UCSF enacts policy mandating open-access research, and a related complaint from me

The chickens are finally coming home to roost.  The price-gouging practices of many academic journals, a practice I’ve often decried here, have driven one large and important university to take  action. According to the webpage of the University of California at San Franciso (UCSF), that university will henceforth constrain all of its faculty and academic employees to publish in journals whose contents are freely and immediately available to anyone.  It has always galled me that although the taxpayers (whose hard-earned dollars are distributed to scientists largely through the National Science Foundation and the National Institutes of Health) fund our work, scientific journals can nevertheless make taxpayers cough up large sums to view the results of that research. That’s unconscionable by anyone’s lights.

“Our primary motivation is to make our research available to anyone who is interested in it, whether they are members of the general public or scientists without costly subscriptions to journals,” said Richard A. Schneider, PhD, chair of the UCSF Academic Senate Committee on Library and Scholarly Communication, who spearheaded the initiative at UCSF. “The decision is a huge step forward in eliminating barriers to scientific research,” he said. “By opening the currently closed system, this policy will fuel innovation and discovery, and give the taxpaying public free access to oversee their investments in research.”

UCSF is the nation’s largest public recipient of funding from the National Institutes of Health (NIH), receiving 1,056 grants last year, valued at $532.8 million. Research from those and other grants leads to more than 4,500 scientific papers each year in highly regarded, peer-reviewed scientific journals, but the majority of those papers are only available to subscribers who pay ever-increasing fees to the journals. The 10-campus University of California (UC) system spends close to $40 million each year to buy access to journals.

. . . The new policy requires UCSF faculty to make each of their articles freely available immediately through an open-access repository, and thus accessible to the public through search engines such as Google Scholar. Articles will be deposited in a UC repository, other national open-access repositories such as the NIH-sponsored PubMed Central, or published as open-access publications. They will then be available to be read, downloaded, mined, or distributed without barriers.

There does, however, seem to be one loophole:

The UCSF policy gives the university a nonexclusive license to distribute any peer-reviewed articles that will also be published in scientific or medical journals. Researchers are able to “opt out” if they want to publish in a certain journal but find that the publisher is unwilling to comply with the UCSF policy.  “The hope,” said Schneider, “is that faculty will think twice about where they publish, and choose to publish in journals that support the goals of the policy.”

So much for forcing Science or Nature to stop charging exorbitant sums to view their papers!

I have another complaint not addressed by this policy: some journals allow scientists to withhold the raw data analyzed in their papers for long periods (over a year) so that other researchers can’t “mine” it and use that data to generate their own papers. The idea is that one’s data is proprietary, and one should have exclusive use of it for a long period of time. This, too, I find unconscionable.  The essence of science is that other workers must be able to repeat your results before those results become credible in the research community.  You can’t do that if someone publishes an analysis and then refuses to give you the data on which that analysis is based.  The selfish desire to “own” datasets, so you can generate lots of papers from them without other scientists being able to peek at the data, is inimical to the progress of science.  In my view, every bit of data on which a published paper is based must be available immediately to any scientist who wants to see it.

h/t: Ollie

41 Comments

  1. Posted May 25, 2012 at 11:35 am | Permalink

    Woo-hoo! The tides, they are a-turnin’!

    (My only beef would be your use of the word, “scientist,” in the last sentence. Replace it with, “person,” and I’m lock-step in agreement with you.)

    Cheers,

    b&

  2. Zugswang
    Posted May 25, 2012 at 11:37 am | Permalink

    “It has always galled me that taxpayers (whose hard-earned dollars are distributed to scientists largely through the National Science Foundation and the National Institutes of Health) fund our work…”

    Just to clarify, even if open-access was a universal principle, would you still dislike the idea of taxpayer-funded research?

    As for the loophole, I imagine that’s a survival strategy more than anything. Publications like Nature and Science won’t go away or change their business model overnight, and it’s going to be difficult to attract new PI’s (or even retain some current researchers) knowing they would be absolutely prohibited from publishing in high impact journals because they don’t comply to a new policy. I’m sure that this will be revised as time goes by, and more universities join the increasingly loud call for journal reform.

    • whyevolutionistrue
      Posted May 25, 2012 at 11:51 am | Permalink

      I didn’t put that very clearly. What I meant (and I’ve fixed it) is that since our work is funded by taxpayers (and I have no problem with that, since scientific research is a boon to societies that can support it), those who pay for it should have free access to the results? I’ve rewritten that sentence to make it clearer; thanks.

  3. bernardhurley
    Posted May 25, 2012 at 11:49 am | Permalink

    It has always amazed me that governments are prepared to spend millions funding research and yet do next to nothing to make sure the results are freely available.

    • Prof.Pedant
      Posted May 25, 2012 at 6:19 pm | Permalink

      My assumption is that at one time monopoly-capitalist solutions to the problem of how to distribute scientific information worked quite well – certainly as well as any other system envisioned – and fit quite well onto the legacy model of publication by scientific societies, etc. But a system that appears to have worked well at one time no longer does. Unfortunately there the realization that things are messed up takes a while to propagate through the system because there is such a strong faith in capitalist approaches to solving resource/information distribution problems. Among other things there seems to be an assumption that businesses that have existed somehow have a ‘right’ to continue to exist, and in their present form. Thus the solutions to the problem that may be easily considered are initially variations of ‘more of the same’.

  4. Posted May 25, 2012 at 11:54 am | Permalink

    I’m glad there are some loopholes. I have a paper under revision that has lots of equations in it. There are only a few journals in population biology that will actually publish the equations in the body of the paper, and not in the Supplementary Material. If I want all readers to see the equations, I strongly prefer them to be in the paper. And I am not aware of any open access journal that will publish the equations that way (no, not even the ones that have “Computational” in their journal title). So given that constraint, I need to use the loophole.

    • vHF
      Posted May 25, 2012 at 12:05 pm | Permalink

      I currently have a paper under revision in PLoS Computational Biology, with a considerable number of equations, and after the first round of reviews nobody seems to mind them. So it is possible.

      In any case, there are degrees of non-OA. Submitting to Am Nat, which belongs to UoC Press is way nicer to the community than submitting to Elsever’s JTB o TPB.

    • ahannaasmi
      Posted May 25, 2012 at 12:09 pm | Permalink

      The Journal of Computataiona Biology? However I believe their open-access option requires authors to pay a (somewhat hefty) fee.

      • ahannaasmi
        Posted May 25, 2012 at 12:10 pm | Permalink

        I meant the Journal of Computational Biology. Sorry for the typo.

  5. ThyroidPlanet
    Posted May 25, 2012 at 12:02 pm | Permalink

    Elsevier publications are free after 12 months. Nature is never free. Working scientists need articles ASAP. Nobody has been keeping scientists from publishing in plos. How do we know the “taxpayers” need to close the gap?

    These facts are always conspicuous in their absence.

    I’ll tell all my neighbors nonetheless about the windfall for us “taxpayers”.

    • vHF
      Posted May 25, 2012 at 12:07 pm | Permalink

      > Elsevier publications are free after 12 months.

      Surely you jest?

      • ThyroidPlanet
        Posted May 25, 2012 at 12:53 pm | Permalink

        Go look one up that isn’t. Cell press is. AFAIK Elsevier is.

        • vHF
          Posted May 25, 2012 at 1:03 pm | Permalink

          But I did. Went to the Journal of Theoretical Biology, selected random issue from 10 years ago and a random article. $39.95.

          Cell Press is an imprint of Elsevier. I repeated the procedure with Trends in Ecology and Evolution, a Cell Press journal. $37.95 this time.

          Are you sure you’re not proxied or VPNed to your research/teaching institution?

          • ThyroidPlanet
            Posted May 25, 2012 at 2:22 pm | Permalink

            I checked cell press a while ago. Cell. My mind filled things in. Can’t do this on mobile babysitting. Sorry.

            Other args stand I think but I should not have posted.

            • vHF
              Posted May 25, 2012 at 2:39 pm | Permalink

              Cell indeed appears to be free (only to read, mind, not to distribute or text-mine) after one year. Having funded, performed, written up, reviewed and edited this research we should be grateful, I suppose.

          • ThyroidPlanet
            Posted May 25, 2012 at 2:23 pm | Permalink

            Also cell is high eigenfactor.

            • MadScientist
              Posted May 25, 2012 at 2:49 pm | Permalink

              Huh? Do you even know what ‘eigen’ means?

              As for public access, Joe and Jane Ordinary should have free access to the results of publicly funded civilian research. The Ordinaries may be bright folk who would love to read such stuff and learn more. As the UCSF folks also pointed out, there are also many scientists out there whose institutions simply can’t afford the subscription fees. Even national libraries are dropping subscriptions (some, like the British Public Library, have provisions in law which guarantee them a free copy but many national libraries simply go without).

              • vHF
                Posted May 25, 2012 at 2:59 pm | Permalink

                Eigenfactor is an established measure of scientific impact, in many ways preferable to the impact factor. So no need to get snippy.

                Good points about the rest. They really are obvious, but need to be made over and over again.

              • ThyroidPlanet
                Posted May 25, 2012 at 5:52 pm | Permalink

                “Joe and Jane Ordinary should have free access to the results of publicly funded civilian research.”

                I would argue using this approach is populist political game play that distracts and eventually undermines the objective – which I actually am for : free-as-in-freedom research results.

                so please, substitute the cynical term “taxpayer” with “citizen”.

          • ThyroidPlanet
            Posted May 25, 2012 at 5:57 pm | Permalink

            OKayyy…

            I see now Trends journals are not free, and that JTB is not free. thank you.

            … but still Cell is free after 12 months.

    • ThyroidPlanet
      Posted May 25, 2012 at 5:59 pm | Permalink

      correction here courtesy of vHF below:

      I said “Elsevier publications are free after 12 months”.

      in fact, the only journals that are in that category are non-Trends journals from Cell Press, e.g. Cell which is of high eigenfactor.

      vHF below corrected me on this, and I appreciate it.

    • Kevin
      Posted May 25, 2012 at 8:42 pm | Permalink

      No. They’re not.

      I deal with Elsevier publications all the time. Papers published in Elsevier journals are NEVER made available for free.

      Please check your facts before posting further.

      • ThyroidPlanet
        Posted May 26, 2012 at 9:02 am | Permalink

        Kevin wrote:
        ” Papers published in Elsevier journals are NEVER made available for free.”

        this is false. Cell Press is part of Elsevier. vide infra:

        Cell press states:

        http://www.cell.com/cellpress/access

        “We provide free access to Cell Press research journals 12 months following publication”

        ThyroidPlanet wrote:

        “the only journals that are in that category [free after 12 months] are non-Trends journals from Cell Press, e.g. Cell”

        that was in response to vHF on his correction. I also admitted to my mind filling in details from a long time ago, and that using mobile while babysitting was a bad idea, and I should not have posted.

        so I think we are all on the same page now. Journal content is free after 12 months for Cell press journals which is part of Elsevier. Not all Elsevier journals are free at any given time. Cell Press is contained within Elsevier.

  6. TheBrummell
    Posted May 25, 2012 at 12:25 pm | Permalink

    I am *right now* procrastinating finishing a manuscript to be sent to Global Change Biology, published by Wiley (whose business model, as far as I can tell, shares much with Elsevier’s). This is an issue that is certainly in my mind.

    I’m glad to see a university make such a bold move, and I’m hoping it will stick and (even better) spread. I don’t know much about the University of California system, do the various universities frequently share policies, or participate in decisions like this? Am I correct in thinking the U of CA system, taken as a whole, represents a very large and significant group of researchers as well as students and instructors?

    Regarding the data thing, it seems to me that researchers are encouraged to publish their data in addition to publishing their results. I’m still trying to wrap my head around this concept (how does one peer-review a database with 10 000 entries in it?), but one effect is one’s data becomes securely attributed to oneself. Somebody else can’t really “steal” your data if it’s been published any more than would citing influential work in a regular paper. Despite my confusion over the process, I’d advocate data publication as greatly preferable to this odd many-month embargo.

    • MadScientist
      Posted May 25, 2012 at 2:34 pm | Permalink

      The different campuses of the UC system are independent; each has their own Board of Regents etc. The only thing in common is that they are established and funded by the state.

      • ahannaasmi
        Posted May 25, 2012 at 6:45 pm | Permalink

        The different campuses of the UC system are independent; each has their own Board of Regents etc.

        No, they have very much that same Board of Regents.

        • Achrachno
          Posted May 25, 2012 at 9:45 pm | Permalink

          Yes, ahannaasmi is correct — all UC campuses share one board of regents. (And if anyone needs some, we’d be happy to give you a few.)

          • Tim
            Posted May 26, 2012 at 1:23 pm | Permalink

            If you’d like to do a straight across trade for the Rick Perry appointed Texas A&M BOR, I think I can get you tens of thousands of A&M faculty, student, and alumni signatures on a petition to make it so.

        • MadScientist
          Posted May 26, 2012 at 4:41 pm | Permalink

          Dang – serves me right for relying on my memory instead of looking it up.

    • MadScientist
      Posted May 25, 2012 at 2:40 pm | Permalink

      Regarding making data public – you need to put it in a format that people can access (preferably with free tools). If it were spectra for example, then you might use the horrible ASCII encoded JCAMP format rather than any of the dozens of proprietary formats which commercial instruments puke out. You also need to publish information on how to get at the data. For example if you use a PostGRESQL database you need to state that, include some brief instructions on how to import the data (since you will likely export it as an SQL script) and state the database schema etc. In other words, your data has to be well documented. Throwing out a few *.csv files with cryptic headers isn’t good enough.

  7. Not a scientist
    Posted May 25, 2012 at 12:53 pm | Permalink

    Forgive me for my lack of knowledge on the business of publishing science journals, but aren’t there still some costs associated with publishing the journal? i.e. editors, copy-writers, copy-setters, etc.? I realize peer-reviewers are typically not paid, but isn’t it acceptable at some point to charge some for the costs to the publishers? $40 million seems excessive, but there must be some other appropriate amount.

    • Kevin
      Posted May 25, 2012 at 8:43 pm | Permalink

      That’s what ads are for.

  8. morkindie
    Posted May 25, 2012 at 1:17 pm | Permalink

    Does this mean that Google Scholar will start putting out and stop being such a tease?

  9. Torbjörn Larsson, OM
    Posted May 25, 2012 at 1:23 pm | Permalink

    The essence of science is that other workers must be able to repeat your results before those results become credible in the research community. You can’t do that if someone publishes an analysis and then refuses to give you the data on which that analysis is based. The selfish desire to “own” datasets, so you can generate lots of papers from them without other scientists being able to peek at the data, is inimical to the progress of science. In my view, every bit of data on which a published paper is based must be available immediately to any scientist who wants to see it.

    The practical problem comes from large observatories like LHC or Kepler, where scientists have spent decades to mass the data sets and doesn’t have the resources to mine them except for the main thrust.

    However, I believe their insight most likely secure results in some form, almost guaranteeing their reward. So I would welcome a push in the direction of openness, and I’m hopeful the problems with balancing rewards can be solved.

    • Torbjörn Larsson, OM
      Posted May 25, 2012 at 1:24 pm | Permalink

      I meant amass the data sets, though *mass” would work too I guess.

  10. MadScientist
    Posted May 25, 2012 at 2:29 pm | Permalink

    I don’t see a need to hold onto data for so long; as far as I’m concerned, by the time an article is published the data is all old – if someone else finds something useful in that data then good for them. If you’re in the astronomical community and you’re using a public facility (as the vast majority of astronomers do) you have 1 year from the observation to stare at the data and send your articles out for publication, then the data becomes publicly available. The 1 year period gives you enough time to gather data from other observations (if necessary) and to write stuff up. I think far too many scientists these days have some crazy ideas about how valuable their data are. Don’t make data public too early – there are unscrupulous people out there who will steal your work if it’s good – but once your articles are approved for publication, don’t keep the data hidden. In fact in many cases my reaction is “I can’t take you seriously if your data is not public – what are you hiding?”

  11. Schenck
    Posted May 25, 2012 at 3:07 pm | Permalink

    This is great, it’s stuff like this that can make a difference, and didn’t Harvard do the same thing recently?

    Only problem is there’s an opt-out, so it’s really a non-policy, if you want to publish in a closed journal, you can.

  12. ThyroidPlanet
    Posted May 25, 2012 at 6:07 pm | Permalink

    [1] the term “taxpayer” is cynical. To gain support for the cause, please substitute “citizen”.

    [2] I think a better way to get the free research results thing to work is via the angle of corporations, industry, etc. will be the real benefactors – akin to what happens when DARPA releases research – and help drive the economy. if anyone wants the government to get behind the free publication idea, driving the economy is a far better tactic than the populist route.

  13. Talib Alttaawiil
    Posted May 25, 2012 at 9:31 pm | Permalink

    does it seem to anyone else that the spirit of capitalism is antithetical to science, scholarship & the free exchange of ideas?

  14. Posted May 25, 2012 at 10:45 pm | Permalink

    It seems trivial to solve the publishing problem in the age of the Internet, so I wonder what really stands in the way? A perceived loss of prestige?

  15. Posted May 25, 2012 at 11:36 pm | Permalink

    Another fine example of ‘sleight of word’. (What we really mean is). Are the science and nature bods using religious tactics???


Follow

Get every new post delivered to your Inbox.

Join 29,483 other followers

%d bloggers like this: