February 12, 2008

Harvard Faculty Adopts Open-Access Requirement

Harvard University’s Faculty of Arts and Sciences adopted a policy this evening that requires faculty members to allow the university to make their scholarly articles available free online.

Peter Suber, an open-access activist with Public Knowledge, a nonprofit group in Washington, said on his blog that the new policy makes Harvard the first university in the United States to mandate open access to its faculty members’ research publications.

Stuart M. Shieber, a professor of computer science at Harvard who proposed the new policy, said after the vote in a news release that the decision “should be a very powerful message to the academic community that we want and should have more control over how our work is used and disseminated.”

The new policy will allow faculty members to request a waiver, but otherwise they must provide an electronic form of each article to the provost’s office, which will place it in an online repository.

The policy will allow Harvard authors to publish in any journal that permits posting online after publication. According to Mr. Suber, about two-thirds of pay-access journals allow such posting in online repositories. —Lila Guterman

Comments

1. Three very important open access mandates have now been passed within less than two months – the National Institutes of Health, the European Research Council and now Harvard. With the Harvard vote, the door is now open for other universities and colleges in the US to adopt similar policies. The NIH and ERC policies also make adoption of mandates by other government agencies in the US and abroad much more likely. It looks like open archiving of peer-reviewed journal literature now on an irreversible course of expansion.

   — Ray English    Feb 13, 03:58 AM    #

2. When will Harvard and other universities do the same for the TEACHING ACADEMICS? If the researchers must make their research papers open acccess, then it would be nice to see all of the professors who write books when they are paid by the univesity also make their books, lectue notes open access (dowloadable pdfs). What is good for the research faculty, is good for the teaching faculty!!
   Also it will make learning material as accessible in the States as it is for the students in India, China and Russia (where cheap international student editions of most text books are readily available to the students for free or for the cost it takes to print and bind.

   — KJJ    Feb 13, 05:17 AM    #

3. This is confusing. On the one hand, the last paragraph suggests that Harvard will not post articles until after they have been peer reviewed by journals and accepted for publication. On the other hand, the policy itself does not say this and permits Harvard to post articles at any time, even before submission to journals. There is a big difference between the two, and the latter could have damaging consequences especially for Harvard junior faculty whose articles might not be considered for publication if already widely available online. By retaining copyright, this policy would also inhibit the efficient clearance of rights for use of articles for teaching purposes through an agency like the Copyright Clearance Center. Imagine how much more complicated the process will be if every teacher has to go seeking permission from every author whose article is wanted for course use, rather than through a centralized clearinghouse. There is a reason that Congress urged the establishment of such clearinghouses when it passed the 1976 Copyright Act. It seems to me that the folks at Harvard haven’t thought through all the consequences of this policy very well.

   — Sandy Thatcher    Feb 13, 07:23 AM    #

4. My question is simple: can Harvard faculty continue to publish in journals whose copyright policies do not permit online downloads from university servers?

   — Jay Clayton    Feb 13, 08:45 AM    #

5. I guess this article is not as clear as I thought (see comment #3 above); I assumed that “open access” and “free on-line” meant eliminating the need to get copyright permission at all. If one wanted to use one of Harvard faculty’s scholarly articles as a course reading, it sounded to me like you’d just provide a link for students to the Harvard repository.

   — sue    Feb 13, 08:53 AM    #

6. You can find the answers you need about post-publication open access at http://www.arl.org/sparc/author/index.html

   — Cy Dillon    Feb 13, 10:00 AM    #

7. Regarding the comment #5 above, I believe Harvard’s online repositories will have clear notes for the use in course reading and future scholar citing.

   The policy doesn’t limit the time when faculty shall provide the electronic form of the article to the provost’s office, so it gives faculty and publishers the flexibility to set the time frame, at least it makes them have to do so.

   — Shanyun Zhang    Feb 13, 10:29 AM    #

8. re: comment no. 4 – I haven’t looked closely enough to answer the questions, but more and more journals are reading the writing on the wall and making it possible for those who create their content to archive their work online. Even Elsevier allows it. Because access means more influence, the influential journals will want to do this to sustain their importance.

   Boycotting Harvard authors is just not the best way for a journal to have an impact.

   — Barbara Fister    Feb 13, 11:09 AM    #

9. Harvard is soooo far behind the times. All that stuffy right-wing conservatism, I believe. Obviously they haven’t heard of:
   http://ocw.mit.edu

   — Ed Montgomery    Feb 13, 12:10 PM    #

10. Ed Montgomery’s analogy is not exactly correct. MIT OpenCourseWare provides course materials (lecture notes, exams, handouts, etc), but this is entirely different from the peer-reviewed research papers that are the subject of the new Harvard policy.

    — ArtC    Feb 13, 12:28 PM    #

11. In regards to comment number 3, for those of you who do not know, Mr. Thatcher is the President of the American Association of University Publishers:
    http://aaupnet.org/news/staff.html

    — Tint    Feb 13, 12:50 PM    #

12. MIT and a few other universities ALREADY make academic materials available online. Check out http://ocw.mit.edu to see their Open CourseWare offerngs.

    — Russell de Pina    Feb 13, 01:11 PM    #

13. Harvard’s initiative is more like arXiv (scientific e-print repository, established circa 1991) than Open CourseWare.

    — sf    Feb 13, 01:33 PM    #

14. OK, I’ll rephrase my question.

    Let’s say a leading Journal says “no” to a reques to change a restrictive copyright policy. Let’s say they say “we own the content and you cannot provide a free copy of it except under a very strict notion of Fair Use, say as material for a course you, the author, teach.” Let’s say the publisher is a for-profit entity, to make it clearer. Let’s say I decide I still want to publish in this journal. If I worked at Harvard, could Harvard tell me not to publish there? Otherwise, isn’t this just rhetoric, really?

    — Jay Clayton    Feb 13, 02:18 PM    #

15. To clarify my previous comment (and reveal that I am also a member of the Board of Directors of the CCC), let me add that of course teachers may simply include the URL to Harvard’s institutional repository for a version of the article. But the real question is: which version is it? Publishers are not likely to allow posting of the FINAL version of the article as printed in their journals; they have made an investment in the processing (copyediting, etc.) for which they can expect a reasonable return. Elsevier and other publishers (like my press) allow posting of an article AFTER peer review but BEFORE final processing; they do NOT allow posting of the FINAL article as such. If teachers are happy to assign earlier versions to their students to avoid permissions costs, that’s fine; they won’t be the archival versions to which citations should be made. For teaching purposes purely, this may not matter much. It also depends on who pays the permissions: if they get passed along to the students, teachers may well decide to link to the Harvard site; if they are paid out of library e-reserve funds or out of a central administration budget, they may not care and will be happy to use the published versions. If they go for the free versions, of course presses will lose income from permissions, meaning that prices for other press publications will rise accordingly to make up the difference in revenue. You push at one place in the system, and an effect pops up somewhere else. Nothing is COMPLETELY free!

    — Sandy Thatcher    Feb 13, 02:19 PM    #

16. Old paradigm.University creates knowledge. Sells knowledge to publishers. Publishers restrict access and sell knowledge back to University (students). New paradigm. University creates knowledge. University disseminates knowledge when and how it sees fit. This move by Harvard is one small step in this direction

    — kgm    Feb 13, 02:22 PM    #

17. Re: comment 16: What publishers also add to the “knowledge” is the expert peer review that serves as an (admittedly imperfect) quality control mark. Nothing I’ve seen in the Harvard proposal indicates a process that will accomplish this same purpose.

    — sijit    Feb 13, 03:20 PM    #

18. I am not opposed to the Harvard policy if the benefits are truly what the Harvard University Librarian states to be the case – more sales of monographs. For many young scholars publishing their works has become more troublesome and tenure committees in some areas are asking for more than one monograph for promotion. Of course, these monographs might be digital with a print-on-demand option.

    In many ways this policy also underscores the need for University Presses to work with their libraries to foster the best possible solution for the age of accessible scholarship and distribution. Posting a paper on an institutional repository doesn’t always mean that it has been peer-reviewed or presented in a way that allows for permanence in citation. The Harvard policy does at least allow an environment in which to work out many issues that will come up including how these postings will alter peer review. Will the Repository in effect become the Journal of Harvard University Research? If other universities follow, will the new environment speed up a focus on rich meta-data creation to allow for a virtual, consolidated site of academic research that is fully searchable and intuitive.

    Harvard University has now taken on the burden that academic publishers have for ages – delivering timely, peer-reviewed research, that is suited to customers’ (faculty and researchers) needs. There will be a large cost to this burden, but will the cost benefit the entire scholarly communication system by reducing the overall price of scholarship? One could envision a system that disables peer-review in certain fields due to open access postings. The university then must assume all the functional roles (including vetting, formatting, fact checking, etc.) of publishing, not just a few (posting and distribution).

    Further, this policy underscores that lack of relevance many scholars have had for their institutional repositories. Scholars are not posting their works to the repositories. Does Harvard’s policy provide the support mechanisms to help faculty post their research? As a publisher, I am well aware of the need to provide authors with services to get their works published in a timely fashion.

    Finally, is the Harvard’s policy a branding initiative more than anything else? In a very competitive global market, having open access to Harvard scholars’ works is an invaluable marketing tool. Therefore, I would assume that other universities will have to follow suit in some way. Those schools who have the funding pools for these repositories will now be able to move their research to the desktops of global consumers. Distance education opportunities and other academic programs will be sensible add-ons.

    Implementing the plan will now be watched by millions and may cost that much, too.

    — Thomas Bacher, Director, Purdue Press    Feb 13, 04:46 PM    #

19. This policy does not attempt to “accomplish” or replace peer review provided by publishers. It simply states that researchers must publish in a journal that allows free posting online after publication, meaning also after peer review, since no journal worth its salt publishes an article without peer review. Opponents of open access always try to argue that open access policies do away with peer review when they do not.

    — rw    Feb 13, 04:54 PM    #

20. Harvard shouldn’t insist that authors retain copyright. Many journals already allow for authors to post their articles in repositories both pre & post-print with the journal still retaining copyright. You can see how some of this works by looking up certain journal titles at the site http://www.sherpa.ac.uk/romeo.php.

    Peer review is essential, and the open access movement in no way promotes circumventing this. But open access is a political movement with passionate opinions on many sides. Though free access to research is important, we need to understand that just like the peer review process, scholarly journals with their exceptional editors and production staff help to create the culture of high quality that we come to expect from research journals. Scholarly editors and staff add value, important value. In the pursuit of “unlimited access” to everything, we don’t want to lose site of how important the people who dedicate their lives to enriching our understanding of the world through their work in scholarly publishing are.

    — fs    Feb 13, 06:48 PM    #

21. From all I know about Elsevir, this sounds sooooo fishy. More flash than bang, you know?

    — marci    Feb 13, 06:51 PM    #

22. In my field it is quite common for publishers to “allow” you to distribute your work from your personal website (with some stipulations that are basically advertising). However, most of them still disallow distribution from repository websites (arXiv, citeseer, etc.), and this would presumably apply to Harvard’s model as well. But citeseer and similar libraries are staggeringly useful exactly because they automatically cross-reference articles, citations, similar documents, etc., from different sources. Being able to find in 10 seconds everyone who has cited the paper you are reading, and how important those works are, is brilliant progress. Of course, publishing companies are still trying to assert their rights to stand in the way of such improvements.

    I’m pleased to see Harvard’s new policy, but it’s still not quite at the level that will permit cross-referencing sites to catalogue all knowledge. Until publishers allow that, their efforts to protect their revenue stream are directly detrimental to science.

    — Ben    Feb 13, 07:36 PM    #

23. Now, I have to ask.
    What format?
    I hope they’re smart enough to post them in ODF…

    — Zach    Feb 13, 08:30 PM    #

24. As to 3: You may still use the copyreight clearing center and pay, if you prefer that, rather than link to the material directly.

    As to 15: This is a misunderstanding concerning journal pricing. Publishers charge as much as they can (“value pricing”). If the willingness to pay decreases, prices decrease. Journals are not priced on a cost plus basis, else they would be much much cheaper. Simply compare book prices with comparable journal prices and you will see what I mean.

    — Ek    Feb 14, 04:10 AM    #

25. This is great news, and absolutely right. Why should the interested be excluded from viewing these things?

    — Roger Pearse    Feb 14, 05:33 AM    #

26. Let’s be clear about this peer review service that publishers provide. In my discipline (math/engineering), the work of distribution to reviewers is provided at no cost by an associate editor, usually working for a university, who receives the paper directly from the author. The reviews are typically provided by academics, again at no cost.

    So the universities pay for the production of the research paper and the reviewing process. Typically too, in my discipline, the paper is produced by the academic in almost camera ready form, often using a style file provided by the publisher. Almost all sales of academic journals not produced by learned societies, such as IEEE or American Math Soc, are to university libraries. In addition some journals ask for page charges from the author before they will publish the paper.

    — grappler    Feb 14, 05:56 AM    #

27. With the ever increasing number of papers in both NATURE AND SCIENCE being RETRACTED, the peer review policies and practices of many journals, even the top ones appear to be in disarray. Part of the problem is that most refereeing is done for free and anonymously. Mafias of authors and publishers can even start new journals and manipulate the measures being developed to measure scientific output and impact. But science has a good way to weed out the bad works. If the work can not be reproduced, then … But in many of the fields like cosmology, geology. the origin of life, the origin of the universe, and evolution, one can really only speculate and hypothesize. A lot of this work is passed off as hard core rigorous science, when it really is not. And this is seen by so many different and confliciting theories and working models, many of which are not self consistent. We need to train the next generation of researchers and students to be critical thinkers and not to believe everything you read in a scientific text book or reseach article. With the Reformation in Europe, religious scholars were all forced to back up their claims. The same is now being done with many opposing scientific theories and models. By allowing every university to allow all of its academics and researchers to publish all of their works, many more opposing and confliciting works will be published. But again no fear, only the ones which pass the tests of scientific scutiny and testing will prevail. Many journals and leaders in various fields of science have held back science in the past. This will be much harder now, so science should actually develop even faster, the disadvantage being that some shaky works appear for a while, before they are shown to be wrong. But even this may not be bad, as the young researchers will be able do discuss how the methodolgies failed to pick up the flaws in logic, analysis, … Science and Nature have had their share of papers withdrawn recently, so there is no perfect system. Peer review is good, but not perfect either.

    — KJJ    Feb 14, 11:44 AM    #

28. What Harvard has done may go well beyond issues of access. My understanding is that Harvard faculty have in fact voted away control over their own work. The motion voted on states:

    “The Faculty adopts the following policy: Each Faculty member grants to the President and Fellows of Harvard College permission to make available his or her scholarly articles and to exercise the copyright in those articles. In legal terms, the permission granted by each Faculty member is a nonexclusive, irrevocable, worldwide license to exercise any and all rights relating to copyright relating to each of his or her scholarly articles, in any medium, and to authorize others to do the same, provided that the articles are not sold for a profit…. The Dean or the Dean’s designate will waive application of the policy for a particular article upon written request by a Faculty member explaining the need.”

    Faculty have not merely agreed to allow open access to their work post-publication. They have signed away their inherent rights as authors to control the fate of their own work. Seems to me that this means that Harvard —- acting in its own best interests, however that may be determined at the time —- may control the time, place and manner of publication (or non-publication) of the work of individual faculty members.

    Under intellectual property law this is known as an author providing “work-for-hire”. It would reduces scholars to the same status as a copy-writer ad an advertising agency flacking toothpaste. How is this a blow for academic freedom and integrity?

    — Peter    Feb 14, 12:19 PM    #

29. As to 28: You seem to have overlooked that the authors remain the copyright holders and grant a NONEXCLUSIVE right to their university. Their remain the owners, at least of the preprint versions, even if the publishers take away some other rights.

    — Ek    Feb 14, 03:13 PM    #

30. I did notice the term “non-exclusive”. It mitigates slightly, but does not obviate the fact that a transfer of rights has taken place, and in such a manner as to at least call into question how much control would really remain with the original creator of the material —- the academic author. The wording cited says that faculty are granting Harvard not only the right to merely post their articles, but to “exercise any and all rights relating to copyright”. The non-exclusive right to post a PDF of your article is one thing; a non-exclusive right to do anything at all with your work is quite another.

    Here’s an analogy. Let’s say you own a house on a one-acre lot. One day your mortgage company gets you to sign over to them a “non-exclusive right to use or build upon” your property, as long as such use is “not for profit”. No threat, it’s non-exclusive, in a good cause, and after all you’ve still got all your rights, right?

    Then, while you’re away on vacation, the mortgage company licenses its rights (for no fee —- how decent!) to your property to a local day care center (non-profit, of course). They tear down your house and build their new facility. Your “rights” to your own property look pretty hypothetical at that point, don’t they? Now try selling your property to someone else —- think you’ll have many takers?

    I realize that not all uses of publishing rights are so severe as to wipe out the remaining value —- in the terms of the analogy above, other uses might be as minor as allowing local kids to play on your property. If you like kids, that might even seem to be a benefit. But the point is that if you sign a grant of rights that is as broad as Harvard’s appears to be, you as an author are relinquishing any and all control, and you are not allowing only those uses of which you approve or see as a community benefit, you are allowing any use that a third party sees fit to excercise —- or that is in its own self-interest. In that sense the faculty is risking something by turning those rights over.

    I will happily stipulate that everyone involved in this issue at Harvard is acting with the best of intentions. However, a good lawyer will advise you not to word a contract assuming the best of intentions, but to fairly secure the rights and interests of both parties. Too bad the A&S faculty didn’t talk to the Law School faculty before voting….

    — Peter    Feb 14, 05:11 PM    #

31. As to 30: It seems still preferable to give Harvard some non-exclusive rights over giving all rights exclusively to a publisher and keeping none.

    Your analogy seems weak. If you want your stuff read, you must transfer some rights. It’s in your interest, but does not apply to your example.

    But even in your example, you would prefer keeping some non-exclusive rights over your property rather than losing it altogether.

    — Ek    Feb 15, 02:37 AM    #

32. Supporting open access journals is laudable, but the way in which the A&S faculty members of Harvard have (apparently) tried to do this problematic. Posts 28 and 30 raise important concerns, but there is one even more troublesome. The policy is a diminishment of academic freedom in that it seems to require faculty to only publish in journals that will allow this other (free) distribution. Most journals will not do this. Thankfully, the policy is probably not legally enforceable, at least for faculty already hired who have not signed an agreement to follow it. It has traditionally been held that materials created by a professor (whether they be course notes, figures, research papers, books) are NOT works for hire, and are the intellectual properties of their creator. This prohibits, for example, a college from taking all of an instructors Powerpoint shows, posting them on a website, and selling it as an on-line course (without consent and compensation to the instructor). The faculty at Harvard who did not support this policy may wish to visit the Statement of Principles on Academic Freedom and Tenure to be found at http://www.aaup.org/AAUP/pubsres/policydocs/contents/1940statement.htm

    There are already plenty of disincentives for qualified men and women to go in to college teaching (low pay, long hours, increasing administrative burdens). Do readers know that many research articles are written during evenings, weekends, and summers, when the faculty member is not being compensated? There is simply not enough time during the 45-hour workweek to get everything done. Perhaps this is not the case at elite schools, where teaching loads are low, support staff extensive, and compensation reasonable. But, at the many other institutions faculty struggle to excel in teaching, research, and service to the community. Limiting these scholar’s choices about how and where to distribute the fruits of their intellectual labors is a slap in the face (no matter how lofty the goal).

    — Ira    Feb 15, 07:56 AM    #

33. I have been trying to ‘educate’ scientists concering so-called Open Access (OA) since it was first proposed by Harold Varmus (then head of NIH). Once again, the depth of the ignorance on this issue astounds me.

    First, essentially all electronic journal articles are already ‘archived’ – for example, the more than 1.8 million articles from over 130 scholarly journals (many of them the top journals in their fields) that are freely downloadable on at Stanford’s HighWire press. Second, electronic archiving is not free but takes a large amount of maintenance, including continual upgrading of hardware and software, and of course personnel. Third, this move potentially short-circuits the peer-review process, which, while not perfect, is a time-tested system that adds tremendous value to scholarly publications (in fact, without peer review one can make a strong argument that the publication is not scholarly at all). Fourth, the OA movement threatens the viability of not-for-profit journals of many scholarly societies, which are the backbone of the scientific publishing system. Fifth, publishers, and especially not-for-profits, have substantial costs associated with publishing (again, thinking that elecronic publication is somehow free is naive in the extreme) and what OA really is doing is shifting the entire cost of publication to the authors (so-called OA or author fees, which currently are $2,500 to $4,000 per article) because OA makes the subscription model nonviable.

    Hope someone clues in the faculty at Harvard and other institutions before they go headlong down this very dangerous path.

    — Lawrence P. Reynolds    Feb 15, 10:38 AM    #

34. As a follow-up to my previous posting, I was informed by email that a recent issue of The Journal of Electronic Publishing
    (Michigan) available at http://journalofelectronicpublishing.org contains a number of articles that are related to the Open Access debate. In particular, see that by Donald Waters, “Open Access Publishing and
    the Emerging Infrastructure for 21st-Century Scholarship.” An excerpt:

    “It is all too easy to focus on the trendy, glitzy, heart-pounding
    rhetoric about the initial step of making materials freely available,
    especially those materials that “your tax dollars helped make possible,”
    and to trust that only good consequences will follow downstream. It is
    much harder to focus strategically on the full life cycle of scholarly
    communications and ask hard questions such as: open access for what and
    for whom and how can we ensure that there is sufficient capital for
    continued innovation in scholarly publishing? One worry about mandates
    for open access publishing is that they will deprive smaller publishers
    of much needed subscription income, pushing them into further decline,
    and making it difficult for them to invest in ways to help scholars
    select, edit, market, evaluate, and sustain the new products of
    scholarship represented in digital resources and databases. The bigger
    worry, which is hardly recognized and much less discussed in open access
    circles, is that sophisticated publishers are increasingly seeing that
    the availability of material in open access form gives them important
    new business opportunities that may ultimately provide a competitive
    advantage by which they can restrict access, limit competition, and
    raise prices.”

    — Lawrence P. Reynolds    Feb 15, 05:38 PM    #

35. Those who defend proprietary publishing seem to overlook the following:

    1. If journal pricing were reasonable (cost plus), there would be no need to introduce open access. The main function of open access, from my point of view, is that it puts a lid on journal pricing and thereby improves the traditional system and make the refereed journals more accessible (and hopefully reduce the number of journals on the low-end sid, thereby increasing usefulness of the journal system).

    2. Those arguing against the Harvard initiative systematically neglect that researchers can opt out of the system. Post 32 illustrates that.

    3. Further, and as to post 34, it is quite right that there is severe market failure in the journal market. This is typically overlooked by those defending the system (and arguing against the Harvard initiative). But the argument that encouragement of supplementary open access (availability of pre-refereed and post-refereed versions of published articles) would raise journal prices is the opposite of what I, as a university professor of economics, would conclude from general market theory and from reading articles by economists on journal pricing. One of the major benefits of the Harvard initiative is to bring journal pricing closer to competitive pricing.

    — Ek    Feb 20, 04:04 AM    #

36. Misunderstandings of Open Access And Harvard OA Policy

    There are too many errors and misunderstandings in the 36 preceding comments to answer them one by one. For a better idea of what Open Access is and what has been happening across the past decade, see the American Scientist Open Access Forum (continuous since 1998).

    For the current Harvard Mandate draft’s strengths and weaknesses (and how to remedy the weaknesses), see this Self-Archiving Policy thread.

    Stevan Harnad

    — Stevan Harnad    Feb 25, 06:35 AM    #

Previous: This Just In: Time Travelers Could Arrive by Summer
Next: Senate Foreign Relations Committee Passes Study-Abroad Bill