Self-Archiving Mandates

Alas, the well-meaning Euroscientist article "Open access in Europe: the bear and the tortoise" is replete with the most common misunderstandings of open access (OA) -- the very same misunderstandings that have kept OA from happening for so many years since it first came within reach.

Despite many hopeful announcements, no OA tipping point has yet been reached. (It isn't even clear what "tipping point" means, if it doesn't mean crossing a threshold as of which 100% OA is within sight and fast approaching.)

ERC "joining" Arxiv means providing partial payment to support the costs of a global repository that has been at the disposition of researchers worldwide since 1991. But it's still only those researchers (mostly physicists and mathematicians) who have been depositing in Arxiv all along who continue to deposit in Arxiv. No tipping point in sight there, for the rest of the disciplines and the rest of the world.

If 62% of ERC-funded articles are OA it is because ERC has mandated OA (but the figure needs to distinguish OA itself, which needs to be immediate, from Delayed Access, which might be 6-12-24 months or even longer, after publication).

The EU Horizon 2020 Framework, too, must clarify and shore up its mandate on the question of the timing of the deposit as well as the timing of access.

Recommendations, Declarations, Statements, Invitations and Incentives to provide OA are very welcome, but alas they do not generate OA itself. Only effective OA mandates, adopted and implemented by research institutions, research funders and universities generate OA. And OA mandates are still few and (more to the point): far too weak (see ROARMAP).

No, the real problem is not the possibility that publishers' copyright agreements with authors can still embargo OA for 6-12-24 months or longer. The problem is that most OA mandates fail to mandate immediate deposit anyway, irrespective of how long they allow access to the immediate-deposit to be embargoed by the publisher. Once authors have done an immediate-deposit, the repositories have a Button that makes it possible to provide almost-immediate almost-OA during any allowable embargo period with one click from the would-user and one click from the author.

The UK, the worldwide OA leader since 2004, has not taken "significant steps" forward on OA recently, but significant steps backward. (The Finch Report and the new RCUK OA mandate "prefers" double-paying to publish in gold OA journals instead of letting UK authors continue to publish in their preferred journals and provide green OA by self-archiving in their institutional repositories). Other countries are in fact doing much better than the UK, most notably Belgium, with the Liège model OA green OA mandate -- the one that all institutions and funders worldwide should be adopting. It requires immediate deposit in the institutional repository as the means of submitting work for research evaluation, and as a condition for research funding.

It is good that the Science Europe Statement favours OA, but as noted, the past decade has demonstrated unequivocally that statements are not enough: Effective green OA mandates (the Liège model) are needed (and the Science Europe Statement is not even a statement in favour of effective green OA mandates). Much more clarity, focus, and specificity are needed in order to get this job done.

And the first step is to stop saying and thinking that the difference between "gold OA" (publishing) and "green OA" self-archiving is that gold means instant OA whereas green means OA within 6 months: Gold OA requires authors to change journals and pay to publish. Green OA allows authors to continue to publish where they choose, at no cost, and to provide immediate Almost-OA regardless of whether and how long a publisher OA embargo is allowed. (And 60% of publishers do not embargo OA at all.)

Yes, developing countries could in principle outpace the EU and the US in providing OA to their own research output, but what both the developing countries and the EU and US need most is access to all of one another's research output -- and most urgently to the research output of EU and the US. Moreover, most developing countries are not yet outpacing the EU and the US in providing OA to their own research output.

The obstacles to OA have nothing to do with the (legitimate) need and desire of researchers to meet the quality standards of the top journals in their fields. And the solution is not just "incentives and support" (already tried many times, many places) but the universal adoption of an effective OA mandate, which is the Liège mandate -- and green.

Alessandro Sarretta asked:

---

"In a blog post you said that 'Elsevier is fully green - Fully green means the refereed, revised, accepted final draft is openly accessible'

"This is correct reading this Elsevier web page that seems an old one._

"But reading the current Article Elsevier posting policy web page, it says:

Pre-print Definition: A preprint is an author’s own write-up of research results and analysis that has not been peer-reviewed, nor had any other value added to it by a publisher (such as formatting, copy editing, technical enhancement etc...).

Elsevier's Policy: An author may use the preprint for personal use, internal institutional use and for permitted scholarly posting.

In general, Elsevier is permissive with respect to authors and electronic preprints. If an electronic preprint of an article is placed on a public server prior to its submission to an Elsevier journal or where a paper was originally authored as a thesis or dissertation, this is not generally viewed by Elsevier as “prior publication” and therefore Elsevier will not require authors to remove electronic preprints of an article from public servers should the article be accepted for publication in an Elsevier journal.

"So the pre-print shall "not have been peer-reviewed"... Could you share your opinion on that?"

---

What matters is the postprint (the "Accepted Author Manuscript [AAM]") not the unrefereed preprint.

Elsevier believes that individual authors should be able to distribute their AAMs [Accepted Author Manuscripts] for their personal voluntary needs and interests, e.g. posting to their websites or their institution’s repository, e-mailing to colleagues. However, our policies differ regarding the systematic aggregation or distribution of AAMs... Therefore, deposit in, or posting to, subject-oriented or centralized repositories (such as PubMed Central), or institutional repositories with systematic posting mandates is permitted only under specific agreements between Elsevier and the repository, agency or institution, and only consistent with the publisher’s policies concerning such repositories. Voluntary posting of AAMs in the arXiv subject repository is permitted.

Please see my prior analyses of this Elsevier double-talk about authors retaining the right to make their AAMs OA in their institutional repositories "voluntarily," but not if their institutions mandate it "systematically." Here's a summary:

1. The author-side distinction between an author's self-archiving voluntarily and mandatorily is pseudo-legal nonsense: All authors can assert, safely and truthfully, that whatever they do, they do "voluntarily."

2. The institution-side distinction between voluntary and "systematic" self-archiving by authors has nothing to do with rights agreements between the author and Elsevier: It is an attempt by Elsevier to create a contingency between (a) its "Big Deal" journal pricing negotiations with an institution and (b) that institution's self-archiving policies. Institutions should of course decline to discuss their self-archiving policies in any way in their pricing negotiations with any publisher.

3. "Systematicity" (if it means anything at all) means systematically collecting, reconstructing and republishing the contents of a journal -- presumably on the part of a rival, free-riding publisher hurting the original publisher's revenues; this would constitute a copyright violation on the part of the rival systematic, free-riding publisher, not the author: An institution does nothing of the sort (any more than an individual self-archiving author does). The institutional repository contains only the institution's own tiny random fragment of any individual journal's annual contents. (ArXiv, in contrast, unlike an institutional repository, is indeed a systematic collection of all or almost all the articles in a number of physics and maths journals: Elsevier hence endorses Green OA self-archiving in arXiv, because although it is systematic, it is "voluntary." The pseudo-distinction is hence that although Green OA self-archiving in a mandated institutional repository is not systematic, it can be embargoed despite the fact that all Elsevier authors retain the right to self-archive in their institutional repositories, because it is not "voluntary.")

All of the above is in any case completely mooted if an institution adopts the ID/OA mandate, because that mandate only requires that the deposit be made immediately, not that it be made OA immediately. (If the author wishes to comply with a publisher OA embargo policy -- which Elsevier does not have -- the repository's "Almost-OA" eprint-request Button can tide over researcher needs during any OA embargo with one click from the requestor and one click from the author.)

End of the gold rush? (Yvonne Morris, cilip): "In the interest of making research outputs publicly available; shorter and consistent or no embargo periods are the desired outcome. However, publishers… have argued that short embargo periods make librarians cancel subscriptions to their journals… The BIS report finds no evidence to support this distinction."

---

I have long meant to comment on a frequent contradiction that keeps being voiced by OA advocates and opponents alike:

I. Call for Disruption: Serial publications are overpriced and unaffordable; publisher profits are excessive; the subscription (license) model is unsustainable: the subscription model needs to be disrupted in order to force it to evolve toward Gold OA.

II. Call for Protection: Serials publications are threatened by (Green) OA, which risks making the subscription model unsustainable: the subscription model needs to be protected in order to allow it to evolve toward Gold OA.

Green OA mandates do two things: (a) They provide immediate OA for all who cannot afford subscription access, and (b) they disrupt the subscription model.

Green OA embargoes do two things: (c) They withhold OA from all who cannot afford subscription access, and (d) they protect the subscription model from disruption.

Why do those OA advocates who are working for (a) (i.e., to provide immediate OA for all who cannot afford subscription access) also feel beholden to promise (d) (i.e. to protect the subscription model from disruption)?

University of Liège and FRSN Belgium have adopted -- and HEFCE and BIS (as well as BOAI-10: 1.1 & 1.6 and the Harvard Open Access Project (HOAP)) have all proposed adopting -- the compromise resolution to this contradiction:

Mandate the immediate repository deposit of the final refereed draft of all articles immediately upon acceptance for publication, but if the author wishes to comply with a publisher embargo on Green OA, do not require access to the deposit to be made OA immediately: Let the deposit be made Closed Access during the allowable embargo period and let the repository's automated eprint-request Button tide over the needs of research and researchers by making it easy for users to request and authors to provide a copy for research purposes with one click each.

This tides over research needs during the embargo. If it still disrupts serials publication and makes subscriptions unsustainable, chances are that it's time for publishers to phase out the products and services for which there is no longer a market in the online era and evolve instead toward something more in line with the real needs of the PostGutenberg research community.

Evolution and adaptation never occur except under the (disruptive) pressure of necessity. Is there any reason to protect the journal publishing industry from evolutionary pressure, at the expense of research progress?

Stevan Harnad

The Director General of the UK's Russell Group of universities, Wendy Piatt, responded as follows to the BIS Committee Recommendations on UK OA policy:

“We welcome the committee’s report, which highlights the vital role that ‘green’ open access can play as the UK moves towards full open access... [T]he committee rightly highlights that ‘green’ open access is a simple and cost-effective way of sharing research. We urge the Government to take note of the calls to reconsider its preference for ‘gold’ open access during the five year transition period."
      “However, we have real reservations about the committee’s recommendation to restrict embargo periods to six months for STEM subjects and 12 months for humanities, arts and social sciences. This will directly limit where researchers can publish, will constrain academic freedom and could potentially damage the international standing of UK universities.”

The substantive recommendations of the 2013 BIS Report (I, II) were:

1. that the Green OA deposit in the institutional repository should be immediate rather than delayed, whether or not Open Access to the deposit is embargoed by the publisher (during any OA embargo the repository's eprint-request Button can then enable the author to fulfill individual user eprint requests automatically with one click each if deposit was immediate),

2. that an effective mechanism for monitoring and ensuring timely mandate compliance should be implemented, and

3. that Gold OA publishing should either no longer be preferred or hybrid Gold should no longer be funded.

The only bearing these three recommendations have on author freedom-of-choice of journal is that they restore it: Whereas Finch/RCUK's current preference for publishing in Gold OA journals would have restricted author freedom of choice, BIS removes that restriction. Authors can publish in whatever journal they wish.

As to the BIS suggestion to reduce the maximum allowable Green OA embargo limit to 6-12 months from 12-24 months: the immediate-deposit mandate (plus the repository Button) largely moots this. As long as deposit is immediate, there would be nothing wrong with allowing an opt-out or waiver (as in the American opt-out mandate models, such as Harvard's) from the maximal embargo limit on an individual article basis, if the author demonstrates that it would restrict journal choice to have to comply with the embargo. The critical thing is that there should be no opt-out from immediate-deposit (which, if not immediately OA, has no bearing whatsoever on author freedom of choice of journals). Authors can publish in whatever journal they wish.

One could hardly have hoped for a better outcome from the Business, Innovation and Skills Committee's Report. If BIS's recommendations are followed then the UK will regain its global leadership role in the Open Access movement -- the role the UK has been playing ever since the pioneering Report in 2004 by Ian Gibson's Parliamentary Select Committee on Science and Technology. That Report had recommended that UK's universities and funding councils should mandate Green OA self-archiving of all peer-reviewed research articles. In the ensuing years more and more of the rest of the world began to follow suit.

The 2013 BIS Report (I, II) now recommends mandating;

1. that the Green OA deposit in the institutional repository should be immediate rather than delayed, whether or not Open Access to the deposit is embargoed by the publisher (during any OA embargo the repository's eprint-request Button can then enable the author to fulfill individual user eprint requests automatically with one click each if deposit was immediate),

2. that an effective mechanism for monitoring and ensuring timely mandate compliance should be implemented, and

3. that Gold OA publishing should either no longer be preferred or hybrid Gold should no longer be funded.

The BIS recommendations now perfectly complement HEFCE's recommendation (as well as those of BOAI-10: 1.6) to make immediate-deposit a condition for eligibility for REF 2020 (thereby effectively recruiting universities to serve as the mechanism for ensuring timely compliance, following the highly successful mandate model of the University of Liège). This effectively fixes the flaws in the Finch Report. The UK's OA policy will now also be compatible with OA policies in the EU, the US and the rest of the world, doing them all one better with its explicit call for immediate institutional deposit and effective compliance monitoring.

Business, Innovation and Skills Committee Select Committee
Announcement No.21
Monday 9 September 2013

NEW REPORT: Open Access: achieving a functional market
Under strict embargo until 00.01 BST on Tuesday 10 September

Government mistaken in focusing on Gold as route to full open access, says Committee

The Government’s commitment to increasing access to published research findings, and its desire to achieve full open access, are welcome, says the Business, Innovation and Skills Committee in a Report published today. However, whilst Gold open access is a desirable ultimate goal, focusing on it during the transition to a fully open access world is a mistake, says the Report.

The Report calls on the Government and RCUK to reconsider their preference for Gold open access during the five year transition period, and give due regard to the evidence of the vital role that Green open access and repositories have to play as the UK moves towards full open access.

The Report recommends that:

• the Government should take an active role in promoting standardisation and compliance across subject and institutional repositories [paragraph 25].

• RCUK should reinstate and strengthen the immediate deposit mandate in its original policy and improve the monitoring and enforcement of mandated deposit [paragraph 31].

• the Government and RCUK should revise their policies to place an upper limit of 6 month embargoes on STEM subject research and up to 12 month embargoes for HASS subject research [paragraph 50].

• the Government should mitigate against the impact on universities of paying Article Processing Charges out of their own reserves [paragraph 64].

• if the preference for Gold is maintained, the Government and RCUK should amend their policies so that APCs are only paid to publishers of pure Gold rather than hybrid journals to eliminate the risk of double-dipping [paragraph 77].

There’s nothing wrong with OA growth in Australia ("Four issues restricting widespread green OA in Australia") that the adoption of the Queensland University of Technology's [QUT's] Green OA self-archiving mandate model by all Australian universities and research funding councils would not fix.

Issue 1 – Lack of data about what Australian research is available OA. The problem is not with knowing what’s OA in Australia. (Well configured repository software plus ROAR will tell you that -- and Google will find it.) A mandate compliance monitoring mechanism, however, is indeed needed. But the ones to monitor compliance are authors’ own institutions, by requiring deposit immediately upon acceptance for publication, time-stamped within days or weeks of the date of the acceptance letter, for all published articles. Immediate-deposit should be a condition for Australian national research funding and performance evaluation [ARC] as well as for institutional research performance assessment, as it is in Belgium by FRS-FNRS and the University of Liege, and as it has been proposed for UK funded research by HEFCE for REF 2020.

Issue 2 – Copyright transfer agreements. It’s always good to agree on fair copyright agreements, but trying to convince publishers to agree to those should on no account be holding up the mandating and provision of Green OA. And for journals that embargo OA, there’s always the immediate-deposit mandate and the repository’s eprint-request Button to provide immediate Almost-OA with one click from the requester and one click from the author.

Issue 3 – The academic reward system. The “academic reward system” is certainly not holding up OA. OA increases research uptake and impact, including citations. And the notion that OA needs some sort of preferential treatment for Gold OA journals, rather than just weighting them based on their track-record for quality, like all other journals, is and has always been complete nonsense, ever since it began to be mooted over a decade ago. The way to provide OA is to publish in the highest standard journal possible for one’s work, and then self-archive the refereed final draft. To pay to publish in a Gold OA journal just because it is OA (rather than because of its quality standards) is to pay for Fools Gold. (There is no OA problem for unrefereed or unpublished work; nor is getting academic credit for such work an OA problem.)

Issue 4 – Improved national discovery services. Discovery tools can always be improved, but they are already pretty powerful. They will not discover OA content that is not there. Hence the only thing that is really needed for OA is effective Green OA mandates, along with effective monitoring of compliance, in order to get it up there, out in the OApen, to be “discovered.”

Yes, there's a flaw in the University of California Open Access (OA) mandate ["Open Contradictions," Editorial, The Daily Californian, 12 Aug 2013], and, yes, it has to do with the fact that U of C authors can opt out of compliance with the mandate.

But, no, the flaw is not that the U of C policy

"allows [authors] to pick and choose where their research goes, thereby creating a divide between those who can afford access to a private academic journal and those who cannot."

That researchers retain their right to choose the most suitable journal for their research is not a flaw but a virtue of any OA mandate. (In the UK, some OA mandates are trying to force authors to choose (and pay) to publish in journals based on the journal's OA policy instead of its quality (peer-review) standards, and that's very bad for both research and researchers -- and certainly unnecessary for OA.)

Nor does the U of C's mandate flaw have anything to do with whether the journal is "private" or "academic." Journals differ in their subject matter and quality standards; there are non-profit and for-profit journals at all quality levels, and that in turn has next to nothing to do with the journal's OA policy -- except that there is a new breed of junk journals that has lately been created on the cheap to provide pay-to-publish OA with low or no peer review quality standards. (See Beall's list.)

There are two ways for authors to provide OA: (1) publish in an OA journal that makes the article OA, often for a fee (this is called "Gold OA") or (2) publish in any journal, but also deposit the final, peer-reviewed draft in the author's institutional repository -- U of C's is called eScholarship -- and set access to the deposit as OA (this is called "Green OA") rather than Closed Access.

U of C has (wisely) mandated Green OA, not (like the UK) Gold OA. Hence journal choice is not at issue for U of C authors: They retain their right to choose to publish in the journal most suitable for their work. What is at issue is whether and when they can make their article OA: Some journals' copyright agreements require authors to embargo OA for 6 months, 12 months, or even longer.

Now we come to the real flaw of the U of C policy: If authors can opt out of the U of C mandate whenever a publisher embargoes OA, this nullifies the mandate and simply allows publishers to continue to determine whether and when articles are made OA.

But there is a very simple and natural solution that moots the publisher OA embargo: U of C needs to add an immediate-deposit clause with no opt-out. This means all authors must deposit their peer-reviewed final drafts in eScholarship immediately upon acceptance for publication whether or not the journal embargoes OA. But in addition, eScholarship should implement the automated Request-Copy Button.

The repository's Button can email one copy of an embargoed deposit to an individual user on request: All it takes is one click from the user to request and one click from the author to fulfill the request. (This is not OA but "Almost-OA.")

Authors retain journal choice, as well as the choice to provide individual access even for embargoed deposits -- but they cannot opt out of immediate-deposit requirement itself: All articles, regardless of journal or journal policy, must be deposited in eScholarship immediately. The author can then either set access to the article as OA immediately, or can use the Button to provide "Almost-OA" during any publisher OA embargo.

Once the one-size-fits-all immediate-deposit mandate (with no opt-out) is adopted by universities and research funders worldwide, not only will it close the "divide between those who can afford access... and those who cannot), but it will help hasten publisher OA embargoes toward their natural, inevitable and well-deserved deaths under the mounting worldwide pressure and demand for immediate OA -- by mandating that all articles must be immediately deposited in repositories and taking publishers out of the loop completely, insofar as mandate compliance is concerned.

None of this can happen if universities continue to allow publishers to decide whether and when authors deposit and provide access, by allowing opt-out from OA mandates.

Aside from the default copyright-reservation mandate with opt-out, always add an immediate-deposit clause without opt-out.

The deposit need not be immediately made OA, but it needs to be deposited in the institutional repository immediately upon acceptance for publication. While access to the deposit is embargoed, the repository can implement the eprint-request Button with which users can request and authors can provide the eprint with one keystroke each.

Deposit should always be done directly by the author (or the author's personal designee: student, research assistant or secretary). It is a big mistake to "submit" the paper instead to the provost's office. At other universities with this style of mandate the provost's office has sat on papers for years instead of depositing them; this is even worse than publication lag or publishers' OA embargoes.

If deposit is instead left to the provost's office, immediate-deposit will not become the natural milestone in the author's research cycle that it needs to become, in order to ensure that the deposit is done at all: The dated acceptance letter from the journal is sent to the author. That sets the date of immediate-deposit and also determines which version is the final, refereed accepted one. The publication date is uncertain and could be as much as a year or more after acceptance.

Mandate deposit immediately upon acceptance for publication, but otherwise, having mandated the N-1 of the author keystrokes required for deposit, in case of embargo, leave the Nth keystroke to the author, in responding to Button-mediated eprint requests.

Put all administrative efforts instead into monitoring mandate compliance -- by systematically collecting the dated acceptance letters instead of the papers themselves, and ensuring that the repository deposit-date is within a few days or weeks of the acceptance date.

See also:

1st-Party Give-Aways Vs. 3rd-Party Rip-Offs
and
Almost-OA: "Frictional Access"

Richard Poynder is absolutely right on every point in his reply to my commentary. (Except possibly one [trivial] one: Richard seems to imply that IEEE has already embargoed the author's final refereed draft. It has not. Richard's is only a speculation that they might, given that some other publishers have done so. Richard is quite right that some other publishers have done so. And his speculation about IEEE may prove correct. But it should be noted that it is still just a prediction…!)

Now to what Richard says about "friction." ("[The request-eprint Button] introduces a different kind of friction into the system, including presumably a time delay...")

The request-eprint Button was created for both EPrints and DSpace repositories in 2006 with six very specific objectives:

1. to make it possible for all institutions and funders to mandate OA without being held back by constraints of copyright renegotiation or embargo length

2. to make it possible for all institutions and funders to mandate the deposit component (if not the OA component, nor the copyright retention component) of any OA policy without the need of an opt-out option

3. to make it possible for authors to provide almost-immediate access ("Almost-OA") to their articles during any OA embargo almost as effectively as by making them immediately OA, thereby maximizing uptake, usage and impact and minimizing losses to research and researchers during any OA embargo period

4. to make it possible for the date of deposit (if not the date of OA) to be dictated by institutions and funders, not by publishers

5. to make it possible for all authors to comply with OA mandates at a fixed, natural, determinate date in their publication cycle and work-flow, and to begin providing OA (or Almost-OA) to their (refereed) findings as early as possible.

6. to make it possible, once all or most institutions and funders worldwide have mandated immediate-deposit, to hasten the inevitable, natural and well-deserved death of all OA embargoes, under the mounting global pressure of OA and its benefits (and author fatigue with the "friction" of having to keep clicking the Button to provide Almost-OA!).

So what the Button introduces is not a delay (the publisher embargoes introduce the delay) but a way to provide access during the delay -- with some "friction" (extra keystrokes for authors) -- but that friction may well help put an end to such gratuitous delays sooner rather than later: Read on...

Publishers embargo (Green) OA in order to prevent their subscription revenue streams from being reduced by the revolutionary technical potential opened up by the online medium for as long as they possibly can, at the cost of research access and impact. The Almost-OA Button and the immediate-deposit mandate were jointly designed long before the Finch Fiasco of 2012, to cover the journals that already had OA embargoes, and any journals that might adopt OA embargoes in the future. It is a prophylactic against OA embargoes.

The perverse (but predictable) effect of Finch/RCUK's reckless new OA policy (of preferring to pay for Gold OA instead of reinforcing the requirement to provide cost-free Green OA) has been to give publishers a much stronger incentive to adopt and lengthen Green OA embargoes beyond RCUK's allowable length limit, and to offer hybrid Gold OA (i.e., keep charging institutions for subscriptions, but allow authors to pay them extra to make their individual article Gold OA) instead, so as to ensure that mandated UK authors are obliged to pay them extra for Gold OA rather than just providing cost-free Green OA.

Immediate-deposit plus the Almost-OA Button will be an antidote to this perverse effect of the Finch/RCUK mandate -- which is why it is so important to adopt HEFCE's proposal to make immediate-deposit mandatory in order to make articles eligible for REF2020.

There is a profound conflict of interest between, on the one side, research, researchers, their institutions, their funders, the vast R&D industry, students, teachers, journalists, and the tax-paying public that funds the research, and, on the other, the publishing industry. Publishing is a service industry that had been performing an essential service to research during the Gutenberg era of print on paper, but is now blocking the natural evolution of research communication in the print-free online era by trying to embargo making refereed research freely accessible to all online.

Publishers will not stop trying to delay the optimal and inevitable for research for as long as possible by embargoing Green OA. OA mandates are the way for the research community to overcome the publishing industry's delay tactics -- and the immediate-deposit mandate plus the Button are their key components.

Harvard/MIT/UC-style copyright-reservation mandates are fine, and welcome, but, as noted, they require opt-out options or authors will not comply. The opt-out is invariably needed for cases where the author's journal of choice insists on embargoing OA beyond the allowable limit and authors (rightly) insist on their journal of choice.

But all that the Harvard/MIT/UC-style copyright-reservation mandates need in order to make them work, optimally, is to add an immediate-deposit requirement (whether or not the article is embargoed), without opt-out. Authors who opt out can then rely on their repository's eprint-request Button to provide Almost-OA during the embargo.

Sale, A., Couture, M., Rodrigues, E., Carr, L. and Harnad, S. (2012) Open Access Mandates and the "Fair Dealing" Button. In: Dynamic Fair Dealing: Creating Canadian Culture Online (Rosemary J. Coombe & Darren Wershler, Eds.)