Definition: "experiment"
Definition: "experiment"

“experiment, noun. An action or operation undertaken in order to discover something unknown, to test a hypothesis, or establish or illustrate some known truth” — Oxford English Dictionary

This week, Cornell University joined four other universities in pledging to provide funds for open access publishers.  The pledge, described as a “compact,” commits institutions to

underwriting reasonable publication charges for articles written by its faculty and published in fee-based open-access journals and for which other institutions would not be expected to provide funds.

The details of this compact are important, although it leaves specific decision-making to each member institution. For example, it does not specify what is meant by “reasonable” charges. It is also designed to be administered as a fund of last resort: faculty are expected to utilize other sources of funding before applying for institutional reimbursement.

The argument for creating these funds is clearly about fairness — as embodied in the name, “Compact for Open-Access Publishing Equity.”  If institutions support subscription-access forms of publication, they should also support free access through author payments.  Stuart Shieber and others have repeatedly used the phrase “leveling the playing field” to embody this principle.

The Cornell University Library issued its own narrative yesterday, which relies on different rationales.  It recognizes the valuable services provided by publishers in terms of managing the certification and distribution services, although it clearly sides with one particular business model, author-pays publishing.

The authors of the Cornell’s compact view their open access publication fund as an “experiment” and, given that this service is designed to help support scientists, this frame will undoubtedly be viewed as acceptable.

But is this really an experiment?  If the creation of a funding line to support a particular form of publishing is designed as a hypothesis, what result are they expecting?  What constitutes a successful or failed experiment?

  • Is the experiment a success when the funds are expended completely within a short period of time requiring the library to find additional publication funds or deny requests?  Or is that a failure?
  • Is money left over in the fund at the end of the fiscal year a success?  Or is that a failure, considering that these funds could have been used for purchasing other sources of information like books and journals?

The language of the Cornell compact is schizophrenic, using words like “compact” and “pledge” together with “experiment.”  Moreover, it still echoes the religious fervor and proselytism found early in the open access debate, as illustrated in last paragraph:

Be an Advocate
If you believe that publishing in open-access journals in order to provide unfettered access to your research is important, please let us know. We are looking for researchers who will encourage others to publish their work in this manner

We should view this compact in its simplest terms — a form of advocacy for a particular publishing model — and drop the scientific objectivity connoted in words like “experiment.”  If this is about fairness, let us talk about how this new model will help or hinder who is allowed to participate in generating new knowledge.  If this is about access, let’s talk about whether this type of publishing results in disseminating scientific results to more readers.  If this debate is about economics, let’s talk about whether Cornell and the four other signatory institutions will save money under this model.

Enhanced by Zemanta
Phil Davis

Phil Davis

Phil Davis is a publishing consultant specializing in the statistical analysis of citation, readership, publication and survey data. He has a Ph.D. in science communication from Cornell University (2010), extensive experience as a science librarian (1995-2006) and was trained as a life scientist.


8 Thoughts on "Cornell Open-Access Publication Fund"

Well stated Phil. It is interesting that they view this as a transitional process but no effort is apparently being given to support subscription journals with an author choice option. They seem to only want to support pure OA journals and not to help others make the transition from subscription to OA. Another aspect of their failed effort to promote a form of publication.

I take your points about semantics, but they work only if you insist on the formal definition of “experiment,” which can also simply mean, as you point out, “to discover something unknown.” Since when have arguments for open access been restricted by the scientific method? There has always been a moral dimension to them, as there should be (compare arguments for a public option in the health care reform debate). “Religious fervor” notwithstanding, in the end the success of this institutional “compact” will be determined by economics–influenced profoundly by the entrenched scholarly reward system perpetuated by these same institutions. This is where we find the real schizophrenia.


Phil raises the right questions:

“If the creation of a funding line to support a particular form of publishing is designed as a hypothesis, what result are they expecting?  What constitutes a successful or failed experiment?… If this is about access, let’s talk about whether this type of publishing results in disseminating scientific results to more readers.  If this debate is about economics, let’s talk about whether Cornell and the four other signatory institutions will save money under this model.”

It’s the usual conflation of the access problem with the affordability problem, as well as the conflation of the solutions: Gold OA publishing and Green OA self-archiving.

Open Access (OA) is about access, not about journal economics. The journal affordability problem is only relevant inasmuch as it reduces access; and Gold OA publishing is only relevant inasmuch as it increases access — which for a given university, is not much. Authors must be free to publish in their journal of choice. Most journals are not Gold OA journals. Nor could universities afford to pay Gold OA fees for the publication of all or most of their authors’ research output today, because universities are already paying for publication via their subscription fees today.

The only measure of the success of a university’s OA policy is the degree to which it provides OA to the university’s own research article output. By that measure, a Gold OA funding compact provides OA to the fraction of a university’s research output for which there exists an affordable Gold OA journal suitable to the author and the research. That fraction will vary with the institution, but it will always be small.

In contrast, a Green OA self-archiving mandate provides OA to most or all of a university’s research article output within two years of adoption.

There are 5 signatories to the Gold OA “Compact” so far. Two of them (Harvard and MIT) have already mandated Green OA, so what they go on to do with their available funds does not matter.

But the other three signatories (Cornell, Dartmouth and Berkeley) have not yet mandated Green OA. As such, their “success” in providing OA to their own research article output will not only be minimal, but they will be setting an extremely bad example for other universities, who may likewise decide that they are doing their part for OA by signing this compact for Gold OA (in exchange for next to no OA, at a high cost) instead of mandating Green OA (in exchange for OA to most or all their research articles output, at next to no extra cost).

What universities, funders, researchers and research itself need, urgently, is Green OA mandates, not Gold OA Compacts. Mandate Green OA first, and then compact to do whatever you like with your spare cash. But on no account spend it on funding Gold OA INSTEAD of mandating Green OA — not if OA is your goal, rather than something else.

See: “Please Commit To Providing Green OA Before Committing To Pay For Gold OA!”

A small thing, but I was curious about who or at what level are the signatories to the compact. It wasn’t easily possible to tell if this was the initiative of a specific department, or a campus-wide commitment on the part of the president or provost.

Also, while several universities have already provided fees funding for OA articles a year or two ago, including UCB, through their libraries or combination of adminstration/library, this particular initiative seems narrow in just what kind of OA it will support. Why would that be?

The decisions to join the compact were made at the provostial level.

The breadth of coverage of the funds is up to the individual universities (as are other issues of implementation), but the argument for the particular approach that Harvard and Cornell are taking is discussed in my background paper in PLoS Biology (

Phil, thank you so much for covering the Compact for Open-Access Publishing Equity and Cornell’s implementation of the agreement, signed by the Provosts of the participating institutions. In terms of the semantics of the COAP implementation of the Compact: when we called this an “experiment,” we did not mean a tightly structured scientific experiment. Even the OED first defines the term “experiment” as “The action of trying anything, or putting it to proof; a test, trial,” and this is the more open-ended spirit in which we are approaching the Open Access fund for Cornell authors.

Quite simply, the objective of the Cornell Open-Access Publication fund is to promote open access to the scholarly research output of Cornell University (regardless of the business model, which needs a community solution). As the signatory from Cornell, our Provost, says: “as part of its social commitment as a research university, Cornell University strives to ensure that scholarly research results are as widely available as possible.”

You express concern that we have decided in favor of author-pays OA before its impact can be understood — and suggest we are being disingenuous in not admitting the foregone conclusion. But you should note that the Compact is only one of numerous “experiments” in alternative publishing models that Cornell (and the other signatories) are pursuing. In the cases of arXiv, Project Euclid, the institutional repositories, our collaborations with the University Presses, etc., etc., we are making scholarly content available in various, disparate ways without knowing what the long-term impact will be on scholarly communications — which models will be taken up, which will fall away, which will coexist, etc. The OED also tells us that the word “experiment” is related to “experience”: without putting new models into practice, without gaining experience with different publishing models, there is no experiment and no concrete knowledge. The Compact facilitates an experience for scholars, libraries, universities, readers.

You certainly asking the right questions: will this or that model “help or hinder who is allowed to participate in generating new knowledge”; which will “disseminat[e] scientific results to more readers”; which will save money? These are questions that many of us have been thinking about for a long time. We should be maintaining an “experimental” outlook and making these considerations about all of our scholarly communications efforts. We need to consider liabilities as well as benefits — these will no doubt be present, in different combinations, in all of the models we consider.

Along with the objective of opening access to Cornell scholarship, I think a primary goal of our implementation of the Compact is to start a dialog among authors at Cornell about the true costs involved in publishing their work. At least initially, I think the success can be measured by the intensity of this dialog.

In terms of Stevan’s issue about “green” before “gold” — we’ve tried that at Cornell and have so far been unsuccessful in achieving a mandate. I’m hoping a successful dialog among our authors about the issue will enable progress toward this goal as well.

Thanks for your thoughtful reply. It is not required to take the word “experiment” in the strict scientific sense. Even in a weak sense, we expect experiments to involve some form of evaluation and it is not clear from your policy or your reply what kind of evaluation will take place with the new OA author fund with the exception of hoping for a “successful dialog among our authors.”

If budget lines for OA journals are to be given the same fair treatment as budget lines for traditional resources, one would expect a similar form of fiscal responsibility and accountability.

Comments are closed.