The only difference between this and previous TWL account donations is what almost certainly got the attention of open-access advocates in the first place, the fact that the publisher in question is Elsevier. This Dutch conglomerate is one of the world's largest academic publishers and controls access to some two thousand academic journals, including the prestigious The Lancet and Cell. Moody's article notes a tweet from open-access advocate Professor Michael Eisen, whose blog and twitter feed are full of complaints about the publisher. In one post, Eisen calls Elsevier "the Dutch publishing conglomerate that has long served as the poster child for all that is wrong with the industry".

From my day job as an academic librarian, I can attest that the complaints about Elsevier by Eisen and other open-access advocates are accurate and well-deserved. There's even a Wikipedia article about the problems in my field created by publishers like Elsevier—the serials crisis, caused by escalating prices of academic journals in a time of declining library budgets and increasing demands for expensive electronic access. This crisis is all the more maddening because the massive profits accumulated by Elsevier and its ilk come from extracting money from libraries and universities based on a product that is written for Elsevier largely for free. Scientific and academic research is generated by academics, most of which is paid for by taxpayers in the form of government grants and salaries for academics working for public institutions, and submitted to academic publishers, who pay nothing in exchange above minor administrative costs. Academic journals are not staffed by employees of publishers like Elsevier, but by other academics who as part of their career portfolio edit the journals and peer review the articles as volunteers.

In a perverse cycle, the publishers then charge libraries for these publications—sometimes twice if they pay separately for a subscription and for access as part of a database. Libraries are charged an "institutional rate" that is far higher than that charged to an individual subscriber; most are academic libraries that are part of the same class of institutions that generate most of this research in the first place. For their role, which is barely more than rent-seeking, the big corporate publishers are raking in enormous profits. Last year, Elsevier reported a profit margin of a whopping 37%. In 2012, Elsevier's director of global academic relations explained the massive profits as "simply a consequence of the firm's efficient operation", a statement that still induces rage three years later in those who understand what's been going on in the academic publishing industry.

Eisen is absolutely justified to write "my concern is not about citing Elsevier articles—it's about helping Elsevier pretend it's interested in the public" or when Professor Peter Murray-Rust told Ars Technica that the accounts were " crumbs from the rich man's table ... It's patronising, ineffectual". This is simply a way for Elsevier to get a bit of good publicity at essentially no cost to them. Where we differ is in what to do next. Open-access advocates would have Wikipedia not provide Elsevier with this opportunity for publicity because of our commitment to free knowledge; but I believe Wikipedia should not heed this suggestion, because while we are committed to open access, our primary obligation is to the readers, even above taking a stand for the open-access movement by rejecting this "gift".

The open-access movement has made great strides in the area of academic journals. Academics are abandoning journals by for-profit publishers like Elsevier in favor of open-access journals and repositories like arXiv. Some of the influential government research-funding agencies are moving in the right direction, although with glacial slowness, by leaning on grantees to avoid locking up behind a corporate paywall outcomes that have been funded by the public purse. On multiple occasions, the editorial boards of Elsevier journals have resigned en masse and set up competing open-access journals.

Access to academic journals has become a significant part of the open-access movement, such as the case of activist Aaron Swartz, who in a sinister course of events was charged with violations of the US Computer Fraud and Abuse Act for downloading journal articles from JSTOR. Despite these insults to the public purse, it is difficult to see how denying 45 active Wikipedia contributors access to the articles in ScienceDirect will serve the cause of open access.

The primary obligation of libraries—whether traditional bricks-and-mortar institutions or library-like online institutions providing information services such as the Wikipedia Library—is satisfying the information needs of the populations they serve. I doubt that Professors Eisen or Murray-Rust would ask libraries at the University of California, Berkeley or the University of Cambridge not to provide access to Elsevier journals like Lancet and Cell in the name of the open-access movement. While libraries such as these can and should support open access, their primary obligation is to the academic and research needs of their students and faculty, not to the needs of the open access movement.

A similar obligation exists for the Wikipedia Library: to help editors write the best articles they can using the best sources they can. As Jake Orlowitz and Alex Stinson wrote for the WMF blog:

“

... the Wikipedia Library has to serve our readers and editors as best we can, and that means giving them as much access as possible to the best research today. Collaboration with publishers is a compromise: editors summarize paywalled content for our readers, sharing information on Wikipedia that may otherwise never be represented.

”

Wikipedia is already doing its part for open access. A recent study discussed in the Signpost's August Recent Research showed that open-access articles are 47% more likely to be cited by Wikipedia. But the fact remains that not every article can be written solely with open-access materials and not every source cited in an article will be available to every reader or editor. Take the article Nyaung-u Sawrahan, which I created as a stub in 2005 with a book from my then-university library. It now lists six books as references, two of which are in Burmese, and none of which appear to be available digitally. Most readers will not have access to these books, let alone the ability to read Burmese.

Eisen warns about a " privileged class" of editors who have access to sources while most readers and editors do not—but that already exists. Different editors read different languages and have access to different materials. Some attend universities with robust print and digital library holdings, while others live in areas with limited library and interlibrary loan access. Programs like TWL don't create those disparities. They help alleviate them by providing some editors with needed sources.

The English Wikipedia is approaching its five-millionth article. That's five million articles in one language alone that we've created and freely donated to the world. While not every source in every one of those five million is freely available at the click of a mouse, if we didn't scrutinize and reference those sources we might not have an article at all on Nyaung-u Sawrahan or any number of other topics. And those articles comprise Wikipedia's biggest commitment to the open-access movement: our readers.