The first caveat is that it may be a rather small experiment; [...]]]> The announcement from JSTOR of their new “Register & Read” program, reported here in Inside Higher Ed and here in The Chronicle of Higher Education, seems like a promising experiment.  It deserves both praise and a couple of caveats, I think.

The first caveat is that it may be a rather small experiment; only 5 of the 50 University of Chicago Press journals, for example, will be available through Register & Read.  The success of the experiment in teaching JSTOR about its potential readership that does not have subscription access may be limited if the size of the sample available is too small, or if the disciplinary focus of what can be read is not broad.

The second important caveat is that this is not an experiment in open access.  Allowing unsubscribed readers “read only” access to three articles from a limited archive every two weeks is nice, and addresses one aspect of the overall access problem.  But it leaves a great deal of that problem untouched.

For example, the Inside Higher Ed piece mentions Aaaron Swartz and his apparent attempt to hack JSTOR and download millions of articles.  Some have claimed that Swartz was attempting to collect these articles for “non-consumptive” text-mining research.  I have no idea if that is true or not, and I disapprove of his methods even if it is.  But the access JSTOR will permit through Register & Read does not begin to address the opportunity for cross-journal research into the characteristics of scholarly output that text-mining offers.  It does not even meet real research needs for academics at under-resourced institutions, since the scope and number of articles is so limited.  At best, and it is a step in the right direction, this program will support one-off research needs and “curiosity” reading.

Nevertheless, there are several things to like about this program.  Most important from my point of view is the recognition it implies of the unexpected reader.  Too many publishers and database vendors have denied for years that there is an access problem at all; that there are readers out there who would like to read articles in JSTOR and elsewhere but cannot do so because they are not affiliated with a subscribing institution.  One can just read the AAP’s recent statement of support for the Research Works Act to find such a fantastical denial.  JSTOR, however, is implicitly acknowledging those potential readers with this new program.  In fact, the Inside High Ed article actually makes reference to the “droves of unlikely visitors” directed to databases like JSTOR by Web search engines.  They are certainly out there, their presence supports most of the strongest arguments for open access, and it is good to see JSTOR acknowledge them.

More than simply acknowledging these unexpected, unsubscribed readers, JSTOR wants to know who they are.  That is the point of registering before one can read under this new program.  JSTOR will collect data to see if individual subscriptions or, one hopes, reasonably priced purchases of individual articles, are possible.  It seems to me that a key element for the future survival of the traditional publishing industry is finding realistic price points for these services; the $39.95 charge that we often encounter today is simply wishful thinking, perhaps intended by publishers to convince themselves that individual sales are impractical more than to actually open that market.  If JSTOR can gather data that will undercut this willful blindness and really begin a discussion of practical access for individuals or non-subscribing institutions, that would be welcome.

I have to note that a colleague suggested to me last week that registration might be a barrier for many people.  For myself, I hope potential users will not see it that way.  The information JSTOR will apparently seek is relatively modest I think, at least compared to what Apple and Amazon and Netflix already know about me!  And JSTOR has been a rather trustworthy partner over time for academic libraries.  What’s more, helping them learn exactly who the folks are that they are currently not serving can have very beneficial consequences.  So even though this is a very modest experiment in access for the under-resourced, I hope individuals will use it and help JSTOR better understand their future business models.

COPE, of course, is the Compact for Open Access Publishing Equity, a plan by which academic libraries create funds to help faculty authors pay the article processing fees (APCs) that some [...]]]> Recently I have had the opportunity to review the first 14 months of Duke’s COPE fund, and it has been an interesting exercise.

COPE, of course, is the Compact for Open Access Publishing Equity, a plan by which academic libraries create funds to help faculty authors pay the article processing fees (APCs) that some open access journals charge as a way to replace subscription fees. Duke established a modest COPE fund in October 2010, and it was re-funded for fiscal year 2011-12.

Now, at the end of November 2011, I find that in 14 months we have reimbursed open access publishing fees for 20 articles written by 18 faculty members. We have six more COPE applications waiting for completion. Assuming that all of them are completed and reimbursed, we will have exhausted the available funding for COPE for the first time.

The authors we have reimbursed are predominately, but not exclusively, scholars in the biomedical sciences. That group is evenly divided between medical school and University-side scientists, but faculty in diverse fields like environmental studies and evolutionary anthropology have also benefited from the fund. The largest group of articles we helped fund were published with journals from the Public Library of Science, followed by Hindawi, Frontiers in Research, and BioMed Central. All of them are, of course, peer-reviewed publications.

With a relatively new initiative like COPE, it is not entirely clear how success should be measured. Is it more successful to spend the entire COPE fund on a campus, or to have it go unused? The answer, I suppose, would depend on the reasons behind the use or non-use.

At Duke we have been very clear that the purpose of COPE is to provide incentives for new models of scholarly publishing and to support open access. As the fund administrator, I am convinced that there is a good deal more open access publishing amongst our faculty than I previously expected. These 25+ articles are very much only the tip of the iceberg. And I cannot say for sure that the COPE fund caused an increase is that type of publishing; I am inclined to think faculty are turning to open access because of its numerous benefits, and that COPE funding is not the strongest factor for most of them. I have been told by authors, however, that COPE is an important example of the University “putting its money where its mouth is,” and I am pleased by that perception.

As open access publishing evolves, different funding models are being tried. It is important to recognize that author-side fees may not be the “winner” over time; many OA journals even now do not rely on such fees, although the best-known ones do. Nevertheless, we are clearly seeing an important transitional movement in scholarly publishing, and a COPE fund is one way for an institution to both encourage that transition and to tangibly affirm a commitment to open access for scholarship.

I expect that at Duke we will reexamine the policies and procedures we have in place for COPE, as we consider how it is to be re-funded, to see if they actually serve the incentive purposes behind the fund. In that task we will gain significant guidance from this recent blog boost by Stuart Shieber of Harvard University’s Scholarly Communications Office, about how funding agencies should assume the task of paying open access article fees. Stuart’s point about funders is important, and I hope that the policies he recommends are widely adopted. But his post is also a cogent and compelling re-assertion of those incentive purposes that COPE is intend to serve and how different policy decisions relate to the overall goals. As such, it provides a helpful guide for anyone considering a COPE fund or considering how to make such a fund more effective.

In its opening paragraph, this I have just returned from the Berlin 9 Conference on Open Access, which was held in Washington, D.C. at the lovely conference center facilities of the Howard Hughes Medical Institute.   It was a fascinating meeting, and quite different in tone from the one I attended last year in Beijing.

In its opening paragraph, this Chronicle of Higher Education report on the conference captures the fundamental difference.  This year the conference was much more clearly focused on the impact of open access on research; rather than talking about how open access will be accomplished, the discussion assumed that open access is inevitable and instead emphasized the differences that the evolution to open will make.

For the sciences especially, it was clear that openness is rapidly becoming the default, because awareness of its benefits is spreading so widely.  This year the partners in the discussion included many working scientists and, significantly, many academic administrators and research funders, who are well-placed, and, increasingly, motivated, to make the transition to open access.  The recent decision announce by the National Autonomous University of Mexico to make a decisive transition to open access is testimony to the impact a commitment by administrators can have.

Some of the most compelling discussion in Washington about the impact of openness centered on the idea of unexpected readers.  For years researchers have assumed that, especially for highly technical work,  all of the people who needed access to their work and could profit from it had access through the subscription databases.  This assumption has probably always been incorrect, but now the promise of open online access has really blown it up completely.  The possibility of unexpected readers, including computers that can make connections and uncover patterns in large collections of works, is now one of the great advantages of OA and one of the primary sources of the expectation for greater innovation.

One very touching story is worth retelling here to make this point.  Philip Bourne, a professor at UC San Diego and Editor in Chief of the journal PLoS Computational Biology, told of a rather remarkable manuscript that was sent directly to him in his editorial role.  He thought it was quite a special work of scholarship, on computer modelling of pandemics, and asked some of his colleagues with expertise in that field for their opinions.  Uniformly it was felt that the article was ground-breaking.  Finally, Bourne met directly with the author and, unusually, urged her to submit it to the journal Science.   You see, the author was a fifteen-year old high school student who had done her research as a visitor in university libraries and, for a while, using a “test” login obtained directly from a vendor.

The point here is not the obstacles to access that this young author encountered and overcame.  The point is that she was not at all the person the authors of previous articles on the topic thought they were writing for.  Yet she made a remarkable advance in the field because she was able to read those works in spite of conventional expectations.

By the way, Science selected her article for in-depth review, which is itself a big accomplishment for even experienced researchers, but ultimately decided not to publish her paper, which will now likely appear in PLoS Computational Biology, as she originally hoped.

In his presentation to the Berlin Conference, law professor Michael Carroll listed five types of readers who should have access to research output, and who do have access when open access becomes the default. On his list of such “unanticipated readers” were serendipitous readers, who find an article that is important to them without knowing they were looking for it, under-resourced readers  (like the high-school author described above), interdisciplinary readers, international readers and machine readers (computers that can derive information from a large corpus of research works).  By the way, the category of serendipitous readers includes all those who might find an article using a Google search and read that work if it is openly available but will encounter a pay-wall if it is not.

Open access serves all of these unexpected readers of scholarly works.  As Carroll summed up his point,  every time we create an open environment, we get unexpected developments and innovations.  We have come far enough down this road now that the burden of proof is no longer on open access advocates, it is on those who would claim that the traditional models of publishing and distribution are still workable.

Earlier this week, only days after it filed its ill-advised lawsuit against the HathiTrust and five of Hathi’s partner universities, the Authors Guild gleefully announced that they had been able to find, with relative ease, the author of one of the books on Hathi’s initial list of orphan works.  You, of course, [...]]]> Dear Mr. Salamanca,

Earlier this week, only days after it filed its ill-advised lawsuit against the HathiTrust and five of Hathi’s partner universities, the Authors Guild gleefully announced that they had been able to find, with relative ease, the author of one of the books on Hathi’s initial list of orphan works.  You, of course, were that author, and the work in question was your 1958 novel The Lost Country.

It is not a comfortable position to be a pawn in a game of “gotcha,” especially when it involves litigation.  What I want to say to you is the same thing I say to faculty authors at the institution where I work: “Consider carefully where your own best interests lie, and manage your copyright to serve those interests.”

In one sense, your situation is quite unusual.  Apparently you still hold the rights in The Lost Country, perhaps because you recovered them from your publisher based on a contractual arrangement.  This was unusual in the 1950′s, when federal copyright did not attach to a work until it had been published, and it is, unfortunately, still not the case for many authors, particularly those who write academic books.  For many of them, rights must be surrendered in order to have a work published in the first place.  So you are ahead of the game in that sense; you have a chance to really manage your copyright for your own benefit.  Congratulations.

It seems clear that your book was included on the list of potential orphans in error.  Of course, inclusion on that list was precisely intended to catch such situations, so the system worked as it should.  Your book has not been included in any distribution of orphan works.  Now you have a chance to decide, however, if you would like to allow a more open distribution.

I am sure I do not have to tell you that libraries, including those that intend to participate in the Hathi Orphan Works project, are not your enemies.  We are in the business of helping authors find readers, which hardly seems like it should be an objectionable activity.  So let’s think for a minute about The Lost Country and what might be best for it and for you.

The sad fact is that The Lost Country has become a pretty obscure work.  Amazon.com shows only two used copies available for sale.  In the Duke Libraries, the last transaction record we have for your novel is in 2004, when our copy was sent to high-density storage.  It has not left the facility once since then, and our system shows no circulations in the prior decade, either.   One of the famous “laws” of librarianship is that every book should have its readers, and the current system, I am afraid, is failing to connect your book to new readers.

It has to be said that the Authors Guild is not going to help you in this regard.  They are not going to publish a new edition of The Lost Country for you, nor will they pay you any royalties on the out-of-print edition.  The Authors Guild simply does not have the ability to create a new market for your book.  Even if they were to succeed in a grand strategy to impose a licensing scheme for orphan works in general, there is no reason to believe that you would profit from it. With such an obscure work, potential users who had to pay a fee would probably just skip the planned use.

Where you can find help for this problem is with the HathiTrust.  Their goal, and the goal of the libraries that plan to participate in the orphan works project, is to make it easier for readers to find works like your novel, which might otherwise languish on shelves or in large warehouses of books.  Digital access to low-use titles through our catalogs will encourage users to discover resources, for study and for entertainment, that they might not have bothered with before.

In your own case, let’s suppose a Duke student has recently seen the Elvis Presley movie made from The Lost Country.  Intrigued, she “Googles” the book and finds that there is a copy held by our library.  But to get it she has to send a request, wait 24 hours or so, then pick it up at one of the library service desks.  Years of experience with college students suggests to me that most just won’t bother; they will move on to something newer and easier to access.  On the other hand, if  that same record that she found with her Google search also contained a link to the book through Hathi, she might read a chapter or two.  She might get hooked.  You will have found a reader.

This is what libraries do; such serendipitous discovery is what we hope for everyday, and it is why we signed up with the HathiTrust.  What Hathi offers to you is the opportunity to continue to find readers for the book on which you worked so hard.

Your “case,” if I can call it that, illustrates two things.  First, that the process of identifying orphan works in the Hathi corpus needs to be tested and refined, which Hathi is committed to doing.  Second, in the rare instance like yours where the process actually turns up an author who does still own copyright, the rational course for that author is to embrace the mission of Hathi and of libraries everywhere of connecting books with readers, and to exercise their right to make their book(s) fully viewable.  Please believe me, that is a much better option than having a book live out its term of copyright on hard-to-access shelves in high-density storage.

Swartz has been indicted on several [...]]]> One of Will Rogers’ best known aphorism is “I only know what I read in the papers.”  In line with Rogers’ irony, if all one knows about the Aaron Swartz case is what one reads in the blogosphere, one knows very little indeed, and much of it wrong.

Swartz has been indicted on several federal charges after allegedly physically and technologically gaining unauthorized access to the MIT network and downloading a huge number of files from JSTOR.  On that everyone agrees.  After that the claims about and arguments based on this event diverge dramatically.

Predictably, many bloggers (an example is this one from the Copyright Alliance) call these actions by Swartz “theft” or “stealing.”  As always when talking about intellectual property, these words are misapplied.  The formal definition of theft from Black’s Law Dictionary is “the felonious taking and removing of another’s personal property with the intent of depriving the true owner of it.”  It should be clear from this definition why we call authorized use of intellectual property “infringement” rather than theft.  What Swartz is alleged to have done did not remove the intellectual property and showed no intent to deprive the original owner of it; he merely made, allegedly, unauthorized copies, which does not have the effect of depriving anyone else of intangible property. JSTOR was never without these files and they have, in fact, recovered the unauthorized copies.

Whenever someone uses the language of theft in reference to intellectual property, they are trying to cover the weakness of their argument, in my opinion.  Let’s just say infringement and talk about both the legitimate reasons to protect IP and the public policy that permits some unauthorized copying.

By the way, Swartz has not been charged with copyright infringement either.  The charges of wire fraud, computer fraud and illegally obtaining information from a protected computer all relate to the hacking itself, not to the downloads.

Another place where serious misrepresentations abound is when we are told (as in this post on the Scholarly Kitchen) that Swartz has “done this before” because of a previous incident where he download large numbers of documents from PACER, a database used by the federal courts.  That incident, however, involved neither illegal access nor copyright infringement.  Although PACER usually charges a fee, Swartz used a computer at a university on which access was being provided for free as an experiment.  And the materials he downloaded – documents from the federal courts – are not protected by any copyright due to section 105 of the US copyright law.  To be sure, Swartz was protesting the fees charged for access to works created at taxpayer expense for the public good, but his actions in that case have no analogy to the behavior charge in this indictment.

One place where there is significant disagreement is about Swartz’s intentions.  Many bloggers simply assume that he intended to release all of the downloaded files to the public, although Swartz claims he intended to do text-mining research with the articles.  He has done such work before, so there is some plausibility to his claim, which may explain when infringement charges have not been brought.  So turning this into a debate about the open access movement is wholly inappropriate.  It is important to recognize that the victim of these alleged crimes was not JSTOR or any of the journals it aggregates.  The victim was MIT.

However fervently one shares Swartz’s goals for greater access to legal and scholarly information and publications, the actions for which he has been charged do not serve those goals.  Quite frankly, Swartz’s actions were not radical enough, in the sense that they did not get to the root of the problem. It is clear that the system of scholarly dissemination is badly broken, and simply hacking it does not change that fact.  The real change, the real solution Swartz (apparently) seeks, will be found only when the academic authors, the original holders of copyright, stop transferring those copyrights to publishers without careful reflection and safeguards on their right to disseminate their own work widely.