SUMMARY: Free/Open Software (notably the first Free Software for creating OAI-compliant Open Access Institutional Repositories, EPrints, created in 2000, distributed under the GNU license, and now used worldwide) has been central to the growth of the Open Access Movement. However, there are also crucial distinctions that need to be made and understood, among the movements for (1) Free/Open source software, (2) Open Access (to peer-reviewed research), (3) Open Data, (4) Creative Commons licensing, and (5) Wikipedia-style collective writing. Open Access (OA) is focussed primarily on refereed research articles. The crucial distinctions revolve mostly around (a) the fundamental difference between author giveaway vs. non-giveaway work and (b) the functional differences between the re-use needs for peer-reviewed research article texts on the one hand, and data, software and other kinds of digital content on the other. (PPT)
Richard Stallman (at the Free Sofware Free Society Conference today in Kerala, India) seems to have come to the mistaken conclusion (from my own talk, presumably) that I am somehow against Free Software!
“Open Access, Free/Open Software, Open Data, Creative Commons and Wikipedia: Commonalities and Distinctions“
But Richard (whom I admire very much) doesn’t seem to understand that what I am actually trying to do, for concrete, pragmatic, strategic reasons, is to very explicitly distinguish the special case of (“gratis” Green) OA from the other 4 “open” cases (free/open software, open data, creative commons licensing, and wikipedia) that resemble OA in some respects, but only partially. The purpose of pointing out this distinction is so that we can at long last reach universal Green (gratis) OA. Universal Green OA will then in turn help strengthen and accelerate reaching the goals of the other 4 “open” movements. Conflating all 5 goals today will not.
The reason Richard does not seem to grasp or accept this is also related to the reason he is so effective where he is indeed effective: He is on an ethical crusade (an ethical crusade in which he is just as right as if he were crusading for providing free health care for all, the curing of all diseases, the remedying of all injustices). But ethical rectitude in principle is alas insufficient to elicit ethical practice, or at least not on a scale that is anywhere near universal: To achieve that, you sometimes have to appeal to self-interest too, at least initially.
For OA, it is simply hopeless to try to get all or even most creators of digital content to provide OA to their creations today if they do not even want to make it freely available. They all ought to want to, perhaps; but telling them they ought to want to (or that it is more ethical to want to) is not enough.
That is why it is essential to have a practical strategy that is aimed explicitly at the subset of creators that, without exception, already want to give away their creations (because it already happens to be in their own interest to do so). Those are the authors of the 2.5 million annual articles that are the target of OA and of Green OA mandates: OA’s target content. They publish their research only for the sake of uptake, usage, application and impact, not for revenues or fees. That simply cannot be said of the authors of most software (or other kinds of digital content) today.
What I was trying to explain at FSFS was just that this special case (of calling on authors to provide OA to their refereed research articles) has to be distinguished from the general case of calling on all authors of all kinds of digital content (whether it be books, data, software, music, movies, or “knowledge”) to make their content free or open.
And the reason is that OA’s target content all consists of exception-free creator give-aways already: No ethical case for openness or give-away needs to be made in the special case of OA’s target content, because its authors already give it all away. Moreover, although most of them won’t go on to do so of their own accord (because they are too busy and/or worried about copyright), most of those authors, when surveyed, state that they would go on to make their give-away articles OA too, willingly, if their institutions or funders were to mandate it: And the evidence is that, when it is indeed mandated, these authors do indeed comply and do it.
So mandates work for author-give-away content. Authors say they will make it OA, willingly, and the actual mandate adoptions confirm that authors do as they said they would do.
There is no reason, however, to expect mandates to work for non-give-away content, today. Authors certainly have not said they would willingly make their non-give-away products (books, software, music, video, data) OA if it were so mandated; nor are there any mandates that test whether they would comply, willingly or unwillingly.
It is not even thinkable today to try to mandate providing OA to content for which its creators not only don’t provide OA spontaneously of their own accord, but don’t want to provide OA, because they don’t want to give it away in the first place (hoping instead to make money from it).
OA mandates have already been sluggish enough so far in just reaching consensus on adoption for just give-away content. What they need is to provide much better and clearer information for authors, their institutions and their public funders on what OA and OA mandates really entail — and what benefits they bring for authors, their research, their institutions, and the public that funds them — rather than an unrealistic and confusing raising of the existing hurdles to reaching consensus on mandate adoption by conflating giveaway content with content that its creators do not (yet) even wish to give away, and for which a credible case based on self-interest cannot yet be made.
Having said that, I of course agree completely with Richard Stallman that if software authors are publicly funded for developing their code, the funder can and should mandate that it be made FS/OS! That is a special case in which OA and FS/OS have far more in common than they do in general. But relative to all the software being written today, the portion that is being developed with public funding is, I suspect, quite small (which is not to say that it should not be mandated to make that portion FS/OS!)