How to make your own work open access

From Harvard Open Access Project
Jump to: navigation, search
  • I first wrote these notes as an online handout for a talk at the Berkman Klein Center on October 23, 2012, and have added updates and revisions periodically since then. —Peter Suber.
    • These notes focus on open access to peer-reviewed research articles and their unrefereed preprints. They do not cover books, theses and dissertations, conference presentations, datasets, courseware, audio, video, multimedia, or source code. But I might add pages on other categories over time. The live audience for the talk consisted of Harvard people, which explains the occasional Harvard reference. But the sources cited will be useful for scholars anywhere. The full title of the talk was, How to Make Your Research Open Access (Whether You're at Harvard or Not).
    • Now that I've got the notes online, I welcome suggestions.
    • This handout should be more useful than my slides from the talk. I keep it up to date, link to the sites I mention, and use complete sentences. But FYI, here are the slides.
    • Also see Chapter 10 ("Self-help") of my book, Open Access (MIT Press, 2012), and the updates to Chapter 10. The book is itself OA.

Publish in an OA journal ("gold" OA)

  • Find a suitable OA journal. Go to the Directory of Open Access Journals (DOAJ) and browse by field.
    • As you browse, bear in mind that some OA journals will be high in quality, impact, and prestige. Some will be low. In this respect, OA journals are just like conventional, non-OA journals.
    • Some will use liberal open licenses, like CC-BY. Some will use more restrictive open licenses like CC-BY-NC or CC-BY-NC-ND. Some will offer only gratis access without open licenses, that is, they will be free of charge for reading but publish under all-rights-reserved copyrights.
    • Some will charge publication fees (also called article processing charges or APCs), and some will not.
  • If the best journal for your purposes charges a publication fee, see whether your funder or employer will pay it.
    • There's no complete list of funders willing to pay publication fees on behalf of grantees. But see the incomplete list from BioMed Central, or the larger incomplete list from the Nature Publishing Group. If the article you want to publish is based on funded research, check with your funding agency directly.
    • Many universities are willing to pay publication fees on behalf of faculty.
    • If you're at Harvard, apply to the Harvard Open-Access Publishing Equity (HOPE) fund. For a Harvard-based project to spread HOPE-like funds to other institutions, see the Compact for Open-Access Publishing Equity (COPE).
    • If you're worried about paying fees out of pocket, here's some background to help you estimate the odds. Although charging publication fees is the best-known business model for peer-reviewed OA journals, it's not the most common. Only about 30% of peer-reviewed OA journals overall charge publication fees. When OA journals do charge fees, the fees are usually paid by funders (59%) or by universities (24%). Only 12% of the time are they paid by authors out of pocket. See Table 4 of the comprehensive Study of Open Access Publishing (SOAP). That means that only 3.6% of authors who publish in OA journals overall (12% of 31%) pay fees out of pocket. At the same time, about 50% of articles published in peer-reviewed OA journals are published in fee-based journals. Hence, if we count by article rather than by journal, then only 6% of authors who publish in OA journals overall (12% of 50%) pay fees out of pocket. These figures are averages, and the rates depend significantly on field and type of employer; see Figure 8 of the SOAP study.
    • If you want to know whether a given journal charges a publication fee, and if it's listed in the Directory of Open Access Journals, then the DOAJ will tell you. If it's not listed in the DOAJ, then reconsider whether it's the right journal for you. For more, see the next entry on assessing the quality of an unknown journal.
  • If you find an otherwise promising OA journal in your field, but have never heard of it, investigate it. Network with trusted colleagues to do so.
    • Find the names of the editors and members of the editorial board. Do you recognize and respect them?
    • Above all, read a good sample of its papers. Are they good, by your own standards?
    • Is the journal listed in the Directory of Open Access Journals (DOAJ)? If so, then either trust the journal or deepen your investigation. Read more here, here, or here about the DOAJ efforts to weed out untrustworthy journals.
    • Also follow the progress of Cabell's, JournalGuide, JournalReviewer, Journalysis, Peer Review Evaluation (PRE), Quality Open Access Market (QOAM), SciRev, and Think-Check-Submit, which are trying to solve the same problem.
    • Don't assume that unknown journals are weak. Low profile does not entail low quality, especially when journals are new, and the average OA journal is significantly newer than the average conventional journal. New journals, OA and non-OA, face the vicious circle of needing excellent submissions to generate reputation, and needing reputation to attract excellent submissions. Don't fault a journal for being only partway through the process of escaping this circle. Look to its quality alone, as far as you can determine it. Even journals excellent from birth need time to develop a reputation in proportion to their quality. When an honest, new, and little-known journal is struggling for visibility, you can help it escape the circle by submitting some of your best work. In the words of Harvard's Faculty Advisory Council on the Library (April 2012), "move prestige to open access."
    • Is the publisher a member of the Open Access Scholarly Publishers Association (OASPA)? OASPA has a good code of conduct requiring peer review, requiring disclosure of a journal's vetting process, fees, and ownership, and prohibiting spam to solicit papers or members of editorial boards. Also see the DOAJ Principles of Transparency. Some honest, high-quality OA publishers do not yet belong to OASPA, and some honest, high-quality OA journals are not yet listed in the DOAJ. But we should encourage them to apply. If your investigation of a particular journal doesn't turn up evidence that you trust one way or another, then follow the rule to avoid publishers who aren't members of OASPA and journals that aren't listed in the DOAJ. Don't hesitate to tell them that you are doing so. That will give them an incentive to join, and live up to their high standards.
    • If an unknown OA journal solicits your work, or invites you to join the editorial board, then you could reply that you'll consider it when the publisher joins OASPA and the journal is listed in the DOAJ.
  • When you find a suitable OA journal, then submit your manuscript, just as you would to a conventional journal.
  • If you don't find a suitable OA journal, check again when you publish your next paper. Things are changing fast.
    • But don't conclude that you can't yet make your article OA. If you don't make it gold OA (through an OA journal), you can publish it in a non-OA journal make the peer-reviewed manuscript green OA (through an OA repository). For details, see the section on green OA, below.

Deposit in an OA repository ("green" OA)

  • If there isn't an OA repository in your institution or field, then consider a universal repository such as Zenodo, OpenDepot, or GitHub.
    • You could also post your work to your personal web site or home page. But repositories are better long-term solutions because they take steps toward digital preservation, provide persistent URLs, and will continue to hold your work and make it OA after you change jobs or die.
    • Non-profit OA repositories are also a better choice than for-profit social-networking sites like and ResearchGate. See the well-written arguments by the U of California and Kathleen Fitzpatrick.
  • To save time, start with your new works and deposit them as they are finished. As you find time, deposit your older works retroactively.
    • When you deposit new works, deposit at the time of acceptance. That's when you have the peer-reviewed manuscript in hand (before you lose track of it) and that's when you're still thinking about widening access to your new work (before you become preoccupied by your next project).
  • Most of the time, you may lawfully make your work OA through a repository even if you published in a non-OA journal. There are two reasons why.
    • First, most non-OA publishers give standing permission for green OA.
    • Second, research institutions can adopt policies to assure permission for green OA even in cases where publishers don't already give standing permission.
    • Unfortunately the two facts above are among the best-kept secrets of OA, and must compete against some of the most widespread misunderstandings about OA, for example, that all OA is gold OA, that publishing in a non-OA journal forecloses the possibility of OA, that the relevant rights always belong to publishers, and that non-OA publishers are doing nothing to adapt to an OA world or accommodate authors who want OA.


  • No matter which path you choose ("gold" or "green" OA), the journal or repository will need permission to make your work OA. But permission from whom? The answer depends on what happened to the rights after you wrote the article.
  • When you write a new article, you are the copyright holder. You needn't apply for a copyright or register the work. It's automatic. If you haven't already transferred rights to others, then permission for OA comes from you.
    • You may authorize publication in an OA journal (gold OA). Just sign the publishing contract.
    • You may authorize OA through a repository (green OA) for an unpublished manuscript or preprint. Just make the deposit. If there's a check box to affirm that you have the right to authorize OA, check it. However, if you want to deposit a published article, then you have probably transferred all or some rights to a publisher. Hence, see next.
  • If you want to deposit a published article in a repository, then the repository will need permission from the relevant rightsholder.
    • If you retained key rights when you published, which is rare, then you may authorize OA through a repository on your own. You needn't consult or involve the publisher.
    • If you transferred key rights to the publisher, which is common, then you will need the publisher's permission.
    • However, most conventional or non-OA publishers give standing permission for author-initiated green OA.
      • To see whether your journal or publisher gives this kind of standing permission, read your publishing agreement. Or look up the journal or publisher in the SHERPA RoMEO database.
      • In most cases, this kind of standing permission for green OA applies to the version of the text approved by peer review (sometimes called the accepted author manuscript or AAM), not to the published edition. To take advantage of this permission, you'll need to put your hands on that version of the text. Tip for the future: whenever you publish an article, always retain the accepted author manuscript, and deposit it in an OA repository at the time of acceptance.
      • SHERPA also maintains a shorter list of publishers who give standing permission for authors to deposit the published editions of articles in an OA repository.
    • If your journal or publisher does not give standing permission for green OA, then try one of these strategies.
      • Ask for permission. Many publishers who don't give standing permission will agree to case-by-case requests.
      • Use an author addendum. An author addendum is a proposed modification to the publishing agreement, written by a lawyer, giving the author the right to authorize OA (and sometimes other rights as well). Because it's only a proposed modification, publishers may accept it or reject it.
  • For standing permission to make your future articles OA, without depending on publishers for that permission, work toward a Harvard-style OA policy at your institution.
    • The Harvard Open Access Project (HOAP) can help you with a policy at your institution. Also see the HOAP guide to good practices for university OA policies. (Disclosure: I write and edit the guide with Stuart Shieber.)
    • Through Harvard-style OA policies, faculty grant the institution non-exclusive rights to their future scholarly articles, including the right to authorize OA through the institutional repository. This assures that faculty may lawfully make their work OA...
      • even when they publish in a non-OA journal,
      • even when the non-OA journal does not give standing permission for green OA, and
      • even when faculty members have not negotiated special access terms or permissions with their publishers.

Translations of this handout

This handout is also available in Greek (September 2013) and Spanish (January 2014). I welcome other translations.