During the first week of December, Nature took a measured step forward in expanding access to high-value scientific literature when the publication announced that research papers from that journal would be made accessible to the public for reading via the Readcube platform if existing subscribers chose to share a link with friend or colleague. The announcement met with equally measured enthusiasm as open access advocates were dissatisfied with system constraints intended to prevent unauthorized diffusion of content outside of that platform.

The Nature announcement contained a quote from Managing Director Timo Hannay characterizing the program as “a convenient, legitimate alternative that allows researchers to access the information they need and the wider, interested public access to scientific knowledge, from the definitive, original source”. (Part of that wider access includes links provided to selected media outlets for use in their coverage of scientific, technical and medical research published in Nature.) Hannay’s blog entry for December 5 provides some additional background on the thinking behind this pilot initiative. Given the popularity of of social networks as a mechanism for content discovery by scientists, faculty and students, one can see the logic in Nature’s approach. By enabling such capability for reading and sharing, they ensure that their content remains visible and available to an interested audience that might otherwise be drawn away to such platforms as Academia.edu, ResearchGate or Mendeley.

Image via Blue Diamond Gallery.

Advocates of open access immediately criticized Nature and its parent company, Macmillan, for this form of “fauxpen access” due to speed bumps hindering further “use” of the material. Readers are not permitted to download or print the full text of an article, or even a simple cut-and-paste for purposes of quotation. Peter Murray-Rust, UK chemist and high-profile open access advocate, noted as well that Nature had failed to build a mobile-friendly version of the site which might preclude practical access to the content by those in the Global South who tend to rely more heavily on mobile devices for accessing Web-based services.

As I hope I’ve made clear, the Nature announcement can be viewed in a variety of lights. On the one hand, it expands access to some portion of the most important and properly-vetted scientific findings. At the same time, the adopted approach leverages protective means that may hinder legitimate downstream use. Reasonable people may disagree as to which aspect outweighs the other.

But there was one aspect I found objectionable in the commentary by OA advocates and that was what appeared to be casual exploitation of the needs of the visually impaired. Multiple blog entries and tweets noted the failure in Nature’s system to support access for that segment of the public. However, these were terse notations, lacking explanations as to whether this was due to incompatibility issues on the particular platform or caused by some other flaw in the system. None of the nay-sayers went into much detail, and yet such incompatibility hadn’t surfaced in earlier open access manifestos — Budapest, Berlin, Bethesda — as being a key concern. Not knowing much about screen reader applications or similar accessibility supports for the visually impaired, I wondered just how big of an issue is this? And for whom?

Even as a casual investigator, I could find specifics on the visually impaired, such as the World Health Organization’s data that says globally, there are 285 million people suffering from visual impairment. Looking a little closer at just one segment of that population, I was able to retrieve an article in Nature quoting NSF statistics (collected in the year, 2000) that put the total percentage of the disabled in the STEM workforce as being about 7% or just under 400,000 individuals. Note that this is not a number encompassing just the visually impaired, but a total of those working in STEM research fields with any form of physical disability. Another result of that casual investigative research linked to a 2007 article from Science which noted that “…the employment rate for scientists and engineers with disabilities is 83 percent, much better than the estimated 26 percent for the overall US population with disabilities.”  Which is why content and technology providers reading the Scholarly Kitchen blog should be re-evaluating their obligation to the needs of this guy!  Standard business rationalizations for slow response (limited corporate resources, insufficient ROI, etc.) are, in my view, unsatisfactory when you factor in the higher concentration of those affected in the workforce.

That said, it is equally poor form on the part of OA advocates to draw upon the visually impaired for political expediency.  Should any advocacy group equate the barriers to access faced by the visually impaired to the barrier posed by a semi-permeable pay-wall? Is the disabling of article downloads in any way comparable to the lack of a screen reader?

If advocating for the disabled is important to the open access community, there is data being collected about the disabled in the STEM workforce that could be better used in moving content providers to action.  It can and has been done. As an example of proven results, some might recall the advocacy movement surrounding the “book famine”, driving the current ratification process of the WIPO June 2013 Marrakesh Treaty. Advocacy in that realm made a difference and there was response from the information community.

My question to open access supporters is “Have you considered how trivializing this particular need for access undercuts your advocacy?” Enlist the aid of one competent librarian at your institution in assembling meaningful data about a truly under-served population, and then leverage that data to win support and drive needed change. Publish a well-researched blog entry on gaps in accommodation for those facing access barriers in your particular lab (ideally without resorting to maudlin sentiment) and thereby deepen the value of your public criticism. Prove that you are intentional advocates for a segment of your own workforce and scholarly community.

To do less is exploitative and unworthy of the open access movement.

Jill O'Neill

Jill O'Neill

Jill O'Neill is the Educational Programs Manager for NISO, the National Information Standards Organization. Over the past twenty-five years, she has held positions with commercial publishing firms Elsevier, ThomsonReuters and John Wiley & Sons followed by more than a decade of serving as Director of Planning & Communication for the National Federation of Advanced Information Services (NFAIS). Outside of working hours, she manages one spouse and two book discussions groups for her local library.


14 Thoughts on "Doing Better With Open Access Advocacy"

“Not knowing much about screen reader applications or similar accessibility supports for the visually impaired, I wondered just how big of an issue is this? And for whom?”

For visually impaired people, most likely.

As for the classic OA declaration not explicitly mentioning visual impairment: you are right that this specifically doesn’t come up in (for example) the BOAI definition. But don’t you think it’s implicit in “free availability on the public internet, permitting any users to read, download, copy, distribute, print, search, or link to the full texts of these articles […] without financial, legal, or technical barriers”?

To be clear, NPG have been really good about stating that what they’re offering here is not open access — they recognise the distinction, and credit to them for not trying to blur boundaries. So the failure of what they’ve provided to meet the requirements of the BOAI is not news, and shouldn’t be a stick to beat them with.

But I do feel that providing the best access for people with visual impairment is very much of a piece with what the earliest OA declarations said — just like, for example, providing the best access for people in economically disadvantaged countries, which is likewise not explicitly mentioned in the BOAI definition but which is understood to be implied and important.

Mike, with due respect, I think treating accessibility issues for the physically impaired is too important to be brushed aside by saying “Well, our support is implicit in our statements”. Such a stance makes it very easy to recruit support on the basis of an emotional appeal while ignoring the need to push for progress on behalf of the disadvantaged group. That’s why I used the loaded vocabulary of “exploitation”.

I have watched with great interest and with some sympathy the emergence of the “open access” movement over the course of the past fifteen years. Points about the economically disadvantaged have been raised multiple times as a supporting argument for why this was a necessary approach to adopt with regard to dissemination of scientific knowledge. But I can count on the fingers of one hand the number of times, I have heard or read public statements by OA advocates that publicly referenced the needs of the physically or visually impaired. That’s why it was so striking to me earlier this month when suddenly the bullet point began appearing.

If the OA community were concerned about this particular lack of access, they could make a difference. But the way advocates treated this concern (burying a bullet point in a list that equates it with the disabling of cut-and-paste) doesn’t persuade me that they are committed to making that difference. Please prove me wrong. Get an OA advocate to examine publicly the issue in depth. Investigate the rationale for why this particular kind of access isn’t just a standard built-in accommodation, one that could be taken for granted. The visually impaired need it more than we need cut-and-paste.

I think treating accessibility issues for the physically impaired is too important to be brushed aside by saying “Well, our support is implicit in our statements”

Do you think treating accessibility issues for those in developing countries is also too important to be brushed aside by saying “Well, our support is implicit in our statements”?

See also clause 2.1.6 (Non-discrimination) of The Open Definition.

If all we needed to eradicate discrimination was a dispassionate ten word statement against it, we’d all be living in a much nicer world. Whether discrimination occurs in the developing countries or in the Western First World, the case for ensuring accessibility for the visually impaired requires assembling data more effectively (and perhaps more volubly) than OA advocates have managed heretofore.

Primarily, the real problem with Nature Publishing Group (NPG) / Macmillan Publishers is not the new “beggar access” (http://www.theguardian.com/us-news/2014/dec/02/nature-archives-free), but

1.) NPG does apparently not even know what Open Access is,
2.) NPG employees occupy positions telling others what Open Access is and if they are allowed to be part of it.

NPG writes on 20 October 2014 about its journal “Nature Communications“ (http://www.nature.com/ncomms/2014/141020/ncomms6523/full/ncomms6523.html):

“The transition to fully open access publishing establishes [note: present tense] Nature Communications as the flagship Nature-branded open access journal”.

„all subscription-access articles will continue to be available [only] to subscribers throughout 2015.“

BOAI: “By ‘open access’ to this literature, we mean its free availability on the public internet, permitting any users to read, …“

Now what is „this literature“ (= research literature)? (compiled from my own observation):
An Open Access article has to meet the BOAI definition for this article (BOAJ).
An Open Access journal meets the BOAJ definition for ALL articles in the journal (see: OASPA, DOAJ)
An Open Access publisher is one that has a minimum of ONE Open Access journal (see: OASPA).

One of OASPA criteria: „The publisher has at least one journal that regularly publishes original research or scholarship which is ALL open access.“ (http://oaspa.org/membership/membership-criteria)

Problems go on with:

„all papers submitted to Nature Communications will, if accepted, be published open access through payment of an APC. The default publication licence is Creative Commons Attribution (CC BY) … although other licence types are available on request.“

„The CC BY-NC-ND and CC BY-NC-SA licenses are available on request“

The CC BY notice at the bottom of paper in „Nature Communications“ contains this information:
„The images or other third party material in this article are included in the article’s Creative Commons license, unless indicated otherwise in the credit line; if the material is not included under the Creative Commons license, users will need to obtain permission from the license holder to reproduce the material.“

So unfortunately, if the article is CC BY this does not mean that (ALL) its content is CC BY.

Carrie Calder is on OASPA’S Board (http://oaspa.org/about/board) and on OASPA’s Membership Committee. Amy Bourke is on OASPA’s Communications Committee (http://oaspa.org/about/oaspas-committees). Both are from Nature Publishing Group/Palgrave Macmillan.

These are the people telling other people how Open Access should be defined and who is an Open Access Publisher and who is not.

More details about the discussion NPG and OA here:

Speaking in a personal capacity, as far as I’m aware, we have 1 blind reader of our magazine who visits the web site. He has been very helpful in tweaking our site to better suit the visually impaired. That’s one out of 120,000 visitors.

I’m not sure exactly what your article is in response to, so I won’t comment on any alleged exploitation. But it’s important to note that using DRM to force users to use a certain reader is almost always guaranteed to bring up accessibility issues where there were none before. You also outlined the issues for mobile users, and there are similar issues for people whose systems don’t support whatever software is required, etc.

These are all important points in any discussion of the merits of a system that uses DRM.

My concern in writing the post had been that accessibility issues should not be lightly glossed over for purposes of a politcal agenda. In serving a research environment, where workforce diversity may mean that an individual (upon occasion) requires some level of accommodation, we need to do better in providing the services that include (nearly automatically) that kind of accommodation. Our colleagues in the OA movement can help drive us towards that goal just as those working in technology can help — if it is made clear that the technology is there and the workforce requires it.

DRM harms accessibility, and that is an important point to make. Hopefully it is OK to make that point without it being considered exploitation.

I agree that some of these protective measures contribute to the accessibility barrier. And I agree that “exploitation” is a loaded word which should be used advisedly. That said, I do not think it unfair to note the ease with which advocacy groups can sometimes evince concern about an issue without ever grappling with it in a meaningful or persuasive fashion. Simply announcing you are against something isn’t sufficient to create change. (BTW, thank you for taking the time to interact here.)

My concerns about accessibility were not made casually. I have been working with a group at Birmingham University who are developing an approach to “speaking” chemical structures in publications. This is a challenge in itself, but the first requirement is to create a semantic representation of the chemistry. There are roughly three cases:

1 the authors and/or publishers provide a semantic representation. This is relatively rare but some NPG journals , among others, provide it. From that we can work, using CML, to experiment with spoken chemical primitives. This is a psychological challenge and we are at an early stage.

2 the publishers provide a pixel diagram (this is the commonest). This can often be converted into semantics but it is hard and requires imperfect heuristics.

3 the publishers make it technical or legally impossible to acquire anything to work with. That is the case here.

I am trying to persuade the community to move from 2 to 1. This is a slow process and most publishers are not interested in creating semantic chemistry. Moving from 2 to 3 is makes our work pointless.

Comments are closed.