get us 100% open access in a sustainable manner.

Agreed.

In our version of Freemium we do provide free and unrestricted access to a complete version of the work, i.e. the full text and any supplementary materials (if any) – we allow allow (and encourage) free and unrestricted share and re-posting of the free file. I’m sure some OA advocates won’t like our model because we only allow read-only access, if you want to download, use offline and have easy copy-paste functions, then you have to pay/have subscription access. But I would argue that, in the absence of any other sustainable funding model, by using Freemium we’ve managed to make 100% of our published works available to all, for free and with no APCs, which is a pretty good result and goes further to achieving Budapest than many other publishers have managed.

Read-only access wouldn’t qualify as OA under any definition of the term that I’m aware of.

Please note that I’m not saying anything against the model you describe; it sounds intriguing, and if it works for you and your authors, then that’s great. But it isn’t what the OA community recognizes as OA.

According to wikipedia, gratis OA is anything that has “free online access” which, by my reckoning, read-only does (especially when being accessed on smartphones and tablets where the concept of downloading is largely redundant) – but, I agree, I’ve yet to meet an OA advocate who thinks that read-only is OA. But the other thing about Freemiium is that the features can evolve and those that were premium could become free when new premium features are available to sustain the premium offer. We’re aiming to keep adding/refining premium features so we can add ‘download’ and ‘copy-paste’ to the free service – but we’re not there yet. Personally, I’m more interested in delivering what we can today OA-wise, on a sustainable basis, than go bust over-reaching ourselves – as they say in France, perfect is often the enemy of the good.

You write “What can’t coexist are a world without any toll access and a world that includes at least some toll access”. I’m not so sure.

You’re not sure whether X and NOT X are mutually exclusive?

Anyway …

Speaking as one of those OA advocates who wants everything to be open: as OA becomes increasingly prevalent (as it is, year on year) we’re going to reach a point where anything that is not open looks second-class. Content mining is becoming more common and easier to do, as tools improve. We’re going to reach a point pretty soon where mining projects just shrug and limit themselves to OA articles, rather than burning a lot of time on negotiations. In that world, no-one is going to want to let their work go into the shrinking segment that prevents its re-use. I think it’s pretty much inevitable now that we’re heading for a world where only a tiny number of toll-access journals can survive: presumably those old stand-outs Science, Nature and Cell. And even they won’t be able to hold out indefinitely.

We’re aiming to add API services to our datasets early in 2015, so anyone can help themselves. We’re also re-engineering our production processes so we can capture our books and journal content in a form suitable for text-mining – which we will also offer for free (despite the cost of getting there!). Personally, I think relatively few will actually data or text-mine and there’ll be a much larger audience wanting ‘finished’ publications and data in a form that is dead-easy to read, understand and re-use. Nonetheless, I agree that it is unlikely that many toll-access journals will survive unless they adopt a freemium model – the policymakers and funders won’t stand for seeing the result of what they fund only locked behind paywalls – funders are the driving force, I don’t think the drive for OA is coming from authors.

Like you, I think the demand for text- and data-mining is somewhat overstated, though history will be the judge of whether these methodologies ever become anything more than a niche area of research.

To address Mike’s point above, if you’re doing a text-mining project and you just shrug and ignore large swaths of the literature because you may have to work to gain access to it, that may result in poor quality results. If I want to look at the last century of research on diabetes, to offer a comprehensive answer, I need to look at all the research, not just the small portion that I can get easily and inexpensively.

“To address Mike’s point above, if you’re doing a text-mining project and you just shrug and ignore large swaths of the literature because you may have to work to gain access to it, that may result in poor quality results.”

It is certainly true that you’ll get better quality results if you can include everything. But as always the issue is one of practicality. If you have a year’s funding, you have to decide how much of that year you want to spend negotiating licences and how much doing the actual work. Time was when the only realistic option was to dedicate a significant proportion of the time to negotiation — maybe more than half. That’s changing– partly because publishers including Elsevier are thankfully becoming more transparent about their terms, but more importantly because of the growing predominance of OA articles. The time will come where for at least some mining projects, it makes more sense just to shrug, use what’s clearly and freely available, and devote all the project’s available time to actual work instead of arguing with publishers and their lawyers.

How would you set up controls for such an experiment and the biases introduced when only looking at a selected portion of the literature that happened to be published in a particular manner?

I would suggest though, at least in the US where it would be fair use, or in the UK where it is explicitly granted under copyright law, that the licensing and lawyers are less problematic than is gaining bulk access to the entirety of the literature. That’s more a technological problem (and one that is currently being addressed) than a legal one.

I have no idea how, in general, you’d control for biases — it would obviously have to be designed as part of the individual study. For many studies, depending on how much effort is available to be expended, it might just be best to state up front what was done, and let the ashes fall where they may.

People like Peter Murray-Rust, who have done a lot of text-mining at the coalface, if you will, say there is no technological problem with downloading articles for text-mining, except where publishers impose one.

Certainly PLOS doesn’t feel the need to throttle or otherwise impede downloads: “Downloads that result from crawling and content mining contribute a trivial amount to the overall traffic at one of the largest Open Access publisher sites and are irrelevant compared to other sources of traffic. This is true both of average traffic levels and of unexpected spikes.”

This does look like a situation where the most helpful thing publishers can do, if they genuinely want to enable the mining that UK law explicitly allows, is to do nothing. Just let people get on with downloading through the usual channels.

Convenience samples do happen, they just tend to reduce the meaning one can derive from the conclusions.

In any research effort, there are costs involved, both financial, and in terms of effort and time. If you are willing to shrug and put in less time, money and effort, it is likely that cutting corners like that will be reflected in what you get out of your studies. It is often hard to get hold of human tissues to study a disease condition. It is expensive to maintain a mouse colony. These are the costs of doing business.

It’s interesting that the safeguards built into the system to thwart alleged nefarious behavior are coming into play here though. Most systems send up an alert when there is a massive amount of downloading, as in the past it usually signaled that someone was up to no good (an attempt to create an illegal mirror site for example). Here we do need to adjust our systems to better understand that there are legitimate needs for these types of behaviors and to differentiate research from malfeasance. Right now that’s at a primitive level of “ask permission and we’ll allow it” but hopefully we can do better in the near future. As you point out, a big advantage of OA is that one doesn’t have to spend any money on such safeguards, as anyone is allowed to do anything they want with the content.

As Mike points out above, the statement is a tautology. A world with no toll access is not a world with some toll access, and a world with some toll access is not a world with no toll access.

That said, one could argue that many of the current arrangements offered by publishers could be classified as “freemium”. Many papers are freely available on preprint servers like arXiv and biorXiv, and then a value added, peer reviewed and edited version is available for a fee in a journal. Similarly, the Green OA path that is common in many journals, particularly for NIH funded articles, offers a premium version available immediately to those willing to pay, and a delayed version that is free for those willing to wait.

In theory, Green OA is a form of freemium, but it’s not really because with Green OA there is no connection between the free service and the premium. In Freemium, financial sustainability is predicated on the conversion rate from free to premium and therefore on the upgrade journey from free to premium. With Green OA, the publisher has no control over the free version and therefore cannot build an upgrade journey for users. I also think that Green OA is fundamentally flawed because it presents a terrible user experience for readers (there is rarely a link between the free version and the VOR and free readers may have to wait months for embargoes to expire) and still locks the VOR behind a paywall – with Freemium, at least our version of it, the VOR is accessible to everyone on publication.

Hence the value of CHORUS (http://www.chorusaccess.org), driving Green OA in context, in the journal and in the case of many journals, making the Version of Record free, and in the case of others, the Author’s Manuscript alongside a link to the Version of Record.

I wrote about this confusion of goals almost exactly a year ago: http://scholarlykitchen.sspnet.org/2013/11/11/open-access-on-the-sea-of-confusion/, noting that this kind of confusion is common in social movements. As I put it then: “Social movements often depend on grand sounding but poorly defined concepts. The fact that the bedfellows are actually speaking different languages goes unnoticed and helps them get along and drive initial progress. It is only when the time comes for actual action that the confusion surfaces and that is where we are today.” I do like Rick’s point that authors may ultimately make the difference, voting with their submissions, as it were.

However, the reason why US Public Access policy does not include copyright issues is not because of any policy decision. The Government’s position, as I understand it, is that they have a “government use” license to the accepted manuscript of every article that reports on research which they funded. Government use includes making the article public, but it does not include making it usable by others. Hence the copyright issue simply does not arise, because the Government has nothing to convey, or to require to be conveyed.

Except of course for the interesting issue as to whether the US Government actually gets such a license, for articles they do not pay for the writing of and which are not specified in the research contract. I have been told by government lawyers that this is an unwritten rule, but the concept of unwritten copyright law is unclear to me, to say the least. To my knowledge the Government claim has never been tested in Court.

I suppose this bit of the correspondence is going on rather a long time so if I am suppressed I cannot complain but I want to return to the assertion I was commenting on. It was from David and read:
“The thing to consider in these discussions is the motivation for a publisher to engage in lengthy and expensive legal actions on behalf of authors. Under traditional copyright, the publisher has a financial interest in enforcing the terms of the copyright and requiring licensing of material. Hence they will send in the lawyers if someone is in violation” Kent has come across such cases but he looks through the lens (in my view) of the tiny number of very high profile medical journals. I know David now works for OUP but I used to run the OUP journals list as it was then. We did not send in the lawyers and at the other publishers I worked for did not do this.

For the record, OUP is currently involved in a long-running lawsuit over a university allegedly violating copyright in its reuse of scholarly materials (http://scholarlykitchen.sspnet.org/2014/10/28/feeding-the-goose-thoughts-on-fair-use-and-the-gsu-decision-2/). And oddly enough, this past week I had to deal with the lawyers over claims made that a journal article violated someone else’s intellectual property. As Ivan Oransky and Adam Marcus recently wrote, lawyers and legal issues are becoming ever more common in academic publishing:
http://www.labtimes.org/labtimes/ranking/dont/2014_06.lasso

While what OA advocates want is certainly interesting, what is actually going to happen is perhaps far more so. At this point I do not see OA going much further as far as free immediate access to non-APC VOR content is concerned. The statements by advocates that success is somehow inevitable notwithstanding. A lot is becoming more open in a lot of interesting different ways, but few of them meet the purist’s standard and I doubt they ever will. Progress is far more important than purity.

What makes you think that there is a consensus on the Berlin Declaration as the only acceptable definition of OA/ Where’s your evidence? There is a long pre-history of the OA movement extending back decades before the Budapest or Berlin statements. We didn’t use the language of OA. but the intent was clear. (I trace some of this history in an article forthcoming in the Journal of Scholarly Publishing.) The Budapest and Berlin statements focus only on STEM journal publishing, and they are therefore self-limiting in that respect. They have nothing to say at all about OA as it applies to monograph publishing, and the history of that part of the OA movement clearly fors NOIT accept the CC BY license as representing any consensus. In fact, if you take a look at the monograph publishing operations (including Toby’s one at OECD), they are much more in line with the CC BY-ND-NC license. That is what the National Academies Press did, it is what we did at Penn State, and it is explicitly the default at the new Amherst college Press. It also does not properly apply to all Gold and Green OA article publishing. Except where the funders mandate use of the CC BY license, I feel confident that the publishers use the CC BY-ND-NC approach instead. And so do many institutional repositories. E.g., the one at Penn State (where I have over 80 of my own articles posted), the default license is CC BY-ND-NC. So, I think it is a vast exaggeration to say that there is such a consensus. If it exists, it does so only for STEM journal publishing and only for publishing where funder mandates require it.

What makes you think that there is a consensus on the Berlin Declaration as the only acceptable definition of OA/ Where’s your evidence?

I had hoped that the carefully qualified language I used in my first paragraph would make it clear that I’m making a provisional assessment of a “hardening” (rather than absolute) consensus, and that my assessment is subjective and based on my own experience in the OA conversational space.

That said, you’re right that the Budapest and Berlin statements focus on journal publishing in the STEM disciplines and have nothing to say about monograph publishing. The OA movement itself has focused primarily on STEM journal publishing for at least the past ten years or so, and has largely ignored scholarly monographs. A debate about whether or not that’s appropriate would probably be a better fit in a forum other than this comment string.

Was it appropriate for the Budapest and Berlin declarations to concentrate on science journals? Yes, because it’s what the framers knew and understood. Is it therefore appropriate to extend those principles to books and the humanities? No — and I admit it’s a trap I easily fall into. I think the great majority of what’s been written about OA for science journals also applies for humanities journals, but we can’t assume a priori that it all does. For books, things are more complicated still (though I suspect nowhere near as complicated as some traditionalists make them).

Sandy, can I give a concrete example of the downside of “ND”? There are at least two typefaces that make it easier for dyslexics to read text (you can google Dyslexie and OpenDyslexic). Would it not be nice if all OA access content were available in this more “accessible” typeface too? Well, technically, if your articles were uploaded in XML (almost all publishers now create XML but almost none publish it), a third party could press a button and create a PDF with a dyslexic typeface. But “ND” means legally they can’t do that as it is a “derivative”…

Thought I’d share an anecdote. Last year, on behalf of UNESCO, we published a book under UNESCO’s CC-BY license in keeping with their OA policy. About nine months later we get a call from the authors about a ‘pirate’ Kindle version of the book that is available for sale via Amazon (the vendor is an entity called Amazon Digital Services, Inc). This version had a different cover and the inside layout was pretty awful. It looks like a machine had scraped the content from the PDF e-book and roughly laid it out again for the Kindle (we didn’t issue a version suitable for Kindles). The authors, who are not UNESCO staff, asked us to contact Amazon and insist that it be removed from their catalogue . . . which, of course, we couldn’t do because it had been published under UNESCO’s CC-BY license. What’s ironic is that the Amazon version has kept the large OA logo that UNESCO stuck on the PDF version that appears on their website (it is not on the version that we published). Makes me wonder if Amazon has realised that there might be money to be made by harvesting content from institutions that have adopted CC-BY OA policies.

Yes, CC-BY is an invitation to mischief of the type you describe here. Why Creative Commons and others are pushing it over CC-BY-NC seems to be capitulation to the wishes of the publisher lobby. What was intended to be de jure open access can be converted to a de facto closed access through the use of optimized discovery tactics, including advertising and DRM. CC-BY is the key to that conversion. Plunket of Tammany Hall would have called this “honest graft.”

Is it “mischief” if it is doing exactly what the license was designed to do? Either use it or don’t use it, but don’t use it and then complain that someone does exactly what you told them they could do.

The call to mischief is not in the existence of this option. Rather, it is in Creative Commons favoring it and characterizing CC-BY as a “Free Culture license” while labeling CC-BY-NC and all other CC licenses as “not a Free Culture license.” This is newspeak of the worst kind.

Comments are closed.