English: Rough seas
English: Rough seas (Photo credit: Wikipedia)

The Open Access (OA) debate is a sea of confusion. These confusions take many forms but the deepest by far is the large number of contradictory models that fall under the term “open access.” When different people use the same words to mean very different things confusion is inevitable. Here is a short list of some of the many OA models that I have observed in the mix:

  1. Freely available journal paid for by author publication charges (APCs).
  2. Free available journal with no APCs, paid for by institution or funding agency grant.
  3. Immediate deposit in a repository, or web posting of freely available article which also appears in a subscription journal.
  4. Immediate deposit in a repository, or web posting of freely available article with no subsequent publication in a subscription journal.
  5. Delayed free access to the article in a journal after an embargo period.
  6. Delayed free access to the article in a repository after an embargo period.

And that’s just a sampling that addresses the question of access. Combine those with all of the different views on copyright and licensing for reuse, and you have an extremely complex matrix. Note that every one of the access models listed above already exists to a significant degree. Plus there is a lot of experimentation going on, which is good, even exciting.

The confusion comes into play in the policy domain where many voices are calling for government mandated open access. But different people are actually calling for very different things using the same language. The result is a great policy muddle.

There is a proper use of the term “open access” which is to refer to the social movement calling for it. But using that term as though it referred to an actual goal or outcome is hopelessly ambiguous and merely adds to the confusion. Using terms like gold OA and green OA does not resolve this confusion because there are not just two kinds of OA, as the incomplete list above makes clear. For each item on our list there are many possible varieties, so a full listing would be enormous.

The point is that it is not clear what we are talking about when we talk about OA and that needs to be recognized. In my taxonomy of confusions open access is what I call a poorly defined concept. This is arguably the worst confusion one can have because it infects everything that is said. That is why it is in the upper left hand corner of my matrix.

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.


38 Thoughts on "Open Access on the Sea of Confusion"

For what it’s worth, the OSTP memo strenuously avoids the use of the term “open access”. It does not appear a single time in the document. Instead, they use the term “public access”. I’m not sure if that alleviates the confusion or just adds another layer to it.

  • David Crotty
  • Nov 11, 2013, 7:10 AM

I think OSTP recognized that OA is a political movement so avoided the term. Their term “public access” is not defined, except by the mandate and that is ambiguous as between items 5 and 6 on my list. CHORUS speaks to 5 while SHARE and PMC speak to 6. Where we will end up is anyone’s guess and there are many battles to be fought along the way. But at this point I do not see “public access” becoming a term of art the way “open access” is.

On the copyright side I am still very curious about the legal basis for the government’s apparent claim to a perpetual license to anything ever written by a researcher based in any way on their funded research. That concept is truly murky and OSTP does not address it. Perhaps it will be tested and defined someday. There must be some limits to the government’s reach.

  • David Wojick
  • Nov 11, 2013, 7:49 AM

“On the copyright side I am still very curious about the legal basis for the government’s apparent claim to a perpetual license to anything ever written by a researcher based in any way on their funded research.”

The OSTP call for public access to papers resulting from funded research is not a copyright issue. Rather it is a condition of receiving funds.

  • Dan
  • Nov 11, 2013, 4:27 PM

If the government is imposing a future license as a condition of funding then that is a copyright issue. The question is whether the government can do that legally? It is an incredibly far reaching license since it apparently covers everything I write for the rest of my life that is based in any way on the research I once did. This looks like taking property from me without compensation, which is illegal, especially if the government then gives it away, which is damaging.

So far as I know none of the non-NIH agencies has these terms in it’s present contracts so they will have to make new rules to do so. That will be an interesting rulemaking.

  • David Wojick
  • Nov 12, 2013, 7:19 AM

Here is the simplest and most straightforward model of Open Access (not mentioned by you).
I never waited for publishers, universities or anyone else, but simply went ahead and paid everything myself out of my own pocket. I still do: http://www.sexarchive.info

  • Erwin J. Haeberle
  • Nov 12, 2013, 5:07 AM

Yours is a web archive so falls under my items 3 or 4 depending on the content. It may be different if you are collecting the content rather than having authors deposit it. My list is confined to journal type articles. If you are posting journal articles how do you handle copyright?

  • David Wojick
  • Nov 12, 2013, 7:07 AM

I do carry journal articles, encyclopedias, textbooks, books and more. Just take a look.
In each case, I obtained the permission of the copyright owners. My MOOCs I wrote myself.

  • Erwin J. Haeberle
  • Nov 12, 2013, 9:50 AM

I suggest a very common type would be a “3a:” Author-paid, freely available article in a subscription journal. My journal offers this because some of our valued authors are under an OA mandate. We do NOT encourage posting the article elsewhere, because (1) it complicates tabulating page views, and (2) it does not automatically link to errata and discussions.

  • Ken Lanfear
  • Nov 11, 2013, 9:36 AM

Of course, the hybrid! Good call. Rats; I was looking for seven and missed the hybrid journal. The hybrid should really be listed as number 2 since it is a variant on the APC model.

  • David Wojick
  • Nov 11, 2013, 12:08 PM

What confusion? ‘Open Access’ is not confusing, it’s ‘Open Access Plus or Minus Whatnot’ that is (OAPMW). I am pretty sure you can create a mighty confusion about anything, if you try hard enough. The ocean of confusion about what ‘publishing’ is, for instance.

  • Jan Velterop
  • Nov 11, 2013, 10:50 AM

The confusion lies in the fact that when two people are conversing and using the phrase “open access,” neither one can safely assume that the other is talking about the same thing. For some people, a document isn’t truly “OA” unless it comports with all criteria laid out in the Berlin and Barcelona statements (including CC-BY licensing). Some people assume that “OA” means Gold; others assume it means Green. Others know (and care) little about funding mechanisms, and use the term simply to mean “freely available to the public for reading.” People regularly refer to OA “mandates” that are not, in fact, mandates at all. And there are other variant understandings of the term as well, as David has laid out. This all does indeed lead to confusion. I don’t know if it’s an “ocean” of confusion, but the ambiguity is real and makes the conversation more difficult than it should be.

  • Rick Anderson
  • Nov 11, 2013, 11:59 AM

The Barcelona Statement. A new one to me. Where do I find it? No wonder some people find it all rather confusing. Open Access Statements being produced out of a top hat. And why is funding important for understanding what open access means? So yes, there is a lot of misunderstanding about the waffle around open access, but that’s self-inflicted by the wafflers. To make the conversation less difficult, when talking about open access let’s talk about access that’s open. No more, no less. How it got to get open is of no consequence to the end-state: open.

If you want to talk about business models that may make open access possible, by all means, go ahead. And if there is confusion about that, well, it’s confusion about business models, not about open access. The fact that some people take pleasure in calling just about anything that’s not pay-walled publishing ‘open access’ surely isn’t the fault of those who first defined what open access was and started to advocate its benefits to the scholarly community.

The situation is not all that different to what the meaning of ‘publishing’ is. Many different definitions and perceptions. Unfortunately, ‘publishing’ was not neatly defined when the word entered the lexicon. Etymology dictates that it has to do with ‘making public’. Open access is making scholarly literature more public, so open access should be more entitled to be called ‘publishing’ than making it available upon payment only, no? By the way (in response to David Wojick), open access publishing can be just as ‘formal’ as pay-walled publishing. The difference is not the formality. The difference is openness of access.

  • Jan Velterop
  • Nov 11, 2013, 2:57 PM

The Barcelona Statement. A new one to me. Where do I find it? No wonder some people find it all rather confusing.

Ha! Yes, you caught me. I got my Budapest mixed up with my Barcelona. Too many OA statements coming out of too many cities starting with B. (Berlin, Bethesda, Barcelona — I propose that the next one come from Bruay-en-Artois).

To make the conversation less difficult, when talking about open access let’s talk about access that’s open. No more, no less. How it got to get open is of no consequence to the end-state: open.

As soon as you get the OA community at large to agree with you on that, the sooner the ambiguity will end. Please note that I’m not celebrating this situation, only pointing out that it exists. It exists not only because different people and groups understand OA differently, but also different people and groups want OA to mean very different things and are very invested in their own definitions and standards. Getting any of them to give up their definition in favor of someone else’s is going to be a challenge. You and I can agree that this is a regrettable state of affairs, but saying “the problem doesn’t exist because it shouldn’t exist” isn’t going to make it go away.

  • Rick Anderson
  • Nov 11, 2013, 5:39 PM

(Berlin, Bethesda, Barcelona — I propose that the next one come from Bruay-en-Artois).

Oops, did it again — typed “Barcelona” when I meant “Budapest.” I plead Monday Brain.

  • Rick Anderson
  • Nov 11, 2013, 5:41 PM

I think you’re right in saying that “different people and groups understand OA differently, but also different people and groups want OA to mean very different things and are very invested in their own definitions and standards.” I don’t think, however, that OA advocates (and I’m not sure if it is those that you mean by ‘the OA community’) understand open access an anything other than access that’s fully open. In my perception there are indeed people who want OA to mean different things, but those people are mainly the ones who benefit – or believe they benefit – from OA not to be straightforward and clear cut. Such as those who like to have more ‘control’, post-publication, than the author has with CC-BY (i.e. the requirement for acknowledgement) and want to retain vestiges of the subscription system, such as NC and ND. NC and ND are not even of much practical value, but they *are* in practice formidable impediments for moving modern science forward, dependent as it is ever faster becoming on machine-supported meta-analyses of large numbers of papers, as simply, humanly, reading them is not feasible anymore in many subdisciplines. The meta-analyses may point to a few articles that do need to be read in detail, but note the ‘few’ in this sentence.

An interesting point about NC not being of practical value: if a CC-BY-NC licence is issued by the author, then why would the publisher be the only one able to use the article commercially? For that to be the case, the author must issue a special waiver for the publisher to be able to, for instance, sell reprints, and I doubt if that happens in most cases. And, anyway, the author can at any time waive the NC stipulation for anybody else, or for the world at large, for that matter. In other words, CC-BY-NC is no guarantee to the publisher to have exclusive commercial rights.

  • Jan Velterop
  • Nov 12, 2013, 7:49 AM

Who says the CC-BY-NC license has to be one issued by the author? It can just as well be issued by the publisher, after a contract has been signed with the author. The practical importance of NC and ND is illustrated by the case of translation. Scientists may not care so much because they speak in mathematical language that is universal and needs no translation, but humanists do care whether their work is poorly translated and CC-BY alone gives them no control over the quality of translations that others make of their work.

  • Sandy Thatcher
  • Nov 12, 2013, 10:24 AM

Technically, the licence can be issued by the © holder, so if the author has first transferred © to the publisher, the latter can issue the licence. Maybe that happens. Now and then. I don’t quite see how NC can prevent bad translations, but perhaps ND can, since it can prevent any translation. (I wonder how well most publishers can judge the quality of a translation, though, especially a translation of difficult conceptual material, and not the relatively ‘simple’ language of science. I wonder also how robust the publications in English are by non-native speakers, as not all publications are written by native speakers of English. The correctness of English can perhaps be judged, but whether it is a faithful representation of the author’s intention in the case she is Japanese, for instance, is anybody’s guess. Reading a book like “Is That a Fish in Your Ear?” by David Bellos is recommended.) There may well be some truth in the idea that for humanities, only ‘ocular access’ and not machine-assisted analysis is important (for which ND might be an impediment). I very much doubt it, though.

As said earlier, NC and ND are throwbacks to the comfortable old days when publishing was still simple, on paper, and on subscription.

  • Jan Velterop
  • Nov 12, 2013, 5:55 PM

There is indeed a formal definition of OA that a small, self-appointed group has declared to be the only acceptable definition, demanding a monopoly on use of the words “open access”. It is unclear how well that group represents or that definition serves the needs of the research community as a whole (see: http://scholarlykitchen.sspnet.org/2013/08/06/is-access-to-the-research-paper-the-same-thing-as-access-to-the-research-results/). It may not be possible for one inflexible definition to serve the needs of all researchers in all diverse fields. Stephen Curry states in a recent post (http://occamstypewriter.org/scurry/2013/11/03/open-access-headaches/) that the BOAI recognizes more than one variant of open access: “The BOAI 10 text explicitly recognises that some variants of open access are more open than others but recommends a pragmatic, can-do approach.”

The confusion results from a difference between one group’s formal definition of the term and common usage of the term by the broader community. It may seem simple and black/white to you, but it is not possible to control an evolving language. One can pedantically try to shout down anyone who uses the term in a manner unacceptable to your definition, but it’s unlikely that over the long term this will be effective.

As for the different flavors of CC licenses, it makes sense that researchers, whose jobs exist in something of a reputation economy, prefer to exert a certain level of control over how their names, reputations and writings are reused by others. NC, ND versions of licenses don’t prevent commercial reuse, they just allow the author to have a say in allowing those reuses. It’s likely why when given the choice, researchers overwhelmingly choose more restrictive licenses. It’s similarly understandable that a company like yours, which seeks to profit from reuse of the writings of others, wishes to reduce restrictions on those reuses and streamline the discovery process. The open question is whether both of your needs are best served by the very blunt “anything goes” CC-BY license, or if there instead could be better licensing terms tailored to meet the needs of both parties. CC-BY does offer a useful open-endedness, coverage for any unexpected future uses that may arise, but at the same time is open for potential abuse.

  • David Crotty
  • Nov 12, 2013, 11:47 AM

Concerning “when two people are conversing and using the phrase ‘open access,’ neither one can safely assume that the other is talking about the same thing.” The same could be said of traditional subscription sources and their plethora of models. There are many traditional models that I have observed in the mix:

* Journals that are available in aggregators such as EBSCO.
* Some journals within the aggregator that has different use terms, such as HBR.
* Journals that are print only, print+online only, online only, some that give the subscriber a choice, but it may be 90% of the cost or more expensive, depending upon the publisher.
* Journals that are available in big deals that we have access to, but we don’t really subscribe to them.
* Magazines that come with an institutional membership, but we really just want the journal, not the magazine.
* Journals that split into two titles, making the second title free for a year, hoping that the institution pays for the free journal for the second year.
* Journals that have different pricing tiers based on the number of Ph.D. candidates (AGU).
* Journals that have different pricing tiers based on Carnegie classificaiton.
* Journals that have different pricing tiers based on the number of FTE in the whole institution.
* Journals that have different pricing tiers based on the number of students in a program.
* Journals that require the librarian to contact the publisher for a quote, where is no list price anywhere, and the price must be negotiated.
* Journals that are split into a dozen or so parts, and the volume numbering for the parts of the journal skips and jumps. (BBA, Elsevier points to a CDL decoder chart for it. http://www.cdlib.org/services/collections/chooseitem/bba.html)
* Journals that have split and merged so many times, one needs a flow chart to keep track, such as http://www.lib.utexas.edu/chem/info/jchemsoc-chart.pdf or http://library.caltech.edu/collections/rpb/chemistry/ComptesRendus.pdf
* Journals that charge differential prices for institutions vs individual subscribers.
* Journals that are treated like books by many libraries, such as annual reviews.
* Journal volumes that are indexed by databases, but the title is considered to be a series title instead of having a title for the individual volume, thus making the volume difficult to find. (SPIE conferences or the Springer LNCS series)
* Conference proceedings. Are they books or journals?
* Standards that come out irregularly. They are periodicals, but not really journals.

And, then there are the models for ebooks and their platforms. Don’t get me started on that. You see, understanding the models that publishers put out is complicated. The Open Access models that we have are not really all that complicated.

  • Joe Kraus (@OAJoe)
  • Nov 12, 2013, 1:51 PM

Concerning “when two people are conversing and using the phrase ‘open access,’ neither one can safely assume that the other is talking about the same thing.” The same could be said of traditional subscription sources and their plethora of models.

You bet. That’s why I would never say anything like “scholarly publishing is not confusing” (à la Velterop), nor would I offer a complicated menu of subscription-publishing variants and give it the heading “SIMPLICITY ITSELF” (à la Harnad). To say that the ongoing OA conversation is characterized by confusion and ambiguity is not to imply that the general scholcomm conversation is a model of clarity and straightforwardness.

  • Rick Anderson
  • Nov 12, 2013, 2:00 PM

I never said that “scholarly publishing is not confusing”. It patently is, particularly to scholarly kitcheners, it seems. What I did say is that the notion of Open Access is clear. All manner of ‘flavours’ of OA are nothing but vain attempts to preserve the old order of print, nothing but trying to keep a grip on e-publishing by preserving the old system in the form of print-analogues. Publishing is yet to enter the age of the internet seriously.

  • Jan Velterop
  • Nov 12, 2013, 6:09 PM

I never said that “scholarly publishing is not confusing”

No, you said “Open Access is not confusing.” To say “scholarly publishing is not confusing” would be to make a claim for scholarly publishing in the manner of (“à la”) your claim for OA.

In any case, though, whether or not OA itself is confusing is a red herring. My point is not that OA itself is confusing, but that the public conversation about OA is characterized by confusion. And you seem simultaneously to agree (“there are indeed people who want OA to mean different things”) and disagree (“what confusion?”), thus handily illustrating one facet of the problem.

  • Rick Anderson
  • Nov 12, 2013, 6:34 PM

Jan, I am fascinated that your concept of OA is different from my entire list. I am referring specifically to the present public policy issue, where governments are struggling to enact OA mandates, not to the broader issue of where publishing is going. Apparently your policy preference is “none of the above” as far as my list is concerned, which is not helpful. Or perhaps you do not wish there to be government mandates. What is your policy position? (This is the question that resolves the confusion.)

  • David Wojick
  • Nov 13, 2013, 12:28 PM

There certainly is great confusion about publishing. In fact the OA confusion is part of it so the publishing confusion is arguably bigger. This is the standard confusion of technological revolution, which I first described way back in 1973 in an essay entitled “The Structure of Technological Revolutions.” During technological revolutions central concepts often go out of focus and need to be rethought, because they embody assumptions that are no longer valid.

It is Chapter 14 in this book (available used for 95 cents and worth it) —


Simply put the Web has raised the issue whether formal publishing is really still necessary. Some OA advocates claim it is not. This is reflected in my item 4.

  • David Wojick
  • Nov 11, 2013, 1:29 PM

I am confused! This post made me realize that I knew so little about this topic. Now I am doing my research, and I feel more prepared knowing that there is ambiguity on the meaning of the term “open access.” For us younger authors, this ocean is quite new; we are up against a lot of complexity and confusion.

  • laurabrimberry
  • Nov 11, 2013, 12:23 PM

Actually Laura, once you see the confusion you are unlikely to be confused by it. You might even find it entertaining in a perverse way. But then concept confusion is my field so I am fascinated by it. It can be painful to watch.

  • David Wojick
  • Nov 11, 2013, 2:21 PM

The confusion is not going to get better. Publisher “apps” are well into development and testing stages. You will be able to link to parts of the article, such as figures, and perhaps even run interactive elements. The licensing/permission/credit issues make my head hurt!

One outcome of the “apps” approach is that the easily copied pdf may become a relic. You will need the publisher’s web services to fully utilize the article.

  • Ken Lanfear
  • Nov 11, 2013, 1:50 PM

Yes the “article” is another central concept that may be going out of focus in the techno revolution. But maybe not because expressed thought has a complex structure that is not easily subdivided. Of course one can always pull figures out but that is not new, nor do the captions always stand alone. I think integrating articles into bodies of thought is more interesting and promising than tearing articles apart. But then that is the area I work in, complex issue analysis and linking.

  • David Wojick
  • Nov 11, 2013, 2:29 PM

Our world is a complex, uncertain, and ever-changing process.

David, when I read your sentence “social movements often depend on grand sounding but poorly defined concepts” I thought, well, doesn’t this increasingly apply to more traditional publishing models as well? Isn’t this a foundation of all social process, an ongoing (re)definition of concepts?

Maybe this process of uncertainty is recursive and cannot be pinned on one party or sub community. Are the cracks of uncertainty, or confusion if you will, there for reasons beyond the movement that poked them open?

In this sense, a crumbling definition of traditional movement is maybe as much the driver of poorly defined new concepts, as the is the new movement?

I’m not sure I’m presenting my thoughts well, but I think of this as ongoing and recursive process rather than a one-in/ one-out type of replacement.

  • Neil Christensen
  • Nov 11, 2013, 3:39 PM

Agreed Neil, confusion is the price of progress. It is the old concept that goes out of focus, as it were. The new concepts are poorly defined precisely because they are emerging. I am not criticizing OA, merely pointing out the present stage of development, especially the policy muddle. What lies ahead we do not know, as the final form of our new concepts is unpredictable.

A glimpse of theory may help. My working definition of a concept is adapted from philosophy of language. A concept is what one has to know or believe in order to use a word correctly. Thus a concept is a body of knowledge and belief. The words we use have the meanings they do because we are trying to say what’s true. So in an important sense it is not up to us what our words mean.

But beliefs change and therefore so do concepts. These changes can be wrenching because beliefs and concepts come in systems, which from time to time get overthrown. We call these upheavals revolutions. Note that scientific and technological revolutions are just special cases which have an especially physical origin. As Joe has pointed out quite eloquently, traditional publishing is already long gone. We are well through the looking glass; the question now is where we will come out? In the meantime our language is necessarily confused.

  • David Wojick
  • Nov 11, 2013, 10:00 PM

Perhaps “open access” is what philosophers have called an “essentially contested concept”: http://en.wikipedia.org/wiki/Essentially_contested_concept. Such concepts are “concepts the proper use of which inevitably involves endless disputes about their proper uses on the part of their users,” and these disputes “cannot be settled by appeal to empirical evidence, linguistic usage, or the canons of logic alone.”

  • Sandy Thatcher
  • Nov 12, 2013, 1:17 AM

An interesting possibility Sandy but at this point the OA issues seem to be resolving themselves, albeit slowly. “Gun control” would be an example of a contested concept and these are typically politically paralyzing because the people are divided on the central beliefs. Mind you if Congress has to resolve the US OA issue it could bog down. This may happen because it is not clear that the non-NIH agencies have the statutory authority to implement the OSTP mandate.

  • David Wojick
  • Nov 12, 2013, 10:17 AM


1. Gold OA
2. Gold OA
3. Green OA
4. Self-archiving of unpublished work (not at issue)
5. Delayed Access (Gold)
6. Delayed Access (Green)

All simple: OA is immediate free access. Author provides it: Green. Journal provides it: Gold. Gold OA is Gold OA whether or not author pays. And Delated Access is Delayed Access, whether Gold or Green.

(And then there’s the Gratis/Libre distinction, which also comes in Green and Gold, paid and unpaid, immediate and delayed: Gratis means free online access; Libre means free online access plus some re-use rights.)

All of this concerns peer-reviewed, published research in the first instance — perhaps also eventually books, data and software. But what authors do or don’t do with unrefereed, unpublished content is not an OA concern.

  • Stevan Harnad
  • Nov 11, 2013, 8:24 PM

By your definition, the Wellcome Trust’s Open Access policy and the RCUK’s Open Access policy are not “open access” because both allow for embargo periods. Perhaps both should be advised to change their terminology.

All of this concerns peer-reviewed, published research in the first instance — perhaps also eventually books, data and software. But what authors do or don’t do with unrefereed, unpublished content is not an OA concern.

Should datasets be peer reviewed as well? Must the data be part of a published paper to make its release valuable to the research community?

Perhaps not so simple after all.

  • David Crotty
  • Nov 11, 2013, 8:54 PM

Stevan speaks his own private language so is an excellent example of my basic point. I noticed this when debating him several months ago in another forum and even mentioned it at the time.

  • David Wojick
  • Nov 11, 2013, 10:04 PM


Jointly coined and defined OA with Peter Suber and the other BOAI signatories in 2002.
Coined Gold and Green OA in 2004.
Refined and updated the BOAI definition (Gratis & Libre OA) with Peter Suber in 2008.

All done publicly, in the interests of the research community and the tax-paying public.

Some private (publishing) interests have tried to redefine OA as paid Gold OA, and Green OA as embargoed access (while themselves imposing the embargo).

It is inevitable that research and public interests will prevail.

  • Stevan Harnad
  • Nov 12, 2013, 11:14 PM

What about publishers that increase the confusion about open access by trying to develop their own CC BY license (Elsevier) or by claiming they are open access although the author has to transfer the copyright of his or her paper to the publisher (IEEE)?

  • Leon Osinski
  • Nov 12, 2013, 10:50 AM

Comments are closed.