Open Access Archivangelism

KG: "I do not think that using the request button is a valid OA strategy. My own experience was that I received few response when requesting an article. The St. Gallen IR manager said that requesters can obtain much more positive results when mailing to the scholar directly."

(1) Michael White reported that the response rates for the email eprint request button at U. Stirling are about 50% fulfillment, 5% refusal and 45% no response.

(2) He also said that some of the no-responses may have been (2a) elapsed email addresses, (2b) temporary absence, (2c) embargoed theses, and (2d) author unfamiliarity with the purpose and use of the email eprint request Button.

(3) He also noted that the response rates may well improve with time. (I would add that that's virtually certain: It is still exceedingly early days for the Button. And time -- as well as the growing clamor for access [and impact] -- is on the Button's side.)

(4) It is harder to imagine why and how the long and complicated (and obsolescent) alternative procedure -- of a user discovering an article that has not been deposited by the author, finding the author's email address, and sending him an email eprint request, to which the author must respond by sending an email and attaching the eprint -- would "obtain much more positive results" than the author depositing the article in his IR, once, and letting the IR's Button send the email requests for the requesters to the author with no need for look-up, and only one click needed from the author to fulfill the request.

(5) The email eprint request Button does not provide OA; it only provides "Almost OA." But that's infinitely better than no OA. And the Button (and the Immediate-Deposit/Optional-Access -- ID/OA -- Mandate, for which the Button was designed) make it possible for institutions and funders to adopt Green OA mandates that neither need to allow exemptions from immediate deposit nor do they need to allow publishers to dictate whether or when the deposit is made.

(If publishers have a say, it is only about whether and when the deposit is made OA, not about whether or when the deposit is made at all. Since 63% of journals are already Green on immediate OA, the ID/OA Button means that an institution or funder can reach uncontroversial consensus on requiring 100% deposit, which then yields at least 63% immediate OA and 37% Almost-OA, whereas the alternative is not arriving at a consensus on mandating OA at all, or adopting a weaker mandate that only provides OA after an embargo period, or only at the publisher's behest, or allows author opt-out. And the most important thing is not only that the ID/OA provides more access and is easier to reach agreement on adopting, but it will also quite naturally drive embargoes into their well-deserved graves, as the mandates and their resulting OA -- and the demand for it -- grow and grow.)

KG: "The Oppenheim/Harnad "preprint & corrigenda" strategy "of tiding over a publisher's OA embargo: Make the unrefereed preprint OA before submitting to the journal, and if upon acceptance the journal seeks to embargo OA to the refereed postprint, instead update the OA preprint with a corrigenda file" is a valid OA strategy because the eprint is PUBLIC."

What makes a strategy "valid" is that it works: increases access, Open Access, and Open Access mandates.

Both the "preprint&corrigenda" strategy and the "ID/OA-mandate&Button" strategy can increase access, OA, and OA mandates, but the ID/OA-mandate&Button strategy is universal: it scales up to cover all of OA's target content, whereas the preprint&corrigenda strategy is not universal, for it does not and cannot cover those disciplines (and individual authors) that have good (and bad) reasons not to want to make their unrefereed preprints public.

KG: "If an article is published then the author hasn't any right under OA aspects to choose which requester has enough "dignity" to receive an eprint. I cannot accept the arbitrariness of such a decision under OA circumstances."

Relax. The reason neophyte self-archiving authors are not fulfilling Button requests is because they are either not receiving them or don't yet understand what to do with them, not because they are making value judgments about who does and does not merit the privilege of accessing their work!

They'll learn: If necessary, they'll learn under the pressure of the impact-weighting of publications in performance evaluation. But my hunch is that they already know they want the user-access and user-impact (from the eager way they do vanity-searches in the biobliography of every work they pick up in their research field, to check whether their own work has been cited). So all they really need to learn now is how the Button works, and why.

Stevan Harnad
American Scientist Open Access Forum

SH: "[T]here is nothing either defensible or enforceable that a publisher can do or say to prevent a researcher from personally distributing individual copies of his own (published) research findings to individual researchers, for research purposes, in any form he wishes, analog or digital, at any time. That is what researchers have been doing for many decades, whether or not their right to do so was formally enshrined in a publisher's 'author-re-use' document."

RQJ: "This discussion strikes at the heart of green OA implementation. Among other things, it's why we have mandates."

Actually that's not correct. What I was referring to above -- authors mailing an individual analog reprint or emailing an individual digital eprint to an individual requester for research purposes -- predates both OA (Green and Gold) and (Green) OA mandates.

The only connection with Green OA mandates is that email eprint requests for Closed Access deposits whose metadata are openly accessible allow users to request -- and authors to provide -- individual one-on-one "Almost OA" during any OA embargo period: That way Green OA mandates can require deposit of the final refereed draft immediately upon acceptance, with no exceptions or opt-outs, no matter how foolish a copyright transfer agreement the author may have signed.

If a Green OA mandate does not require immediate deposit, then it is completely at the mercy of publisher OA embargoes: The author deposits only if and when the publisher stipulates that he may deposit, because all deposits are OA. If, instead, immediate deposits are required in every case, without exception, but where OA is publisher-embargoed the deposit may instead be made Closed Access during the embargo, rather than OA, then the email eprint request button allows the author to provide "Almost OA" on an individual case by case basis for the Closed Access articles during the embargo.

But if the mandate instead requires deposit only after the publisher embargo has elapsed, that means the only access during the embargo period is subscriber-access. That means a great loss of potential research usage and impact.

RQJ: "I believe Harnad is likely incorrect as a matter of law (at least in the US), but ultimately this may end up as a court case that gives us more explicit guidance.

If researchers sending individual reprints and eprints to individual requesters for research purposes has not gone to court for over a half century, it is difficult to imagine why someone would think it will go to court now: Publishers suddenly begin suing their authors for fulfilling reprint requests?

RQJ: "Note that "research findings" (which are the stuff of patent or academic integrity if protected at all) are very different from their expression in text, which is what is transferred through the copyright agreement."

We are not talking about research findings, we are talking about copies of verbatim (published) reports of research findings: sending them to individual requesters, as scholars and scientists have been doing for over half a century (since at least the launch of Eugene Garfield's "Current Contents" and "Request-a-print" cards):

Swales, J. (1988), Language and scientific communication. The case of the reprint request. Scientometrics 13: 93–101. "This paper reports on a study of Reprint Requests (RRs). It is estimated that tens of millions of RRs are mailed each year, most being triggered by Current Contents..."

RQJ: "Note also that "what researchers have been doing for many decades" is disputable -- arguably what researchers did anteXerox was distribute the 100 or so offprints of their article that they got as part of their Faustian bargains."

They could also mail out copies of their revised, accepted final drafts.

And whether or not any of that was "disputable" before xerox, it certainly wasn't ever contested -- neither with the onset of the xerox era, nor with the onset of the email era.

RQJ: "Note also that courts would be under strong conflicting pressures if a case like this ever actually got heard. On the one hand, Harnad's point is good that courts would want to identify ways to find for those sympathetic scholarly authors. On another, anyone who has been following the RIAA (or remembers Eldred) knows that some of the courts also have tried to find in favor of the owners of the copyrighted works and in favor of sanctity of contract."

Notice that in all other cases but this very special one (refereed research journal articles) both author and publisher were allied on the same side of the copyright/access divide: both wanted to protect access to their (joint) product (and revenues) from piracy by third parties.

In stark contrast, in this one anomalous case -- author give-away research, written purely for maximal uptake, usage and impact, not at all for royalty revenue -- the publisher and the author are on opposite sides of the copyright/access divide, and publishers would not be suing pirates, but the authors of their own works (and not "works for hire!").

I would say that the differences from all prior cases are radical enough here to safely conclude that all prior bets are off, insofar as citing precedents and analogies are concerned.

And I would say that the de facto uncontested practices of millions of scholars and scientists annually for decades since well into both the photocopy and the email eras bear this out.

And although individual reprint/eprint request-fulfillment by authors is definitely not OA (though it is a harbinger of it), the growing clamor for OA today is surely making it all the harder for publishers now suddenly to do an abrupt about-face, endeavoring to contest individual reprint/eprint request-fulfillment by authors after all this time -- and now, of all times!

RQJ: "On a third hand, the institutional employers of the researchers might well try to assert WmfH or other compulsory license theories that trumped the publisher's copyright."

You are thinking here about what institutions (and funders) could do to force the issue insofar as OA is concerned (and I agree, they do have an exceedingly strong hand, and could and should use it if it proves necessary).

But that is not even what we are talking about here: We are just talking about the longstanding pre-OA practice of individual reprint/eprint request-fulfillment by authors, for research purposes...

RQJ: "On a fourth, there's the public interest in "the Progress of Science" and a dearth of good empirical data as to which copyright regimes actually do promote that progress."

All worthy and worthwhile, but probably not necessary, as neither individual reprint/eprint request-fulfillment by authors nor Immediate-Deposit/Optional-Access (ID/OA) mandates are copyright matters:

RQJ: "...Will it ever go to court? Maybe not. The publishers might win their particular case but lose the war by triggering a revolution."

What is the "it" that you are wondering about? Over 90% of journals are already Green on immediate, unembargoed OA self-archiving in some form (63% for the refereed postprint, a further 32% for the unrefereed preprint).

So are you wondering whether the non-Green journals will try to sue their authors? No, they won't. At most, some may try to send them take-down notices, which their authors will either choose to honor or ignore.

But that isn't even what we are talking about here: We are talking about individual reprint/eprint request-fulfillment by authors, for research purposes: Wouldn't the time for authors to worry about that have been 50 years ago, before they began doing it, rather than now, when they and their children and grand-children have already been doing it with impunity for generations?

Stevan Harnad
American Scientist Open Access Forum