Self-Archiving Mandates

SUMMARY: NIH's Acting Director, Raynard Kington, writes that "NIH [is] open to closer collaboration with institutional [repositories]... [D]irect feeds from [institutional repositories (IRs) are] worthwhile [but] raise important technical and logistical challenges..."
    All technical and logistical challenges to designating Institutional Repositories (IRs) as NIH's preferred locus of direct deposit (followed by "direct feed" to PubMed Central (PMC)) can be successfully met (most already have been):
(1) The SWORD transfer protocol has already solved the problem of automatically exporting IR deposits to other repositories.
(2) "Author approval": Authors are mandated by NIH to deposit, and NIH specifies the locus of deposit.
(3) "Copyright permissions": If copyright is not an issue with PMC deposit, it is even less of an issue with direct institutional deposit in the fundee's own IR.
(4) "Quality control": The IR deposit can be exported by "direct feed" (via SWORD) to PMC, where exactly the same quality controls can be performed as are now being performed by PMC.
(5) "Formats for electronic transfer": The SWORD protocol does the electronic transfer, and the format for deposit of the author's final, refereed, revised draft is exactly the same.
    The benefits of NIH/institutional collaboration on direct feeds will be enormous, and will far exceed the current reach of the NIH mandate. This should also be cited in the defense of NIH's historically invaluable public access policy against the Conyers Bill's attempt to overturn it.

---

In his "Analysis of Comments and Implementation of the NIH Public Access Policy," Dr. Raynard Kington, Acting Director, National Institutes of Health (NIH), writes that

"direct feeds from [institutional repositories (IRs) are]... worthwhile... but... raise important technical and logistical challenges regarding author approval, copyright permissions, quality control, and formats for electronic transfer. The NIH remains open to closer collaboration with institutional [repositories] and will consider this issue as the Policy matures."

It is virtually certain that all technical and logistical challenges to designating Institutional Repositories (IRs) as NIH's preferred locus of direct deposit (followed by "direct feed" to PubMed Central (PMC)) can be successfully met (and most already have been: see below). The benefits of NIH/institutional collaboration on direct feeds will be enormous, and will far exceed the current reach of the NIH mandate (which is now restricted to the 80,000 articles a year resulting from NIH funding, no more, no less).

The NIH mandate touches the institutions of every one of NIH's fundees. If the NIH mandate preferentially encourages its fundees to deposit their NIH-funded output in their own respective IRs (with direct feed to to PMC therefrom, instead of direct deposit in PMC, as now), it will also motivate their fundees to deposit -- and motivate their fundees' institutions to mandate the deposit of -- the rest of their institutional output in their IR too, not just the NIH-funded fraction of it. Not so if the 80,000 NIH articles must be directly deposited institution-externally (in PMC): That has the exact opposite effect, competing with and complicating, hence demotivating institutional deposits and mandates. (And we must not forget that the institutions are the universal providers of all research output: funded and unfunded, across all disciplines.)

The "technical and logistical challenges" for "direct feeds" from IRs to PMC have already been largely met:

(1) The SWORD transfer protocol has already solved the problem of automatically exporting IR deposits to other respoitories.

(2) "Author approval" is not a problem at all: Authors are mandated by NIH to deposit, and NIH specifies the locus of deposit. Currently that designated locus is PMC. The recommendation here is that the preferred locus of deposit should instead be the author's IR, with the deposit then automatically ported by "direct feed" (via SWORD) to PMC. (Institutional deposit will in fact simplify deposit for authors, increasing their motivation to comply with the NIH mandate.)

(3) "Copyright permissions" are not a problem either: Authors are mandated by NIH to deposit and NIH specifies the locus of deposit. If copyright is not an issue with PMC deposit, it is even less of an issue with direct institutional deposit in the fundee's own IR. (Publisher embargoes can be -- and are -- implemented by IRs just as they are by PMC.)

(4) "Quality control" is not a problem either. Authors are mandated by NIH to deposit their final, refereed, revised draft, and NIH specifies the locus of deposit. The IR deposit can be exported by "direct feed" (via SWORD) to PMC, where exactly the same quality controls can be performed as are now being performed by PMC. (The IR direct-deposit protocol can easily be made to conform to the PMC direct-deposit protocol: they are almost identical already.)

(5) "Formats for electronic transfer" is not a problem either. The SWORD protocol does the electronic transfer, and the format for deposit of the author's final, refereed, revised draft is exactly the same.

It is very welcome and timely news that NIH's Acting Director is "open to closer collaboration with institutional archives." The sooner a collaborative deposit policy, with IR deposit and direct feed to PMC can be adopted and announced, the sooner its potentially enormous knock-on effects will begin to make themselves felt in helping to wake the "slumbering giant" -- the US and global network of universities and research institutes, not only the NIH-funded ones, but all of them: the universal providers of research, worldwide -- to create their own IRs (if they don't have them already) and to mandate the deposit of all of their own research output into them, not just NIH-funded research.

This global enabling effect of the NIH mandate for accelerating and facilitating universal OA should also be cited in the defense of NIH's historically invaluable public access policy against the Conyers Bill's attempt to overturn it.

(And the other research funding councils worldwide, too, should be encouraged to consider the enormous potential OA gains -- at no loss -- from stipulating IR deposit rather than institution-external deposit in their own OA policies as well.)

Stevan Harnad
American Scientist Open Access Forum

Science Foundation Ireland (SFI) (funder-mandate):
http://www.sfi.ie

Repository Portal:
http://www.irel-open.ie/

Self-Archiving Policy:

1. All researchers are required to lodge their publications resulting in whole or in part from SFI-funded research in an open access repository as soon as possible after publication.

2. The publication to be so deposited relates to accepted peer-reviewed publications, conference proceedings and technical reports.

3. Publisher’s copyright, licensing and embargo policies must be respected.

4. The protection of intellectual property, in any form, arising from the publication takes precedence over open access depositing and indeed over any form of publication.

5. The time by which open access depositing is to take place ranges from immediately on publication, if possible, up to six months from the date of publication.

6. Institutional and disciplinary repositories should be used in preference to an author’s own website. In the Life Sciences, PubMed Central and/or a future European PubMed Central should be used. Information on the Irish universities’ institutional repositories is available through IREL-Open

7. The publication to be deposited is to be the publisher’s version, if permitted, or the author’s final version. It is required that there be a link from the deposited version to the publication site, and also that the DOI (Document Object Identifier) be used.

8. On publication, whether in print or online with a DOI, metadata must be deposited in an open access (institutional, disciplinary, other) repository. Metadata comprises at a minimum the full bibliographic data.

9. Suitable open access repositories should make provision for long-term preservation of, and continuing free public access to, their contents.

10. Research monographs are within the scope of peer-reviewed publications. They must be deposited in an open access repository as soon as possible after publication, subject to the copyright which has been assigned to the publisher.

11. Wherever possible supporting research data should be made available in an open access repository at the time of publication of the research publication. Research data includes associated metadata and supporting documentation.

12. In the interest of reproducibility of scientific results, wherever possible, software should be made available in an open access repository at the time of publication of the research publication. Software is understood to include documentation on usage, and on methods and algorithms employed.

[As the Conyers Bill in the US seeks to undo the good done by the NIH Public Access Policy, here is some sunnier news from Spain. It is no exaggeration that open access to health research advances research progress and saves lives, whereas the Conyers Bill seeks to protect publisher access to profits at public expense. Iberia has different priorities. Many thanks to Eloy Rodrigues and Alicia Lopez Medina for posting this information about Spain's Draft Open Access Law on the American Scientist Open Access Forum.]

---

Draft of the National Law of Science in Spain includes Open Access.

A draft of the National Law of Science in Spain has been released and includes a chapter dedicated to Open Access.

Article 33 of the draft states (translated from Spanish):

1. The agents of the Spanish System of Science and Technology will promote the development of institutional or aggregated repositories to allow open access to the publications of their researchers

2. The researchers, whose research had been granted with public funds shall make publicly available a digital version of their own final draft (peer-reviewed and accepted for publication) as soon as possible, but not later than 6 months after the official date of the publication of this final draft.

3. This final version will be made publicly available in open access repositories, disciplinary or institutional.

4. The public digital version deposited in the repositories can be used in the evaluation processes by Public Administrations.

The inclusion of Open Access in this document represents an important and significant commitment in support of Open Access and repositories in Spain.

The draft document is available online from https://lcyt.fecyt.es/

SUMMARY: Imre Simon asks:
Why are Institutional Repositories (IRs) near empty unless mandated, whereas Central Repositories (CRs) like ArXiv and  CiteSeerX appear to be full without a mandate? 
    Here is the answer:
    (1) Authors deposit papers directly in Arxiv, whereas CiteseerX (like Google Scholar) is harvested from authors' websites.
    (2) The crucial factor is central vs. institutional locus-of-deposit. Search is always at the CR level.
    (3) These CRs (Arxiv for physics, Citeseerx for computer science) are fuller than IRs because: (3a) An entire discipline is bigger than an institution. (3b) The global unmandated deposit rate is about 15% of OA's total target: all annual journal articles, across all disciplines and institutions. (3c) But deposit rate is the ratio of deposits to total output, which is much bigger for an entire discipline than a single institution. Physics and Computer Science have been depositing, one centrally, one institutionally, unmandated, for years, but OA's problem is all the disciplines that are not.
    (4) Locus-of-deposit and mandates are closely related issues.
    (5) Deposit mandates can be either funder or institutional mandates.
    (6) Funder mandates only cover funded research, and not all research is funded. 
    (7) But all research output is institutional.
    (8) So if all institutions mandated OA, that would generate universal OA.
    (9) So what is most needed is universal institutional OA mandates.
    (10) Funder mandates would help far more if they could facilitate the deposit not only of the research they fund, but all research.
    (11) To do this, funder mandates need only change one small detail. This would lose none of their funded content, but could help gain the rest of the output of each of its fundees' institutions.
    (12) Funders need to stipulate the fundee's own IR as the preferred locus-of-deposit for complying with the funder's deposit mandate.
    (13) The fundees' deposits can be harvested to CRs from IRs.
    (14) The issue of search and functionality at the harvester level is a red herring. 
    (15) The special features of the few disciplines that began spontaneously self-archiving long ago, unmandated, have nothing to do with the IR vs CR deposit-locus issue; hence unmandated CRs do not offer a viable alternative to universal IR mandates.

---

Imre Simon wrote (in the American Scientist Open Access Forum):

"It is an unquestionable reality that unmandated IR's [Institutional Repositories] remain all but empty.

"[In contrast] ArXiv, CiteSeerX, Repec and SSRN are the four examples of large thematic repositories [Central Repositories, CRs] I know of which are populated without a mandate.

"One wonders why?"

The answer is highly instructive. Let me try to map it out as 15 simple points, one following from the other:

(1) There is a profound difference between (1a) Arxiv (and perhaps also SSRN), which are Central Repositories [CRs] in which authors deposit papers directly, and (1b) CiteseerX (and partly also Repec), which are harvested CRs, their papers and metadata being harvested from local repositories, usually at the author's host institution, where they have been directly deposited. Harvested CRs are like OAIster -- or, for that matter, Google Scholar!

(2) The difference is crucial, because central vs. institutional locus-of-deposit is what is really under discussion here; no one is disputing that navigation and search are done, and should be done, at the central level, irrespective of whether CR deposit is direct or CR contents are harvested.

(3) There are several reasons why these particular CRs (Arxiv, Repec, SSRN, and the biggest of all, Citeseerx) are fuller than IRs:

    (3a) An entire discipline is bigger than a single (multidisciplinary but local) institution

    (3b) These CRs contain only the deposits of those individual authors and disciplines that do deposit spontaneously, unmandated; these amount to about 15% of OA's total target output, and that is well known. The problem is the remaining 85% -- which will be pretty homogeneously represented in each individual multidisciplinary institution's IR (each one 85%-empty if unmandated).

    (3c) But there is a systematic denominator bias here, for the success of an IR in capturing its institutional research output needs to be reckoned as the ratio of its annual deposited papers to the total annual paper output for that institution, whereas for the success of a CR must be reckoned as the ratio of its annual deposited papers to the total annual output for the discipline or disciplines the CR covers (worldwide)! For certain disciplines and subdisciplines -- such as High Energy Physics, Astrophysics, Economics and Computer Science -- this ratio will be quite high. But those are not OA's problem disciplines, because they are depositing already, whether centrally or locally, unmandated, and have been doing so for years. OA's problem is all the disciplines that are not doing so, for those are the main basis of the 85% emptiness of IRs.

(4) The reason all this matters -- and the reason it is so important not to conflate direct-deposit CRs with harvested CRs, nor to conflate deposit locus with search locus -- is that the locus-of-deposit issue is very deeply interrelated with the issue of mandates.

(5) Deposit mandates can be funder mandates or institutional mandates.

(6) Funder mandates only cover funded research, and not all (perhaps not even most) research output is funded; moreover, this would be true even if all funders already mandated OA.

(7) In contrast, (virtually) all research output (and hence all of OA's target content) is institutional. Institutions are the universal research providers.

(8) So if all institutions mandated OA, that would generate universal OA.

(9) Hence if all of OA's target content is institutional output, it follows that, inasmuch as the 85% of research that is not being deposited spontaneously will be deposited once it is mandated, what is most needed is universal institutional OA mandates.

(10) Funder mandates already help, for their subset of OA's total target content, but they would help far more if they could facilitate the deposit not only of the research they fund, but all research: in other words, if funder mandates could help induce institutions, too, to mandate OA, for all of their own research output, not just the subset mandated by the funder.

(11) In order to be able to do this, funder mandates need only standardize one implementational detail, one that does not lose any of their own target content, but has the potential to extend the reach of the funder mandate to touch the rest of the research output of each one of its fundees' institutions.

(12) Funders need to stipulate the fundee's own IR as the preferred locus-of-deposit for complying with the funder's deposit mandate (with an interim backup repository like DEPOT -- which was created to host deposits until the depositor's institution sets up an IR of its own, to which the DEPOT deposits can then be automatically exported: currently, DEPOT has had only 66 deposits in its nearly 2 years of existence, and that is because most UK funders are either requiring CR deposit or leaving it open which repository their choose fundees deposit in).

(13) The contents can be harvested to CRs from IRs.

(14) The issue of search and functionality at the harvester level is a red herring. (Citeseerx is a perfect example of the functionality of a CR that harvests from distributed IRs.)

(15) Nor do the special features of the few disciplines -- such as computer science, the first, and physics and economics, which took spontaneously to self-archiving long ago, without waiting for a mandate -- have anything to do with either (a) the IR/CR issue, or (b) viable alternatives to mandates, because no one has so far demonstrated any alternatives (apart from waiting and waiting) that can generate the 85% of content missing from IRs, and from OA as a whole.

Stevan Harnad
American Scientist Open Access Forum

SUMMARY: In “Authors: I don’t care where you deposit, just do it”, Gavin Baker suggests that it does not matter where authors deposit their papers to make them Open Access (OA): in an Institutional Repository (IR) or a Central Repository (CR). Nor, more importantly, does it matter where authors' funders mandate that they should deposit them, because IR deposits can be harvested to CRs and vice versa. I point out that this apparent symmetry between IRs and CRs with respect to the harvestability from one to the other (in either direction) is irrelevant today because most of the target content for OA is not yet even being deposited at all, anywhere: In other words, authors are decidedly not "just doing it." Nor are institutions -- the universal providers of OA's target output, both funded and unfunded, across all fields -- "just mandating that authors do it." Apart from the tiny number (about 30) that have already mandated deposit, institutions are the "slumbering giant" of OA, until they wake up and mandate the deposit of their own research output in their own IRs. Not all research output is funded, but all research output is institutional: Hence institutions are the universal providers of all OA's target content. Although not many funders mandate deposit either, the few that already do (about 30) can help wake the slumbering giant, because one funder mandate impinges on the research output of fundees at many different institutions. But there is a fundamental underlying asymmetry governing where funders should mandate deposit: As Prof. Bernard Rentier (founder of EuropenScholar and Rector of U. Liège, one of the first universities to adopt an institutional deposit mandate) has recently stressed, convergent funder mandates that require deposit in the fundee's own IR will facilitate the adoption of deposit mandates by institutions (the slumbering giant), whereas divergent funder mandates that require CR deposit (or are indifferent between CR and IR deposit) will only capture the research they fund, while needlessly handicapping (or missing the opportunity to facilitate) efforts to get institutional deposit mandates adopted and complied with too. The optimal solution for both institutions and funders is therefore: "Deposit institutionally, harvest centrally".

---

The well-meaning recommendations of Gavin Baker (GB) in “Authors: I don’t care where you deposit, just do it” are based on some implicit misunderstandings and unrealistic assumptions about the status quo today (OA < 15%) and what is practically required to get us from that status quo to 100% OA, at long last.

Locus-of-Deposit Matters for OA Mandates. Getting us to 100% OA at long last is precisely what OA deposit mandates are all about. And that in turn is what makes locus-of-deposit the crucial matter that Prof. Rentier has fully recognized it to be: The lesson for GB is that if you care about getting authors to deposit, you need to care about where you mandate deposit!

Symmetry of IR/CR InterHarvestability Is Irrelevant Without Deposits, and Deposit Mandates. GB makes the simple, indeed simplistic assumption of symmetry, insofar as harvestability is concerned: There are Institutional Repositories (IRs) and Central Repositories (CRs) and they are all interoperable, hence inter-harvestable: “What difference does it make where the OA content is deposited? It can be harvested in either direction!” The difference is very simple: Most of it is not deposited anywhere, and what we are talking about is choosing the fastest and surest policy to get it all deposited (and hence OA).

Next to No Deposits Without Deposit Mandates. Everyone (including GB) is at last coming to realize that the way to get OA's target content to be deposited and thereby made OA is by adopting deposit mandates, because most researchers (85%) are simply not depositing, whereas most researchers say they will deposit, willingly, if mandated; and it turns out that when it is indeed mandated, most researchers actually do deposit.

Who Should Mandate Deposit, Where? The apparent symmetry of cross-harvestability to/from IRs and CRs (once everything is indeed deposited) certainly does not answer these two questions. A closer examination of the question of who is to mandate deposit, and where discloses profound and interrelated asymmetries. The two candidates for mandating deposit are authors' employers (i.e., their institutions) and authors' funders. And the two candidate loci for deposit are the authors' own institutions (IRs) or various thematic or geographic collections (CRs).

Not All Research Is Funded, But All Research Is Institutional. There are tens of thousands of research-active universities, research institutions, and major R&D organizations worldwide. These are the "institutions" and they generate virtually all of OA's primary target content: refereed journal articles. There are also a few thousand major research funders, public and private, that fund some (but by no means all) research output.

Asymmetry I: Funder Mandates Only Cover Part of OA’s Target Content. And here we have the first profound asymmetry that GB overlooks: All of OA's target output is produced by institutions (unaffiliated researchers represent such a minute fraction of the total that it would be absurd to base overall OA policy on this tiny special case). But not all -- in fact perhaps not even most -- of OA's target output is funded. So if CRs are to be filled by funder mandates, they will not be filled with all and perhaps not even most of OA's target content even if all funders were to mandate it -- and, crucially, most funders still don't mandate deposit at all.

The “slumbering Giant”: Most Institutions Do Not Yet Mandate Deposit. The fact that most institutions don’t mandate is the practical gist of the entire locus-of-deposit issue: How does locus-of-deposit affect probability-of-mandate adoption (and compliance)? Without the mandates (and mandate compliance), we do not have the OA.

Funder Mandates Can Wake the slumbering Giant – If They Stipulate Institutional IRs As Locus-of-Deposit. Now the coverage-asymmetry we have already pointed out -- that all research originates from institutions but not all research is funded -- already suggests that there may not be parity between IRs and CRs: If funders were the only ones to mandate deposit, and they either stipulated that papers had to be deposited in a CR, or they left it open-ended whether they were deposited in a CR or an IR, then that would leave all unfunded (institutional!) content undeposited, and no form of CR-to-IR harvesting could fill the IRs with their missing content. This would remain true even if (mirabile dictu) all funders were to mandate deposit. It is the institution/funder, IR/CR coverage asymmetry, which is irremediable by harvesting.

Institutions Are the Universal OA Providers. Now note that the reverse is not true, and this is another aspect of the coverage-asymmetry: If (mirabile dictu) all institutions were to mandate deposit, then that would indeed generate 100% OA, and funders and CRs would indeed be able to harvest all their target content from the CRs. The trouble is, and remains, that all institutions do not yet mandate deposit; only a tiny number of them (about 30) do today.

Funder Mandates Reach and Bind Multiple Distributed Institutions. It is likewise true that only a tiny number of funders (also about 30) as yet mandate deposit. (Perfect symmetry so far.) But because the number of funders is far smaller than the number of institutions, and because some of the mandating funders are rather big ones, the actual proportion of mandating funders (as well as the amount and proportion of OA content that they cover) is substantially higher than the current proportion of mandating institutions (and the amount and proportion of OA content that they cover).

Asymmetry II: Funder Mandates Can Jump-Start Institutional Mandates – If Locus-of-Deposit Is Convergent Rather than Divergent. This is the second IR/CR asymmetry, and the most important one, for with a simple change in the stipulated locus-of-deposit of these few funder mandates (and future ones) -- from CR or IR/CR deposit to IR deposit -- funder mandates can, at no loss to themselves, help to jump-start a burst of IRs and IR mandates from the “slumbering giant,” the distributed worldwide set of institutions: the universal providers of all OA content. Every single funder mandate in fact touches on many institutions: the institutions of each one of its fundees. If funders mandate CR deposit only, or even if they leave it open whether deposit is in a CR or an IR, they miss the opportunity to systematically motivate and facilitate IR-creation and mandate-adoption by institutions. Suppose funders instead stipulate the fundee's own IR as the designated locus of deposit (allowing, where needed, the option of depositing in interim back-up repositories like DEPOT, [currently languishing with only 66 deposits!], explicitly designed to host deposits temporarily for institutions that have not yet set up IRs of their own, ready to be exported to the depositor's institution automatically as soon as they have set up their own IR):

Funders Motivate Institutions to Mandate and Authors to Comply. The result of funder mandates designation the fundee's IR as locus of deposit is that funders greatly increase the potential scope and power of their deposit mandates, far beyond only their own funded output: With institutional authors depositing funder-mandated papers in their own IRs, not only are authors thereby motivated to deposit the rest of their output in their IRs too, but their institution is motivated to capture the rest of its institutional output by mandating deposit too. (The slumbering giant awakes!)

Collaboration and Convergence Versus Competition and Divergence Between Institutional and Funder Mandates. If funder mandates thus collaborate and converge on locus-of-deposit with institutional mandates, by stipulating IR deposit (at the unitary local source: the universal provider), instead of competing and diverging needlessly on locus of deposit, nothing whatsoever is lost and everything is gained. Automatic harvesting is indeed there, ready to collect funder-mandated IR deposits into the funder's preferred CRs where desired, because the deposits themselves have been done: Institutions are well placed and only too willing to cooperate in monitoring and ensuring compliance with the grant fulfillment conditions of the funders of their researchers. And the more that institutions are involved in seeing to it that their researchers do their depositing in compliance with funder requirements, the sooner the token will fall for them, so they realize that the very same solution can and should be applied to all their institutional research output, funded and unfunded, in all areas, by adopting an institutional mandate too.

Institution/Funder Complementarity and Synergy in Reaching Universal OA. It is from the fundamental asymmetry underlying the question of locus-of-deposit and direction of harvesting that the institution/funder complementarity and synergy emerges: It has next to nothing to do with the symmetry of harvestability, given the deposits, and everything to do with ensuring that the deposits actually get done! (And this is without even mentioning the third asymmetry underlying the absurdity of institutions having to back-harvest their own institutional output because it has been deposited willy-nilly institution-externally instead of being deposited institutionally in the first place, and then automatically exported wherever else desired.)

I will reply to the rest of GB's posting via quote/commentary:

If requiring deposit is left to universities, their repositories will only contain publications by their researchers.

(1) But the total of all institutional research output is the total of all research output, hence 100% of OA’s target content. (“Institutions” includes all universities, research institutes, and major R&D organizations.)

(2) Not so for the total of all funded output.

(3) And the fundamental problem for OA today is that the institutions – the slumbering giant, the universal providers – do not yet require deposit. Only a tiny number of them do.

(4) If (as is simply assumed here) all institutions did mandate deposit for all of their research article output, we would have 100% OA.

(5) The issue, therefore, is: How can locus-of-deposit be used to help induce all institutions to mandate deposit?

Since some researchers have multiple institutional affiliations, and since any given publication may be authored by researchers across multiple institutions, it is easy to see that researchers will ultimately have to deposit their publications in as many repositories as there are institutions involved in their research.

We are talking about one, single locus of direct deposit, particularly for funder-mandated deposits, and whether that one deposit should be in the author’s own IR or a non-institutional CR, if the purpose is not just to deposit that one paper to make it OA, but to wake the slumbering giant to mandate depositing all institutional output to make it OA.

For an author affiliated with multiple institutions, that still only means a one-time IR deposit, in the IR of one of his institutions, followed by export to the IRs of the rest of his institutions. That has nothing to do with the question of symmetry for the locus of deposit of funder mandates that cover multiple institutions.

This trend seems to rest on the naive notion that, in the Internet era, it is somehow still necessary for researchers to conduct their work solely through the channels of a university.

It is not clear what futuristic scenario GB may have in mind, but the reality today is that researchers are employed and paid by their institutions to conduct and report their (funded and unfunded) research.

It is understandable that universities may wish to host a complete collection of the research published by their faculty, but nowadays that can easily be accomplished by importing it automatically from the more complete collections of the distributed Web.

Yes, just as manna from heaven could be harvested to feed the hungry! The trouble is, most research (85%) is not yet deposited anywhere on the distributed Web today, hence nowhere to be harvested to or from, because deposit has not yet been mandated. We are talking here about how locus-of-deposit, particularly for funder mandates, can increase the probability of waking the slumbering giant (institutions) to mandate deposit of all of their output, funded and unfunded, across all disciplines, institutions, languages and nations. That way we might one day indeed have a “complete collection” of OA research output globally.

Recall also that universities are not the universal providers of all research output. There will always be independent scholars, as well as publications by authors in government, non-profits and think tanks, and corporations.

Government, non-profits, think-tanks and corporations are all “institutions,” can all create their own (individual or consortial) IRs, and can mandate deposit in them. And funders can mandate deposit in those IRs by those of their fundees working at those institutions.

The OA philosophy is global. It cannot be reduced to a single university.

No one is reducing OA to a single university. OA content is distributed across a global network of institutions, the universal providers. And the idea is not reduction but expansion – of the practice and policy of depositing the research output of all those institutions in their own IRs, through collaborative, convergent rather than competing and divergent deposit mandates from both institutions and funders.

Institutions: Require that researchers ensure their publication ends up in your IR, wherever they initially deposited it. If they are accustomed to depositing in a funder or thematic repository, they’re responsible for working with the IR manager to ensure their work gets harvested. Better still if the IR 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 IR manager.

All splendid alternatives for institutions that already mandate deposit! But the trouble is that the giant is still slumbering, and we are talking about how to wake him. If he were already awake -- i.e., if all or most or even many institutions were already mandating deposit for all of their research output -- then locus-of-deposit would be far less crucial, and GB’s symmetry assumptions and alternatives would be perfectly fine.

The trouble is that that is not at all where we are today. The giant is fast asleep; indeed he has grotesquely overslept! And this panorama of alternative ways to do what he is not doing, nor thinking of doing, is not going to wake him up and induce him to do it.

Whereas convergent IR mandates from funders have a good chance of serving as the wake-up call...

Funders: Require 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.

In the current status quo, when only a tiny proportion of funders is mandating deposit at all, and an even tinier proportion of institutions is doing so, this advice is perhaps a tad premature... for existing IR content.

But it would certainly be a good idea to couple IR deposit mandates from funders with clear instructions on how the deposits may be automatically exported to the funder’s preferred CR (if any), as well as how compliance with the deposit mandate is to be verified as a fulfillment condition of the grant, with the collaboration of the fundee’s institution, IR software (for a funder-name metadata tag) and IR management.

This seems more flexible and accommodating to all preferences. In particular, it’s superior to the suggestion that funders should require authors to deposit in their institution’s repository, because [1] not every institution has an IR. Indeed, [2] not every funded researcher may even have an institution. The ultimate goal is opening all research, regardless of where the authors work or who funded the research.

[1] Every institution is just (a) a piece of free software, (b) a $1000 linux server plus (c) a few days of sysad set-up time from having its own IR. In the meanwhile, there are interim alternatives like the DEPOT or multi-institution consortial IR solutions like the Whiterose Consortium.

[2] Unaffiliated researchers are a red herring, when we are talking about a collaborative policy that will greatly enhance the probability of mandating deposit for the overwhelming majority of OA’s target output that is produced by researchers affiliated with an institution. Moreover, universities have a long and admirable tradition of hosting the work of unaffiliated scholars (and some IRs already do).

In summary: The symmetry between Institutional Repositories (IRs) and Central Repositories (CRs) with respect to harvestability the harvestability from one to the other (in either direction) is irrelevant today because most of the target content for Open Access (OA) is not yet being deposited at all. Convergent funder mandates that require IR deposit will facilitate the adoption of deposit mandates by institutions (the universal research providers), whereas divergent funder mandates that require CR deposit (or are indifferent between CR and IR deposit) will needlessly handicap (or fail to facilitate) efforts to get institutional deposit mandates adopted and complied with. The universal, synergistic solution for both institutions and funders is: Deposit institutionally, harvest centrally.

Gavin Baker comments:

"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."

Noble wishes. But alas not being fulfilled -- and it's been a long wait. That's why specifying institutional locus-of-deposit for existing funder-mandates matters: to help make these wishes come true instead of just continuing to hope they will come true of their own accord.

"[P]ractice 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"

Indeed. But exhortations to mandate are not effective either! So this is an exhortation to make a specific modification in existing (funder) mandates to make them more effective in generating further (institutional) mandates.

"Where he’s wrong: Dr. Harnad repeats...“Not All Research Is Funded, But All Research Is Institutional”... he weakens this statement to 'virtually all'..."

Not all researchers have fingers -- only "virtually all" do: That does not mean it cannot be mandated that the deposit keystrokes need to be done.

"[N]ot all academic journal literature is written by authors at colleges, universities, or research institutions"

Fine. This changes nothing regarding the potential benefits of funder mandates stipulating IRs as the default locus of deposit. As noted, where the fundee has no IR, they can and should deposit in a back-up CR such as DEPOT.

"Institutional mandates won’t touch these [unaffiliated] authors (by definition). Funder mandates may touch some, though I’d suspect that most such research is unfunded"

That makes this case even more irrelevant. This is about funder-mandates inducing fundees' institutions to adopt institution-mandates. No IR? Or no Instiutution? Deposit in the DEPOT -- and let's not let this minuscule number of exceptions obscure the potential benefits of funder mandates stipulating institutional deposit as a rule!

"(Harnad notes that some funders do offer “back-up” repositories for grantees without IRs, but without endorsing this flexibility.)"

I did indeed endorse it, many, many times: And, in case of doubt, I endorse it here too.

"Harnad also deftly re-defines the traditional meaning of “institution” in “IR”, to mean “any organization whose employees might ever produce research”... As a result, he reduces further the number of “unaffiliated” researchers"

Yes. And your point is...?

("Every institution is just (a) a piece of free software, (b) a $1000 linux server plus (c) a few days of sysad set-up time away from having its own IR. In the meanwhile, there are interim alternatives like the DEPOT...")

"Doesn’t it make more sense to tell your employees to deposit in a higher-visibility thematic repository?"

Most definitely not. It makes immeasurably more sense to mandate that both fundees and institutional employees deposit in their own IR (or DEPOT) and let high-visibility CRs harvest the deposits thereafter, because that will generate more institutional mandates. That is in fact the very essence of the point I am making.

"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."

That's the symmetry-fallacy again, which is exactly what I was refuting: There's no point "calling for" ("exhorting") the export of nonexistent (because undeposited) content. The point here is about funder-mandates designating institutional deposit in order to generate institutional-mandates (waking the "slumbering giant", "more bang for the research buck," or whatever you want to call it) so as to generate all that missing content!.

"Harnad also understates the resource requirements for adequately maintaining an IR. I’ll let Dorothea Salo answer this point."

Umm -- I'd rather this point was answered by the managers of the few successful, well-stocked IRs that are actually doing what OA IRs are for -- rather than doing anything but (and at high cost)...

"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."

No, institutional mandates would offer broader coverage (indeed, universal coverage, if universally adopted). But only 30 have been adopted so far; and so what we are talking about here is how funder-mandates stipulating institutional deposit could induce more institutions to mandate deposit for the rest of their total output, funded and unfunded.

"(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.)"

Again, the point is not that a funder generates more research than a single institution (though most do), but that a funder generates research at many institutions. That's what will give them "more bang for the buck" if they mandate IR rather than CR deposit. (And we are talking about making this small but important change in existing funder-mandates, not about just wishing for more funder-mandates.)

"[D]o 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."

Sounds like you are and are not sanguine, Gavin!

And funders do have enormous "leverage" with their fundee institutions. But we are not talking about direct leverage to require fundee institututions to create IRs or adopt deposit mandates -- just the indirect leverage that would result from specifying IRs instead of CRs as the funder-mandate's default locus of deposit.

"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."

Agreed! So what are we arguing about?

(Yes, "repositories of last resort" do need to provided -- though it's not clear why they have to be provided by the funder; nor how many of them are really needed. There's plenty of room, for example, in DEPOT, which currently has only 66 deposits after almost two years of existence, simply because funders are mandating direct deposit in their own CRs instead of the fundee's IR -- for which DEPOT was specifically created to serve as the back-up.)

"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."

Here's the way: import/export/harvest them from the IRs (and DEPOTs). And the fundee institutions will be more than happy to collaborate in monitoring and ensuring compliance with the funder's grant fulfillment conditions.

"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."

Yup, that's it! So what are we arguing about, Gavin?

"[Or] Encourage grantees to deposit in their IR, if they have one."

Now that, in contrast, would be useless, just as encouraging ("exhorting," "calling for," "wishing") authors to deposit -- or funders or institutions to mandate -- is useless. What we are talking about is a tiny but concrete and specific change in the implementational details of actual, adopted funder mandates, so as to require institutional (or, as last resort, DEPOT) deposit, in order to generate more institutional mandates. Merely "encouraging" it will generate nothing.

"[T]he risk of backlash if IRs are poorly managed, a factor over which the funder can have little control..."

Why are we speculating about the possibility of full, ill-managed IRs when what we are actually faced with is virtually all IRs (whether well-managed or ill) empty! Necessity is the mother of invention: Fill those IRs with their valuable intended content, and institutions will nurture them as the invaluable assets they will prove to be. Keep fussing instead about the current or future management of idle or nonexistent IRs and you will have years more of idle or nonexistent IRs.

"Funder repositories come with their own cost: the increased potential for government interference in science that comes with greater centralization"

Let's not get carried away! We're talking about published journal articles, made openly accessible, free for all. If they are deposited in each researcher's distributed IRs, they can thence be harvested into multiple CRs. If there is government "interference" (whatever that means) in governmental CRs, there will be plenty of other harvested CRs (e.g. google scholar!) where those came from.

Stevan Harnad
American Scientist Open Access Forum

      

Funder mandates from Canada (FRSQ), Norway (NRC) and Spain (Asturias) now bring the planetary total of Green OA Mandates to 65! (Three new ones in one day; thanks to Peter Suber for news of two of them.)