Open Access Archivangelism

SUMMARY: Free online access to refereed research ("Gratis OA") is already within the global research community's immediate reach, because it can be mandated by research institutions and funders. "Libre OA" (Gratis OA plus various further re-use rights) is not within reach because not all researchers want to allow it, hence it cannot be mandated without allowing author opt-out. Yet global Gratis OA is very likely to encourage authors to go on to provide Libre OA too. The worldwide research community should accordingly grasp what is already within its reach, by mandating Gratis OA, rather than over-reaching for what is not yet within its grasp, delaying the immediate benefits of Gratis OA for research uptake, impact and progress.

---

Peter Murray-Rust and I are — and have always been — on the same team. Our disagreements have not been about the ultimate goals, but about the immediate means of reaching those goals. "Gratis OA" is free online access; "Libre OA" is free online access plus various re-use rights. So Gratis OA is a necessary condition for Libre OA: Libre OA is more than Gratis OA, but you cannot have Libre OA without Gratis OA.

But we do not yet have Gratis OA! Less than 20% of yearly refereed research output is OA at all. So the strategic difference between Peter and me is very easy to state and to understand: It is easier to ask researchers, institutions and funders for less than it is to ask them for more, especially when most are not yet providing the less, let alone the more.

How can researchers be induced to provide at least Gratis OA? Their institutions and funders can mandate that they self-archive their refereed final drafts in their institutional repositories immediately upon acceptance for publication. That is Green, Gratis OA. Making journals OA (Gold OA) is in the hands of the publishing community, not the researcher community, hence Gold OA — whether Gratis or Libre — cannot be mandated; only Green OA can be mandated. Moreover, Green, Gratis OA mandates are in far less conflict with either the policies of most publishers or the desires of most authors.

Hence (by my lights) the overwhelming priority today for those who seek OA worldwide should be to see to it that Green, Gratis OA mandates are adopted by institutions and funders worldwide. The rest — Libre OA and Gold OA — will eventually come, once we have mandated universal Green, Gratis OA. But not even Green, Gratis OA will come if we needlessly over-reach now, and insist on more, when we do not even have less.

As we approach universal Green, Gratis OA mandates worldwide, search and harvesting will become incomparably more powerful than it is now. (It is already very powerful now, with Google Scholar, Citeseerx and other new search engines, despite the sparseness of the OA content base (<20%). As OA content becomes less sparse, harvesting and search will become all the more sophisticated and powerful, and the "global bibliography" Peter recommends will assemble across the global network of distributed in institutional OA repositories of its own accord, part of the repository deposit procedure and tagging. The problem today is content, not search.)

As to Peter's concern about the right to re-use figures in Gratis OA articles: It is already possible (and easy) to write a java or perl script today that will call up a figure embedded in a Gratis OA document. Instead of literally reproducing the figure in another work, as in the Gutenberg era (which required permission), the online era allows us to embed a pointer URL that has virtually the same effect -- mediated by one click by the user, to get to the locus of the figure in the Gratis OA article.

As to other Libre OA uses (e.g., data-mining): Those researchers today who (for some reason I find rather difficult to fathom) feel they need advance statements, formal and explicit, of their "harvesting rights" (when everyone else today is happily crawling and harvesting the entirety of web gratis content with impunity) will have to wait patiently until we have Gratis OA; once we have mandated it, Gratis OA's own benefits and potential will induce more and more researchers to seek and provide Libre OA, formally. Over-reaching by asking for more today, when most researchers are not yet even providing Gratis OA, nor being mandated to provide Gratis OA, will not motivate them to provide Libre OA.

In closing, I would like to remind everyone that we are just beginning to think of freeing research from the constraints of the Gutenberg era of Closed Access; throughout the Gutenberg era 100% of research was (and 80% of it still is) neither Gratis nor Libre. In print days, you could not even access a paper if your institution did not have a subscription to the journal in which it was published (and if you did access it, all you could do was read it, and use the information -- not re-use, re-mix, or re-publish the text or figures). The online era made it possible for researchers to make their papers accessible to all potential users, not just those whose institutions subscribed to the journal in which it was published. The further idea of various re-use rights — (and note that there are a number of different levels or degrees of potential re-use rights, all the way to making the document public-domain) — was not even thinkable prior to the online era, when we did not even have Gratis OA — because of the inescapable economic constraints of the Gutenberg medium.

So if Libre OA feels urgent now, it is only because the online era has made Gratis OA possible. But before we try to reach the farther possibility, surely we should first seize the benefits of the nearer possibility that the online era has already opened up for us (free online access), for that is a proximal goal that we already have within our grasp the tried, tested and effective practical means of reaching (Green Gratis OA mandates by research institutions and funders), rather than continuing to ask for more — without any tried, tested and effective means of getting it.

Peter could perhaps cite the possibility of adopting stronger Green OA mandates — copyright reservation mandates like Harvard's (about which I am sceptical, because of their opt-out clauses) — but Peter is sceptical about mandates in general (whereas I am only sceptical about mandates that needlessly raise the goal-posts while mandates themselves are still sparse worldwide and successful consensus on adopting them is still slow to reach).

The practical question to be asked of anyone who is desirous of immediate Libre OA rather than Gratis OA is hence this: How do you propose to persuade researchers to provide it?

Stevan Harnad

EnablingOpenScholarship