The initial idea was brilliant in its simplicity: get bloggers around the world to write about a female role model in technology. The pledge attracted 1,700 signatures, and more than 500 published posts have been recorded so far.

I’ve had the pleasure to meet and work with a number of outstanding women in technology and information policy, including the young women of Students for Free Culture (such as Elizabeth Stark and Karen Rustad); Laurie Taylor, now acting director of the University of Florida’s Digital Library Center; the remarkable women of SPARC, Heather Joseph and Jennifer McLennan; many others throughout the world of libraries and open access, including Prue Adler, Julia Blixrud, and Karla Hahn of ARL, Donna Okubo of PLoS, Heather Morrison, and others; Heather Ford, formerly of iCommons; and many others. Wendy Seltzer‘s Chilling Effects Clearinghouse inspired me as a tool against Internet censorship and repression. Jessica Litman’s Digital Copyright was a constant companion when I studied the DMCA as an undergrad. I even read danah boyd‘s and Dorothea Salo‘s blogs.

But when I thought of someone to profile, one women stood out in my mind: Gigi Sohn. I think it’d be realistic to call Gigi one of the most important women in American tech policy. Gigi and Public Knowledge have been instrumental in many of the most important legal and political battles of the era, from the successful suit to overturn the broadcast flag to orphan works, Net neutrality, and more. Her hard-nosed approach gets results even as it wins admirers.

So, cheers to Gigi, and to the past, present, and future women of tech. Happy Ada Lovelace Day!

Let me make a few statements, just to get them out of the way:

• Every research funder should mandate OA to the research it funds.
• Every institution that produces research (college, universities, research institutions) should mandate OA to the research it produces.
• Every researcher should ensure her research is OA.

These obligations are independent and co-equal. This particular discussion is at the level of policy — the first and second points — since practice has shown that policy is significantly more effective than unenforceable exhortations to authors in general. So our question is not whether to mandate OA, but how.

Where he’s wrong: Dr. Harnad repeats a claim that I dismiss verbatim in my original post: “Not All Research Is Funded, But All Research Is Institutional.” In the body of his post, he weakens this statement to “virtually all”. But my point — that not all academic journal literature is written by authors at colleges, universities, or research institutions — remains. I don’t know what percentage of authors are independent scholars or employees of non-profits, government, or corporations, but I’d hazard a guess that it’s at least 5% across all fields, and probably as high as 10% or 15% in some fields. (Harnad doesn’t offer a number either, but claims it’s “such a minute fraction” it’s not worth worrying about.) Institutional mandates won’t touch these authors (by definition). Funder mandates may touch some, though I’d suspect that most such research is unfunded. Regardless of where the author works, we need OA to academic journal literature. Where non-institutional authors are funded, we need funder mandates to ensure OA to their research. Since these authors don’t have an institution, they don’t have an institutional repository, so they need the flexibility to deposit in a funder or thematic/disciplinary repository instead. This is, to me, the minimum sane policy for funder mandates — not just for non-institutional authors, but for authors whose institution doesn’t offer an IR. (Harnad notes that some funders do offer “back-up” repositories for grantees without IRs, but without endorsing this flexibility.)

Harnad also deftly re-defines the traditional meaning of “institution” in “IR”, to mean “any organization whose employees might ever produce research” rather than the commonly-understood sense of “an institution of higher education or research”, i.e. colleges, universities, and research labs. As a result, he reduces further the number of “unaffiliated” researchers. While some non-HEI/research organizations that produce research may establish repositories, as Médecins Sans Frontières did, many won’t, especially as the research output declines. (If one employee per 1,000 publishes an article per year, does it make sense to host and manage your own IR? Doesn’t it make more sense to tell your employees to deposit in a higher-visibility thematic repository?)

My point about authors with multiple institutional affiliations also goes unheeded. Harnad suggests that this situation “only” requires one deposit, “followed by export to the IRs of the rest of his institutions”. Well, now we’re talking about export, which is exactly what I call for between IRs, thematic, and funder repositories. There’s no difference between exporting from one IR to another and between exporting from a thematic repository to an IR.

Harnad also understates the resource requirements for adequately maintaining an IR. I’ll let Dorothea Salo answer this point. Since I spent much of today waiting for my email server to resume functionality, I know that servers don’t just maintain themselves, even if you didn’t dedicate any resources to customer service or development.

Where I was wrong: Since more research comes out of institutions (as a whole) than is funded (as a whole), institutional mandates offer broader coverage. (But a given funder may sponsor significantly more research than a given institution produces, and there are fewer funders than institutions, so they give more “bang for the buck” as initial targets.) To the extent that funders are ahead of institutions (and they generally are), it is useful to use any leverage that exists to encourage institutions to deploy IRs and mandate deposit in them. Why must institutions deploy IRs? For the same reason that funders must offer their own repositories: you can’t require authors to deposit somewhere that doesn’t exist. Just as a funder shouldn’t mandate deposit without offering a “repository of last resort”, neither should an institution. Having established that both institutions and funders should offer repositories (since they should both require self-archiving, and you can’t require deposit without offering a “repository of last resort”), we return to the earlier question: do funders have any leverage to encourage institutions to deploy IRs and mandate deposit? I am less sanguine about this than Harnad, but flooding IRs with deposits certainly could have the effect of building institutional constituencies for IRs and thus for institutional mandates. (However, this assumes the IRs are well-managed — otherwise, you’re breeding dissatisfaction with them — and while funders can have measure of control over the management of their own repositories, they can’t do so with IRs. You have to hope that institutions will decide the requisite resources to maintenance. You entrust your infrastructure to a third party. Enforcement now includes not just deposit by the author, which is easy to measure, but adequate maintenance by the institution, which is less easy.)

Therefore: Is there some promise in using funder mandates to prod universities toward institutional mandates? Yes, there’s some. Is it a sure thing? No. Do funders still need to provide a “repository of last resort” to deal with exceptions? Yes. Is it still easier for funders to monitor compliance and ensure adequate levels of service (including preservation) by requiring grantees to get their work into the funder’s repository, one way or another? Yes.

So I’m willing to shift a bit from my prior position of the symmetry of funders and institutions vis-à-vis repositories. But I still don’t see the IR warranting the level of primacy that Harnad would assign to it.

As for policy recommendations for funders, then, I’m agnostic between these two options — but admit either is preferable to my earlier suggestion:

• Require grantees to deposit in their IR, if they have one. Require also that grantees ensure their publication ends up in your repository, wherever they initially deposited it. If they are accustomed to depositing in an institutional repository, they’re responsible for working with your repository manager to ensure their work gets harvested. Better still if the repository allows users to set up automatic harvesting themselves, either for a single publication or for all the author’s future publications, if they don’t want to talk to the repository manager.
• Encourage grantees to deposit in their IR, if they have one. Require also that grantees ensure their publication ends up in your repository, wherever they initially deposited it. If they are accustomed to depositing in an institutional repository, they’re responsible for working with your repository manager to ensure their work gets harvested. Better still if the repository allows users to set up automatic harvesting themselves, either for a single publication or for all the author’s future publications, if they don’t want to talk to the repository manager.

There are costs and benefits to either, in my view. Requiring offers the potential benefit of building constituencies for IRs and institutional mandates. But it also comes with the cost of introducing another step to the process, and runs the risk of backlash if IRs are poorly managed, a factor over which the funder can have little control.

(Aside: Funder repositories come with their own cost: the increased potential for government interference in science that comes with greater centralization, since the largest funders are public agencies. As I’ve written before, this is a minor but non-trivial consideration: minor because published research, by definition, is available elsewhere, and because funding decisions are more of an “interference” than tampering with a non-exclusive database of results could be; but non-trivial because dissemination ought not to be restricted or tampered with, particularly as repositories become more important as a dissemination source. Luckily, we can erect firewalls for academic freedom, as I’ve proposed. In addition, we have friends like the Scientific Integrity program at the Union of Concerned Scientists as watchdogs for such policies when they are adopted.)

Update: Things I forgot to say:

• LOCKSS: lots of copies keeps stuff safe. As long as we can have export/cross-deposit down to a few clicks (or dropping an email to the repository manager with a link to the item to harvest), there’s little harm in having multiple identical copies, and some benefit in terms of preservation.
• Heather Morrison’s comments succinctly point to some of the benefits of both thematic/disciplinary and institutional repositories. (Some qualities of each apply to funder repositories.) This is one of the (previously) unstated assumptions in my argument: if authors are depositing somewhere, it’s because they see a benefit to doing so, so let them keep on doing it, and make it simple for them to ensure it ends up in your repository, too.
• Peter Suber’s comments in his February newsletter (#3 and #13) summarize well the gist of what I concluded in this post and the previous. Not surprisingly, that’s where this whole thread started: Bernard Rentier’s post, to which my previous post was responding, was in turn a response to Suber’s comments. But even if we ended up in the same place, it’s (hopefully) with a more thorough understanding than we started with.

In a sense, this tension between scientific independence and state power (especially the power of the purse) is nothing new, as any schoolchild familiar with the trial of Galileo knows. But the timing of this story — just a few days before the new NIH public access policy takes effect, requiring funded researchers to deposit their manuscripts in a central database — begs comparison.

(First, though, it’s worth nothing the biggest difference: the funder in this case is USAID, which primarily funds development programs, unlike the NIH or similar agencies which primarily fund academic research. The latter, unlike the former, can be expected to have a culture of healthy respect for academic freedom. The former’s policy of censorship has a decades-old history and, while protested by reproductive rights activists, could be expected to draw little attention from the research community. By contrast, a new policy of censorship by an agency like the NIH could be expected to spark outrage by researchers.)

No doubt some skeptics of open access will take the opportunity to again insinuate that OA equals government censorship, as the American Association of Publishers’ consultant so famously suggested. Others, who prefer institutional repositories to those organized by funders or discipline, may add the supposed threat to academic freedom to their litany of arguments in favor of IRs (though others, in conversation, have suggested the same logic applies against IRs, by increasing the influence of the university relative to the faculty member in the dissemination of research).

In reality, funder repositories are no greater interference in the conduct of science than the funding itself (likewise with institutional repositories). Nevertheless, I’ll suggest three strategies to minimize the risk:

1. Recognize the importance of Peter Suber’s second principle for university open access policies. The principle, as developed by Suber in the April 2008 issue of the SPARC Open Access Newsletter, is: Universities should not limit the freedom of faculty to submit their work to the journals of their choice.

   Situations like that of the JHU database call into question Peter’s assertion that “[funders] have an interest in making that research as useful and widely available as possible, and virtually no competing interests.” (See also point 2, below.)

   We must remember that the imperative is OA, not OA through a particular venue. Preserving the author’s right to publish where she wishes, and to self-archive where she wishes (non-exclusive), doesn’t prevent undue interference with academic freedom, but it leaves other options open and mitigates the damage (see also point 3, below).

2. Erect firewalls for academic freedom when institutions and public agencies take new roles in disseminating research. (Really, this applies to any new entrant in the field of disseminating information, but state actors and institutions which employ faculty have the most concentrated capacity for influence.)

   For the avoidance of doubt, concern for academic freedom is not an argument against OA, or an argument for universities or research funders not to act in support of OA. But we should be conscious to erect basic bulwarks against the erosion of that freedom.

   In the case of NIH‘s PubMedCentral, a simple policy to the effect that “All NIH-funded research accepted for publication in a scientific journal shall be included in PubMedCentral without prejudice to the topic or conclusions” would establish a formal rule against situations like the Hopkins database finds itself in.

   Another basic firewall should be expectation that, where information is removed or edited for any reason, that change should be transparent to the reader. (In this case, the Hopkins database contained no notification of the change.) Even in Google in China tells users that their search results are being censored.

3. Use copy-permissive licenses on research publications to ensure maximum dissemination.

   Imagine a similar situation in which a database’s search results are censored. A mirror site could easily copy the database’s content and provide an unfiltered search interface. To be fair, an external search engine like Google could do so, too; but while that might solve this problem, imagine instead that the database is deleting publications altogether, and that (for whatever reason) the database contains the only copy of the article on the Web.

   Giving permission to copy makes censorship less effective: as the saying goes, lots of copies keeps stuff safe.

   • Aside: Some people have little regard for eliminating permission barriers as a goal of OA. They aren’t sure precisely why someone else should be able to copy, translate, format-shift, or mashup their research. But that’s exactly the point of eliminating permission barriers: it enables uses we can’t even imagine (in addition to the ones we can, which are important and valuable in their own right).

     I see it as the corollary to seemingly “harmless failures” in security systems: eliminating permission barriers to research publications may be a seemingly benefit-less success. With apologies to Ed Felten, having identified a benefitless success, there are two ways to proceed. The first way is to think carefully about the possible benefits, and realize that some of it is actually useful. The second way is to think, “This looks like a benefitless success, but we should do it anyway. Maybe good can come of this.” The first way protects you if you’re clever; the second way always protects you.

Update: Hopkins has reversed its decision. The sponsoring school’s dean released a statement, saying in part,

The Johns Hopkins Bloomberg School of Public Health is dedicated to the advancement and dissemination of knowledge and not its restriction.

That is precisely the right response. It’s a dedication I wish academics would live up to more often (see: progress on open access to research, open courseware, adoption of FOSS in higher ed, etc.)

The House recently passed a bill, HR 2102, which would create a reporter’s privilege, and a companion bill in the Senate has been reported out of committee, S. 2035. However, the President has indicated he would veto the bill.

Before the House voted on the bill, Josh Wolf spotted some changes from earlier language. He notes:

When the bill was first introduced in May, it was not without its weaknesses, but it’s [sic] broad definition of who was covered under the bill and its limited exceptions made for a robust bill that I could easily support.

By August the bill had been amended to refine who would and wouldn’t qualify as a journalist. The new language stated that only those who who derive “financial gain or livelihood” from their journalistic activity would be covered. Journalism students would no longer be covered, nor would many bloggers. [...]

Since then, two more amendments have been introduced that will further erode the bill. One introduced by Representative Boucher (VA) and Pence (IN) (sponsors of the original bill) adds that the journalist must depend on his craft for a “substantial portion of the person’s livelihood or for substantial financial gain” [emphasis by Josh].

I was struck by a comment on the article:

I noticed in the Washington Post article that the large newspaper conglomerates all support the bill with the amendment. Mmm?

My inner political scientist awoke. Here was James Q. Wilson’s theory on diffuse vs. concentrated benefits: the relatively few and concentrated commercial journalism operations were effectively defending their interest; the many and diffuse bloggers and students were not. And Mancur Olson’s logic of collective action: the relatively small, homogeneous, tight-knight community of large journalism organizations could effectively mobilize — it was even potentially a privileged group, where it would be rational for one actor to expend resources to pursue the good even if no others followed suit. The larger, more heterogeneous, more disconnected group of amateur journalists could not effectively mobilize.