Last week in popular media, Alexandra Elbakyan got a lot of screen time (also known as free advertising) and the response has been interesting. For those that have not been paying attention, Elbakyan runs Sci-Hub, a site that provides illegal access to over 47 million scholarly journal articles.

You can read about Elbakyan’s mission in her own words here, here, and here. She sincerely believes that she is above the law.

see no evil monkeys

“I developed the website where anyone can download paywalled research papers by request. Also I uploaded at least half of more than 41 million paywalled papers to the LibGen database and worked actively to create mirrors of it. I am not afraid to say this, because when you do the right thing, why should you hide it?” Elbakyan told Torrentfreak.

Despite a court injunction, Sci-Hub is still up. Forty-seven million articles are still illegally posted. For details on how this all actually works, you can read David Smith’s post here.

Elbakyan is believed to be somewhere in Russia, is a citizen of Khazakstan, and has moved her website out of the reach of US authorities. Now that I have tried to briefly identify the problem, I would like to explore the reactions of different communities to Sci-Hub.

Advocates of Open Access

A potential tragedy lurking in the background of this issue is what damage it will do to the larger open access (OA) movement. Advocates for OA, admittedly, have a good story to tell: science should be open to everyone. It’s incomplete, but still a good story. But advocacy can be a hard thing with a movement that has little organization. Loud individuals can appear to speak for the majority and you never know what you will get.

While many serious OA folks have been cautious, or silent, on the issue of Sci-Hub, mainstream media coverage, which may or may not be aided by a public relations firm working on Sci-Hub’s behalf, links the OA movement to support for Sci-Hub in almost every article. Here’s one example from Vox:

“Opponents, particularly the publishers of journals, argue that there are all sorts of costs that go into publishing science and those costs need to be recovered. Without publishers, they argue, there would be no arbiters of science. Enter Alexandra Elbakyan. Hailed as a hero by open access acolytes, she’s on a mission to make the world’s science accessible to all.”

Other posts supporting Sci-Hub can be found here and here. Others aren’t quite sure what to make of it.  Sci-Hub isn’t doing any favors to the OA community either:

Sci-Hub tweet

Publishers are warned all the time that free content is not OA content. Certainly stolen content still under copyright of the publisher, is not open access content.

Heather Joseph, Executive Director of SPARC, an organization funded by libraries to advocate for OA, spoke on NPR’s Weekend Edition recently claiming that lack of access to pay-walled content has “forced” researchers to find a work-around to access what they want and Sci-Hub is delivering.

What Joseph and Elbakyan have left out of their interviews is that there are all kinds of ways for researchers (and the public) to legally access papers, some of it free: interlibrary loan; free or low-cost access to developing countries via HINARI, Research4Life, EIFL, INASP; or even using Google Scholar to see if there is an accepted manuscript version hosted on the authors’ website, a university open repository, a funding agency repository, or a social sharing site. Low cost options include DeepDyve or article rentals which can be as low as $1.

Of course, saying that high prices drive people to break the law is sort of a losing argument if you don’t then condemn the theft in the next breath. It was disappointing that Joseph did not condemn the illegal actions of Elbakyan in her NPR interview, especially given Joseph’s role as chief “lobbyist” for public access mandates on behalf of the Association of Research Libraries members.

The self-proclaimed OA advocate that has been promoting Sci-Hub relentlessly on Twitter is Jon Tennant, the Communications Director for ScienceOpen. He even posted a bookmark on his Github page that automatically adds the Sci-Hub web extension to any DOI link you happen upon.

This all seems pretty radical for the employee of a company that appears to have an honest approach to providing authors with an OA solution. ScienceOpen has recently started approaching publishers about partnerships around indexing of content. I am not sure that having your Director of Communications actively openly advocating for copyright infringement against those partners is helpful in building those collaborations.

The OA movement has made great strides in the last decade. Folks are starting to come around to thinking that not all OA journals are predatory or chock full of junk science. Public access to federally funded research, something threatened for years, is now a reality. No question, some people are doing things right. But when the people who are doing things right don’t condemn the folks that want to burn the place to the ground, their message goes up in smoke right along with it.


This whole situation is leaving librarians in a bit of a quandary as well. Sci-Hub is accessing publisher platforms via “donated” credentials coming from university library patrons. The libraries have paid for the credentials that their patrons are “donating” to Sci-Hub. At least one library also claims that Sci-Hub, or someone on their behalf, obtains credentials through fraud and email phishing scams.

Libraries sign license agreements and terms of use that do not allow them to provide access to folks outside of their pre-defined community, which usually consists of students, faculty, and increasingly, alumni. When patrons of those libraries “donate” their login credentials, this violates those license terms. Unfortunately for the libraries, it’s not hard to tell which ones are unwitting (or otherwise) participants in the illegal activity.

Many publishers watermark the PDFs with information about from where and when the article was downloaded. It is not unusual to see the university name or the actual user if they are registered on the platform included in the watermark. At a minimum, the IP address is included. It is quite easy to see where Sci-Hub is pulling from and where LibGen hosted content originated. We have found our own PDFs with watermarks from University of California, University of Colorado, Florida Atlantic University, Pennsylvania State University, and so on.

The access gateways and proxies used by libraries are also used to connect to other university services. Potential copyright infringement liability aside, one would think that librarians and administrators would not be happy knowing that lawbreaking organizations now have access to their private networks and potentially, all the confidential information about students and employees that they contain.

In this article posted by the Chronicle of Higher Education, librarians are again slow to condemn the service and instead have chosen to focus on the “broken system.” The universities have a responsibility to detect and stop this behavior of their patrons. Is it not up to a university to expect a simple code of conduct from their students and faculty?

Publishers Whose Names Don’t Start with “E”

If you hang around on Twitter and follow folks who are passionate about open science, you will quickly notice a pattern — everything Elsevier does is evil! The big dark dirty secret is that almost all scholarly publishers do things the same way as Elsevier. We have similar platforms, similar tools at our disposal, and similar concerns in many cases.

All that said, Elsevier takes almost all the flack and truth be told, many other publishers appreciate that. Flying “under the radar” is what scholarly publishers do best.

Many of us use the same company to find pirated content online and many of us are wasting time sending loads and loads of take-down notices to Sci-Hub and LibGen. Now is the time to take the next step.

It may have taken a while but publishers have certainly embraced the OA model and are proponents of giving authors choices, even if those choices aren’t popular with the OA folks. Never has there been a time when authors can choose a license or negotiate terms with a publisher for journal content like they can now. Re-use of content is more relaxed, prior use of content is taking off! Pre-print servers are popping up and publishers are NOT slamming the door.

Publishers are also adapting to the public access mandates. Some are even funding a “clearinghouse” to make things easier and cheaper for federal agencies while also helping to make FREE versions of content more discoverable and accessible to the public.

I’m not saying that all is perfect and I am not calling the publishers “progressive”. I will admit that some of this comes with a lot of kicking and screaming. What I am saying is that a line needs to be drawn. Stealing content and reposting it for free is over that line.

Elsevier is leading the legal fight against Sci-Hub and LibGen, and thank goodness they are. Publishers who have the means should be joining them. Those that don’t should be speaking out. If Elbakyan actually has over 47 million articles, then she probably has your stuff. I highly doubt her usage is as high as she claims but with this publicity tour, it will go up.

So what can publishers do? Work with your library administrators. Work with your platform providers on access control issues. Talk about the consequences of Sci-Hub (the list below is a start).

Potential Consequences 

Talking about the consequences of Sci-Hub is another post altogether. But I have been collecting a few as I discuss the issue with others in the industry.

Credentials. As David Smith has pointed out, the security of the library credentials is poor and allowing criminal elements into the system should be terrifying. Publishers could insist on more robust (perhaps expiring credentials) from the universities as part of the future licensing agreements. This would be inconvenient, at best, for the universities.

More restricted download counters. Most publishers have fraud detectors built into the platforms that monitor heavy downloads from within an institution. The threshold is different depending on the publisher. If the fraud monitor is tripped, the entire university loses access until the publisher and the library have a chance to clear things up. The future could hold more stringent monitors with more disruptions of service.

More restrictions on the use of content. Like PDFs? Say goodbye to them. Publishers could opt to change access models restricting the download of PDFs on servers or just eliminate the PDF altogether. The PDF has always been a “leakage” problem and publisher have been trying to get rid of them for years. They are expensive to create, offer limited use of advanced features, and don’t help us keep the “eyeballs on the page.” The problem is that they are insanely popular with almost all scholarly content users. Publishers are investing a lot of money in readers (ReadCube, colwiz, etc.) and in personal library storage systems (Mendeley, ReadCube, colwiz, etc.). Eliminating PDFs and instead only providing an online collection of links could be coming to a desktop near you.

Financial implications. While it may be a favorite past-time for some to think about the destruction of commercial publishers, the societies go down in flames with it. Elsevier, Wiley, Sage, Springer, Taylor & Francis may make a lot of money but they also help hundreds of societies make money on their publications in order to serve up continuing education, k-12 outreach, professional development, etc. Independent societies are much more vulnerable to harm from Sci-Hub than the big corporations.

There is no question that Sci-Hub is illegal. They are keeping the site live despite a legal injunction. Elbaykan is happy to talk to the press and make specious legal arguments, but has never actually had the courage to show up in court. As the OA advocates, librarians, and publishers try to create a more open and collaborative environment, they should condemn this solution and realize the harm their silence will cause.

Angela Cochran

Angela Cochran

Angela Cochran is Vice President of Publishing at the American Society of Clinical Oncology. She is past president of the Society for Scholarly Publishing and of the Council of Science Editors. Views on TSK are her own.


167 Thoughts on "A Funny Thing Happened on the Way to OA"

Dear Angela,

Thanks for linking to my post. As it says on the website, and also on my Twitter handle, those comments and opinions are my own and do not reflect those of my employers, including Imperial College London. I have also never announced that I am an ‘OA advocate’, but I do support open access in general. The GitHub link again is personal, and built by others – I simply forked it. My actions are independent of those of ScienceOpen, and that has always been clear.


Also, just to make it absolutely clear, ScienceOpen does not support SciHub.

She sincerely believes that she is above the law.

I think that is a bit of a misinterpretation. Nothing Elbakyan has said seems to communicate the kind of arrogance of exceptionalism that this implies. It’s more that she sees what she’s doing as obeying a higher law (whether correctly or incorrectly is, for the moment, beside the point). In other words, it’s not that she thinks she is anything special, but that what she’s doing is special.

And by “a higher law” she doesn’t just mean an abstract sense of what is morally right, but very concretely article 27 of the UN declaration on human rights.

As you noted, I’m not really sure what to think about all this (for those who don’t want to my whole article on this, the punchline was “Heck if I know”. But I do think we at least owe Elbakyan a proper reading of her motivation, and I don’t think “above the law” really captures it.

Article 27 – Universal Declaration of Human Rights

Everyone has the right freely to participate in the cultural life of the community, to enjoy the arts and to share in scientific advancement and its benefits.

Everyone has the right to the protection of the moral and material interests resulting from any scientific, literary or artistic production of which he is the author.

The second clause certainly seems to relate to copyright. But you will notice that it specifically refers to the rights of authors. So where this applies to scholarly publishing at all, I take it to be a statement against mandatory copyright transfers, rather than a reinforcement of publisher-owned copyright that have been taken away from authors.

Nothing is mandatory. If you don’t want to sign a copyright transfer, there are many other options available.

That has not been my experience. I have had many long, tedious conversations with publishers on this subject, and they have not generally been keen to observe Article 27 part 2.


To be fair, I am referring here to experiences that I had some years ago. Ever since I Became Enlightened, I only submit to open-access journals, so the issue doesn’t arise.

Of course, publishing open access is a way to retain copyright over your work as an author, and works at the article level. I’ve found this to be a much more constructive approach as an academic, and this is I think the future of how open access and copyright will look. But this is the future, and Sci-Hub is about the past..

But this is the future, and Sci-Hub is about the past.

That is a very important point, and much overlooked. The enormous majority of open-access schemes that have been proposed and implemented related to work being done now, to be published in the future. It’s nice, and not completely unrealistic, to imagine a future in which all scholarly publishing is Gold OA, and publishers are appropriately paid for services rendered without needing to lock up knowledge. But even in that utopia, there’s the small matter of the last 300 years of scholarship, most of which remains under lock and key. Until that changes — and I don’t know of even any serious proposals towards fixing it — Sci-Hub and similar systems will have a purpose.

“If you don’t want to sign a copyright transfer, there are many other options available.”

Somehow, publishers always “forget” to present those options.

She is quite clear below that her goal is to ‘correct’ the law by abolishing copyright, so she in fact opposes that UN listed human right. It is therefore incorrect to cite it as a basis for her actions.

For decades I have been trying to persuade my literary friends to write a play. The basic plot can be transported to a variety of environments. In the present context Angela has a child with incurable disease X. In several years time the child dies. A few months later Alexandra comes up with the cure for X. She publicly states that, had she had better access to the literature, she would almost certainly have found the cure years earlier. Angela realizes that her child’s death was partly her own doing. The curtain falls. [Full disclosure: With the support of some publishing houses circa 1994 I initiated the online discussion group Bionet.journals.note (now defunct), the mission of which seems to have been similar to that of The Scholarly Kitchen.]

Thanks for your interest in Sci-Hub !

I would like to point out that some of the arguments you’re making a flawed, and emotional rather than rational.

for example, you say:
‘Elbaykan is happy to talk to the press and make specious legal arguments, but has never actually had the courage to show up in court’

Yes, Elbakyan pretends to be courageous, but has never actually had the courage to jump from the skyscraper roof…
Even further, to show up in court, I would have to spent money and time. And since I was not going to win that particular trial anyway, I made a perfectly rational decision to invest these money and time working on Sci-Hub.

Regarding talking to the press, I take any opportunity available to explain my position and motivations for creating Sci-Hub, as well as to let people know what is going on.

‘she is above the law’
That would be a mistake to think our laws are perfect. Sci-Hub is not breaking the law, but pointing out particular spots in the law system that are not perfect and need corrections. When people do not see anything immoral in free science and education websites, why should them be defined illegal?

I do not see Sci-Hub as destructive, because no subscriptions were cancelled so far. If universities feel it would be just to support publishers, they can continue their subscriptions. What’s wrong?
But as you agree, subscription-based model is outdated, and publishers are switching to OA increasingly. Sci-Hub makes no damage to publishers operating within Open Access model.

‘Independent societies are much more vulnerable to harm from Sci-Hub than the big corporations’

Yes, I’ve seen this article, but it makes no sense. The article says Sci-Hub would be damaging to non-profits that depend on subscription sales. But that is self-contradictory. How can non-profit depend on sales? The traditional model for non-profits is to collect donations.
Regarding scholarly societies, maybe they can try collecting membership fees?

Again thank you for your rpecious time!
Best regards,

Is copyright per se bad law or just for journal articles? I am trying to understand the scope of the crusade.

Alexandra. Thanks for stopping by. I am not interested in debating with you whether what you are doing is illegal. You have made your point clear and I have made mine.

You are completely wrong about non-profits. Non-profits sell billions of dollars worth of products and services all over the place. The difference between a non-profit and a for-profit has to do with what happens to the money it makes. Here is some info on that:

Professional and learned societies publish journals and sell subscriptions (or make money on APCs) in order to fund their mission. Many of these societies also have membership dues. Libraries use download activity as a way to determine future subscriptions. If library patrons go to your site instead of the publisher site, library usage will go down. I know you don’t care about that but one consequence is that libraries could cancel subscriptions due to low usage. I am not saying that will happen because I have my doubts about your usage claims. Anyhow, libraries are more likely to cancel subscriptions offered by the smaller non-profit publishers. It is difficult to unbundle a poorly downloaded journal from a package offered by large commercial publishers. So on this way, non-profits are more at risk of damage than commercial publishers due to your activities.

I do not agree that the subscription based model is outdated. I think there is plenty of room in the market for lots of business models.

A non-profit that depends on sales can hardly be differentiated from for-profit, commercial organization.
The true non-profits collect donations for their cause, and then use collected money to make the world better.

Anything other has to be labeled ‘non-profit’ with a great caution.
I highly doubt that publishers who restrict people from information access (i.e. those who need paywalls and subscribtions) can promote themselves as ‘non-profits’

That’s all I wanted to say on a non-profit topic.

The ignorance in this statement is a bit staggering. I will try to be polite in my response. You would do well to spend some time learning about not-for-profit (NFP) organizations and their need to be self-sustaining. Any NFP that cannot generate a surplus of funds is unlikely to survive for very long, be able to adapt to changing conditions or launch new initiatives. Putting the important work that such organizations do at risk out of ignorance is extremely irresponsible.

I see that you have not found any better argument to reply than accuse your opponent of some ‘ignorance’. I will take this as a kind of a personal attack. When personal attacks become involved, it is better to stop the discussion.

Alexandra, please hear this from someone who is not your enemy: the term “non-profit”, at least in the USA, has a specific meaning, and it simply does not relate to what the sources of income are. It’s only to do with governance. This is merely a point of terminology: I have a lot of sympathy with your sense that there’s something wrong with a non-profit that obtains its funds by preventing access to scholarship; but doing so does not that they’re not a non-profit.

You know, when some organization advertises itself as a non-profit, people are not going to imagine a greedy publisher that restricts access to information with paywalls. When hearing ‘non-profit’, people are going to think about all those organizations that help poor people in Africa to get education and food, and et cetera.

But those publishers who use paywalls do the reverse: they deny these people their chances.
That’s why I’m saying that is not fair to advertise these publishers as ‘non-profits’, even though they can be non-profit by some formal definition.

Yes, that’s why this article has just defined Sci-Hub as a black side warrior.

Dear Sir,

You have to check Wiki facts about this company:

Cancer Research UK’s work is almost entirely funded by the public. It raises money through donations, legacies, community fundraising, events, retail and corporate partnerships. Over 40,000 people are regular volunteers

They can sale something, but they do not depend on sales entirely, and even further, do not depend on restricting access to important medical information. Donations are listed as the first source of their funding.

And similarly, most research societies that publish journals raise funds through a variety of means–selling subscriptions, membership dues, meetings, etc. Again, please take time to learn more.

They may collect their funds from variety of sources, but the one based on restricting access to important scientific information through government-forced Internet censorship is not an ethical business model. And is not neccessary when there are other options, including those you have listed.

Thanks, that’s a response that can actually lead to dialogue. I’m glad you are now willing to acknowledge that not-for-profits do indeed use different methods including subscriptions to raise funds for their charitable works. It is clear that you have ethical issues with some of these methods, but that’s a different statement than denying that they exist. It is also important to realize that those NFP organizations are the most vulnerable in the publishing space, and that your actions put them at much more risk than it does the big corporations. This may not dissuade you from your course, but it is worth considering.

To the best of my knowledge, Reed Elsevier PLC, Reed Elsevier NV, and RELX group are not really NFP organizations nor charities. Am I wrong?

To the best of my knowledge, Reed Elsevier PLC, Reed Elsevier NV, and RELX group are not the only publishers of scholarly literature, nor are they the only publishers whose copyrighted materials can be found on Sci-Hub.

Setting aside the ad hominem attack which does your argument no credit, the choice of Cancer research UK as an example to support your view is curious. CRUK is likely to be a beneficiary of the initiative that Dr Elbakyan’s is taking as it will allow more of the donations it receives to go towards research and less towards commercial and even NFP publishers. On a broader note you have given her the choice of being either a liar or ignorant. I’d say she is right not to take you seriously.

Your concern trolling is noted. When a criminal visits a professional forum and tries to spread misinformation in an attempt to justify her criminal actions to the very people she is stealing from, then clearly the most important thing to consider is making sure her feelings do not get hurt.

As for CRUK, first, it is unclear how much, if any of their funding goes toward journal subscriptions. Funders are able to clearly specify how their funds can and cannot be used by researchers and their institutions. Clearly, the literature is a vital tool for the successful researcher. Is it wrong for a research funder to support paying for the tools and services that their researchers need? Should we similarly try to steal from test tube manufacturers or service providers like DNA sequencing companies because that would also help stretch out research funding?

On a broader note you have given her the choice of being either a liar or ignorant. I’d say she is right not to take you seriously.

She has stated something that is factually incorrect. That either means that she knows it is factually incorrect and is still using it to justify her criminal activity (which would make her a liar) or she does not understand the situation as well as she claims to and does not know that her claims are incorrect, which would make her ignorant:
adjective ig·no·rant ˈig-n(ə-)rənt
: lacking knowledge or information
: resulting from or showing a lack of knowledge

Are you suggesting a third alternative here for her factually incorrect statements?

When a criminal visits a professional forum and tries to spread misinformation in an attempt to justify her criminal actions to the very people she is stealing from …

David, this is unworthy of you. Here, I feel you have yourself crossed the line into trolling.

See my recent piece What is Alexandra Elbakyan’s motivation for creating and running Sci-Hub?. Whatever one’s conclusions about Sci-Hub, I don’t accept that demonising terminology like “criminal” can possibly be appropriate in this situation. At the very least, the situation is much more complex than that. Aren’t you the one always telling me to pay more attention to nuance?

Sorry Mike, no, I stand by my word choice. Words have meaning. In this case, Elbakyan is clearly violating international law, and has clearly stated that she has done so and intends to continue to violate the law.
adjective crim·i·nal ˈkri-mə-nəl, ˈkrim-nəl
: involving illegal activity

You are getting into arguments over her motivation for violating the law, but there is no doubt that the law has been violated.

Seems to me you’re picking nits over the difference between a legalese definition of a word and the common language definition of a word. See my definition above. Has a law been violated? Then the word is appropriate.

Well, David, this is rather a tedious side-issue, so I will leave the last word to you and let it die. But it seems to be that you were the one who, to justify your use of the emotive words “criminal”, resorted to a pedantic definition; then when it turned out that the definition didn’t actually apply, you decided it was too pedantic.

I think it was always inevitable that this thread would descend into name-calling at some point. I just didn’t expect it to start with you. It’s neither accurate nor helpful.

And with that, I’m out of this sub-thread.

Note also that since the site solicits “donations” from users, it would qualify under your definition of “criminal” anyway.

Sorry Mike, the term is accurate. Laws have been broken. If pointing that out is too much for your delicate sensitivities, so be it. If only Elbakyan cared as much about not hurting the feelings of others as you clearly do. I’d rather have someone call me an accurate term that I didn’t care for then steal from me.

I will chip in on the question of “criminal” from a legal point of view. To call someone a criminal when they have not been convicted of a crime is quite simply wrong, and may be defamatory. David may wish to use the word in a popular sense but unless and until the owner of Sci-Hub is convicted an offence which carries a judicial penalty (as opposed to damages), it is in law wrong to call them a criminal. Does David think it is OK to call a lawyer a criminal because, thanks to their incompetence, they get sued for damages by one of their clients?

To add to this pedantic subthread, I’ll also note that whatever laws Elbakyan is or isn’t breaking, “stealing” isn’t the right word to use. Copyright infringement is not theft.
A red herring is something that misleads or distracts from a relevant or important issue. It may be either a logical fallacy or a literary device that leads readers or audiences towards a false conclusion. A red herring might be intentionally used, such as in mystery fiction or as part of rhetorical strategies (e.g. in politics), or it could be inadvertently used during argumentation.

From what you have said, either you are ignorant or you are deliberately lying. I am trying to give you the benefit of the doubt and assume you actually believe the things you are saying. However, those things are flagrantly incorrect. As I said above, please take some time to educate yourself more about how NFP organizations work before causing them further harm.

Refusal to listen to facts does not make one any less ignorant.

As I told before, I have nothing to discuss with you further because you resorted to a nasty kind of personal attacks, simply calling other people ignorant instead of providing proofs is not going to make any interesting discussion.

Running and hiding doesn’t make you any more right. As I said, please do some work to better understand the nature of NFP organizations.

Nobody is running away. I simply prefer polite discussion, not the one you’re trying to start.

But rather than respond to the substance of my comments, you choose to continuously complain that I am pointing out your ignorance. If you have extensive experience running a successful not-for-profit organization, or an extensive knowledge of the various business models that such organizations use, it is not evident from your comments. Several here have tried to offer correct information on this subject which you seem unwilling to accept.

Dear Sir,

I replied to substance of your comments already.
And please learn to be polite on your discussions.
I know it may be hard for you when you understand you’re wrong, but this is not the reason to resort to personal attacks.

“How can non-profit depend on sales?” That is beyond simple-minded!

I wonder whether it would have been possible to express your disagreement in a more constructive way? Perhaps something like “The term ‘Non-profit’ refers to the governance structure and aims of an organisation, and doesn’t in itself constrain the activities by which the organisation brings in funds”?

Are you familiar with the distinction between “profit” and “revenue”?

In classic social media style, Sci-Hub is now asking for donations to keep the site running. There is a donate button and contributions are made using Bitcoin.

This has been the case for a long time — but I don’t understand the connection with social media.

Can anyone envision a system where the work that publishers do and get paid for could be done elsewhere instead? For example in the colleges, universities and research centers where this knowledge is usually produced?
In other words, why the big, expensive round trip through corporate publishing? Couldn’t an organization on the scale of EduCAUSE handle all or most of this?

One could go way out into the imagination and think about the idea of a university starting their own publishing activities, maybe even call it something like a “university press”.

Most colleges, universities and research centers cannot achieve the necessary economies of scale which is why I specified “on the scale of EduCAUSE.” This cannot be something that only the elite institutions do and still obviate the need for expensive commercial publishers.

I suspect Sandy Thatcher will join the conversation shortly to talk about Project Muse and JSTOR.

Not to mention HathiTrust! And, gee, we have the Power 5 football conferences making billions of dollars, don’t we? Isn’t that scale enough?

Elbakyan evidently has a very simple-minded understanding of how publishing works. Her actions pose a threat to university presses as well, and these are non-profit publishers supported by universities but generally expected to recover about 90% of their costs through sales, including subscriptions to journals for those presses that publish journals. Reducing that income will have either of two consequences: 1) putting that press out of business, or 2) forcing the university to increase its subsidy to the press, which would mean taking money away from other uses (such as support for the library) or increasing fees (such as raising tuition for students). Is Elbakyan comfortable with those consequences?

Since Elbakyan puts herself out of reach of effective legal action, how about mounting a “denial of service” attack on her website? What would she do–go to court to challenge such action?

Since Elbakyan puts herself out of reach of effective legal action, how about mounting a “denial of service” attack on her website? What would she do–go to court to challenge such action?

Sure, if you want to throw away what claim you have to the moral high-ground.

(BTW., this has already been done. Sci-Hub has survived DoS, and likely will do again.)

Surviving DDoS is not a problem, but DDoS attacks cost money. And since we all know how cautious are publishers about their money, I highly doubt any attack will happen at all.

Moral high ground? If you know anything about the history of the civil rights movement, you’ll know that the moral high ground was achieved by those who broke the law but who also accepted the consequences of doing so by going to jail. Elbayan has not achieved that moral high ground by refusing to accept the legal consequences. She wants to eat her cake and have it too!

Sandy, the point here is that publishers, including yourself, seem to assume that they have the moral high-ground with respect to Sci-Hub. Attempts by publishers to DoS the Sci-Hub service will eat away the high-ground from under your feet. For that reason, I think it is an poor strategy, and you were in error to propose it.

Note that this is entirely independent of who really is in the right morally. I’m just taking about public perception here.

Not to mention the legal ramifications, quoting from (

Many jurisdictions have laws under which denial-of-service attacks are illegal.

In the US, denial-of-service attacks may be considered a federal crime under the Computer Fraud and Abuse Act with penalties that include years of imprisonment. The Computer Crime and Intellectual Property Section of the US Department of Justice handles cases of (D)DoS.

In European countries, committing criminal denial-of-service attacks may, as a minimum, lead to arrest. The United Kingdom is unusual in that it specifically outlawed denial-of-service attacks and set a maximum penalty of 10 years in prison with the Police and Justice Act 2006, which amended Section 3 of the Computer Misuse Act 1990.

Seems ironic that DoS attacks would be illegal against sites that are themselves illegal. If those harmed cannot fight back, what are they to do? Gee, maybe drone attacks? Hire Blackwater operatives?

Sandy, I find it very distasteful that you would contemplate violent military action as a way of preventing people from reading and using scholarly research. I can only hope that this is a joke; if it is, then it’s one in poor taste. (Imagine your reaction if an OA advocate were to suggest car-bombing Elsevier headquarters. I hope we’re all better than that.)

Yeah, I was joking, Mike. I thought that would maybe be obvious once I started suggesting Blackwater operatives, but I guess you lack a sense of humor. And what on earth makes you think I’m a fan of Elsevier? I guess you haven’t read my many articles promoting OA, have you? They’re all available (OA, by the way) at Penn State’s IR.

I am glad to hear that it was a joke. I couldn’t tell because, so far as I could determine, your earlier suggestion of a DoS attack on Sci-Hub did appear to be serious.

It’s worth noting that if sites are egregious enough to attract serious attention (eg piratebay etc) then for most people, they can be effectively removed from the web. There are various mechanisms to enable this to happen (they also take time). Personally I’m not a massive fan of such things because of slippery slopes and all that. The technology to do that is indifferent to the quality of the argument or the legitimacy of the site. But it’s pretty clear that when you look at the evasiveness over how the access credentials are obtained, this is not some principled stance – it’s a VERY shady operation indeed.

But it’s pretty clear that when you look at the evasiveness over how the access credentials are obtained, this is not some principled stance – it’s a VERY shady operation indeed.

I don’t think we can know that at all. If I were running a site that used other people’s credentials, and if those credentials were willingly donated, I would consider it my responsibility to those people not to disclose the fact.

And I don’t think “this is not some principled stance” is at all a tenable position. You only have to look at the nature of Elbakyan’s arguments — the appeal to human rights laws, for example — to see that a principled stance is exactly what it is. They may not be principles that they agree with, but they are evidently ones that Elbakyan believes in very strongly.

And how do you know she’s really not just evoking those principles as a way to justify her unprincipled behavior? Do you know her personally? Why should we rake her words at face value? Do you think the Kim Dotcom was fighting for a principle with his Megaupload?

How do I know Elbakyan is not lying about her motivation? I suppose I don’t know, any more than I know you aren’t lying in your comments, or David Crotty in his, or Tom Reller in his. My policy is to assume people are telling the truth unless it can be shown that they’re not. I thought that was pretty universal.

The other half of this equation is: what do you think Elbakyan has to gain? She is putting herself in the firing line of a big, well-resources and litigious corporation in return for … what? A couple of BitCoin donations? I can’t see any credible motive for her actions other than the ones she claims, namely to make scholarly research freely available to the world. Doesn’t Occam’s Razor suggest to you, as well, that is the real reason?

Remember the history of Napster and Mendeley? They caused enough trouble to their industries that they were finally bought out, Napster by Bertelsmann and Mendeley by Elsevier. It is hardly far-fetched to think that Elbakyan might be hoping for a similar lucrative buyout from the publishers she has annoyed. Remember that the founders of Napster and Mendeley also talked the language of morality in their defense, but had no trouble selling out when the time was ripe.

Sandy, I can see that you’re very emotional about this issue, and this has caused you to speak loosely in your comment here: to which I refer by link because threading on this blog has been troublesome for years.

There was never anything shady or wrong in what Mendeley did, thus we never needed to defend ourselves from anything. How VC-funded companies work is that they either go bust, go public or get acquired. It’s only a sell-out if the company stops doing what customers loved it for and we haven’t done that. Mendeley was valuable enough to enough people we weren’t going to fail, but our niche also isn’t big enough that it made sense to go public. We continue to provide our valuable service as a part of Elsevier and though we were forced to make some minor changes that we probably wouldn’t have otherwise, we are now in a position to show the industry how services can be a growing part of revenue, even as subscription revenues falter. Did you notice RELX posted gains for 2015 ahead of expectations? Take a look at what drove those gains and come back here.

We did and do feel very strongly that Open Access is the right thing for the research community precisely because it enables the creation of more valuable services. This has been our argument from day one and it is our practice today. Everyone in our industry, commercial and society publishers both, should pay close attention to the signals that are being broadcast by researchers because those signals tell us what valuable services need to be created. Because we have no intrinsic right to exist, we must demonstrate value to them. If our value proposition is so weak that a little website cooked up by one person can threaten it, we need to work on our value proposition! To continue your example of music sharing, what stopped the reign of Napster et al, wasn’t their acquisition, it was iTunes and the like. Access to your music on-demand on your device of choice is what people wanted and Sony was institutionally incapable of capturing that market, so Apple did and they became the biggest company ever. The path to success for a for-profit company will always be, “Give customers what they want.” As Steve Jobs famously said, this may not be what they say they want, so it’s our job to find that out, and we’re getting a very loud message right now.

I have some possibly ignorant but persistent thoughts on the non-profit thing, about which I would love feedback. The difference between a for-profit entity and a non-profit one is that when people don’t want what a for-profit offers anymore, the company goes away or re-invents itself. A non-profit, however, can continue for years pulling in donations from people who aren’t the consumers of the product. I sympathize with the plight of scientific societies, who have provided valuable services to their membership for years, and with university presses, which provide an outlet for authors who otherwise would be ignored by big commercial labels. I don’t think you have much to worry about, though. You might need to do some fundraising a little differently, but as long as people still think you’re worthwhile, you’ll be fine. Commercial enterprises are in the process of re-invention and clearly we’re not moving fast enough to stay ahead of where the customer wants us to be in terms of ease of access and discovery. That’s a business challenge we are taking on and in which we believe we will be successful.

So, Dr. Gunn, are you saying that the owners of Mendeley received no monetary reward from the sale to Elsevier, that it was all done in the spirit of altruism?

If Napster was done in by iTunes, then why did Bertelsmann buy it? A large company like that is not usually guilty of buying valueless properties.

Yes, i know all about providing valuable services for OA I sit on the board of the CCC and was an early advocate of having the CCC do just that. One of its mist successful recent ventures was providing a service to normalize4 the formats of journal articles so that they could be more easily subjected to text and data mining.

University presses in the US do not spend much time trying to raise money, apart from the occasional effort to apply for an NEH Challenge Grant (which I successfully did at Penn State). Their operating costs are largely (90% on average) covered by the revenues they generate, and the rest comes from a university subsidy.

As a poor student, I had to budget and fix my own meals. When I made money, I could afford a cook, or a meal delivered. Academics used to curate their own articles and share the task with their collegues. Journals eased the effort to distribute and publishers were willing to provide the value added service. Somewhere along the line, even the worth of where the article was published took the burden off faculty evaluations and pub/perish was born. Now authors are caught in this circle and think more about the quantity of their oeuvre and impact factors, often dominant in the decision even how to split a thin piece of research into multiple articles- all making the service providers, the publishers happy to facilitate, creating new journals at a lively pace to feed the “caucus race”

The rise of the Internet leads back to reconsidering this situation and Sci-Hub, Open Access etc are raising issues that question how much is truly behind those pay walls that should be open. Is the form and format as well as the entire system supporting the publishing arena a house of cards on the edge that is not just questioning the paywall of the obliging publishers but once fully in the open will it question the entire academic caucus race.

I wonder if Mike Taylor would explain to us whether he thinks a counterpart to Sci-Hub for the humanities and social sciences would not also be a good thing and, if it is a good thing for journal articles, why it would not also be a good thing for books in those fields. I should think that Article 27 would apply equally well to scholarly books as to scholarly articles. Why not free those to the world at large also?

Sandy, if you read what I’ve written about Sci-Hub, it will be apparent that I have not yet even come to a conclusion on whether it’s a good thing for science articles. I’m certainly not going to leap ahead to conclusions for these other issues.

I’d like to chime in here with a couple of recurring themes I continuously try to point out to reporters. (Though sadly, they never seem to make it into the article). (1) Is that AE and too many others conflate the submitted manuscript with the final published article. This relates to the unfortunate age-old soundbyte, “what the taxpayer paid for, the taxpayer should read.” What the taxpayer (or funder) actually paid for is the work, the summary and manuscript. Anything else beyond the point of submission the taxpayer didn’t pay for. Those taxpayer funded manuscripts, have been, are, and will forever be available to read by Joe Taxpayer, free from any publisher interference (Glamour journals excluded). Indeed more and more people are using repositories for this purpose.

The only reason anyone sees any use to what SciHub holds is because they host branded, final published articles – specifically not the submitted versions the taxpayer funded. If they just want the work the taxpayer funded, they can go to a wide variety of repositories, email the author, go to a library, etc. So, if, as too many people say, that publishers add no value, there is no need for SciHub, because that’s a repository of value-added articles. The whole “they don’t add value” line of reason is an utter contradiction. If publishers didn’t add value, SciHub wouldn’t exist.

All to conclude, that undermining Journals, publishers and the investments they make to ensure the system operates effectively, and results in the articles academics demonstrate they want to read, seems awfully counterproductive. If all this knowledge were truly free, with no incentive to do all the things publishers do (or manage, or regulate, or host, etc.) between submission and publication (and beyond), it would cease to be the desired knowledge people like AE try so hard to obtain.

And (2) Elsevier and other publishers have a wide variety of free-to-low cost access programs in place today already. A list of those from Elsevier (including industry initiatives like Chorus) are available here ( @Forsdyke might be particularly interested in PatientInform. (

And there also are a wide variety of other available version and access programs via green OA. Not to mention Gold OA.

This is all so simple, but it gets lost in advocate rhetoric. We all need to keep working to get people to understand the difference between what academics do, and what publishers do in support.

Actually, Tom, in many, perhaps most cases the taxpayer does not pay for the manuscript. There are two reasons for this. First it is often written well after the grant ends. Second, the budget for the grant probably seldom includes hours for writing articles. I have never seen one that did. As one Program Officer told me: “We do not pay people to write articles. We pay them to do research.” (Mind you this is just a conjecture, as these are empirical questions.)

So who do you think is paying for the articles to be written, David?

It seems pretty clear to me that, whatever the wording of the grant says, scholars consider the writing of the paper as a part of the work that they’re expected to do.

But if that’s not the case then I think we have to say that scholars write papers out of the goodness of their own hearts — which makes it if anything even more reprehensible for them to be locked up.

I think in most cases their institutions pay for the writing. Also, expectation is not compensation. This is important. You are changing the subject, that is, deflecting.

Even if you’re right (which I don’t accept) that would only mean that the institutions would have a claim to the copyright. It doesn’t get us anywhere towards your desired goal of justifying the publishers’ ownership of the paper.

I stated no such goal. (Where do you get this stuff?) The clear issue is what taxpayers pay for.

Eliminate PDF. Make them read from the screen. Suppress Printscreen button. Forbid to take photos of the screen. Make them come to your office to read. Don’t publish. Read it yourself under blanket with a flashlight.

Brilliant! The blanket and the flashlight is perfect. In all seriousness, my point with this post is that there are harms and consequences of Sci-Hub, even if the Sci-Hub creator herself doesn’t understand that. Systems that have been in place for literally hundreds of years move slowly. Not everyone in academics and research care about open access. I know that is hard for some to believe but it is true. If everyone did care, the entire system would be easy to change. When the nuclear option is thrown down (take it all and give it away to everyone else), those working to effect change should condemn the illegal actions and use this as a learning experience to move forward.

Any technical obstacle can be circumvented, so eliminating PDF will not work against Sci-Hub, unfortunately 🙁 Perhaps you do not understand that, but for me it is kinda obvious.

By all means, feel free to share with us the technical details on how you are taking all the PDFs.

“My point with this post is that there are harms and consequences of Sci-Hub.”

I think we can all agree that that is true.

I rather hope we would also all agree that there are benefits of Sci-Hub.

The question is: which is the greater? I don’t think that’s obvious at all, and I suspect it won’t be possible to know for some years yet. One possible outcome would be that this cutting of the Gordian Knot induces a very rapid transition to a 100% gold ecosystem — a consequence that some would welcome (me included) and others very much not.

Yeah, this attitude in the article really struck me, too; I think responding to this by making content *even harder* to access, whether via DRM or other means, is really tone-deaf.

I am not a fan of DRM. I am all for making content easily accessible in whatever formats are wanted to those who have paid for it.

Thanks all for this “interesting” discussion. I’m not sure I have much to add than I am off to watch Godzilla (2014). YMMV.

Well I was going to (and might still do later), but I’m off to nurse my rather unwell neighbour until her son arrives. #terminalcancersucks

“What Joseph and Elbakyan have left out of their interviews is that there are all kinds of ways for researchers (and the public) to legally access papers, some of it free: interlibrary loan; free or low-cost access to developing countries via HINARI, Research4Life, EIFL, INASP; or even using Google Scholar to see if there is an accepted manuscript version hosted on the authors’ website, a university open repository, a funding agency repository, or a social sharing site.”

So let me play the devil’s advocate here for a bit.

If no one’s bothered by all those other free ways to legally access papers, what makes this one additional way to do it such a big deal? How is Sci-Hub not perceived merely as a technological refinement of an ages-old practice of scholarly sharing? Because that’s what it essentially is, the only difference being that Sci-Hub now allows the “author” to give you the key to his library instead of having to manually retrieve the papers himself to hand them to you one paper at a time.

I am glad you asked.

1. Inter Library Loan is accommodated in the licensing agreements with libraries.
2. HINARI, Research4Life, INASP, EIFL are programs to provide access to academic institutions or research centers in developing countries with limited budgets. Different programs have different rules about eligibility. For example, Kazakhstan is not considered a developing country and does not qualify for free access but Ethiopia is. You can look these up for eligibility.
3. Many (most?) publishers allow authors to post the accepted manuscript–typically after an embargo period. Manuscripts are very unpopular with readers but if you need the research and don’t have a subscription, there you have it.

What publishers (most) don’t allow are the PDFs that they copyedited, tagged, typeset, created metadata to accompany, deposited metadata for maximum discoverability, hosted, etc., to be posted with no compensation to the publisher for having provided those services.

Your responses sound to me like a reason for Kazakhstan to welcome Sci-Hub with open arms.

As signatories to the Berne Convention, I’m not sure this is legally feasible.

My point is that, were I a Kazakhstanian medical researcher without access to articles that I needed for my work, it would be cold comfort to know that researchers in Ethiopia did have access. I think it takes some kind of disconnect from reality not to see that Kazakhstani researchers in this position will welcome Sci-Hub with a frank and simple joy quite out of keeping with the complex issues we’re discussing here. From their perspective, it’s simply the difference between being able to do their research and not.

Of course, I don’t for that reason expect the Kazahkstan’s government to officially adopt a pro-Sci-Hub position. But you can bet your intestines that its researchers will.

I have no idea why Kazakhstan is not on the developing country list. Perhaps this is a leftover anti-Soviet era mistake. They certainly are a poor country.

Because Kazakhstan is not a lower income country, but an upper middle income country. This is World Bank GDP classification.

And for the record, Chile is high income country ( And guess what – most of our “high” GDP is concentrated in the upper 0.1%, where a few families enjoy. If we were to take away the richest 1% from Chilean GDP, we would be like Congo.

Paywalls ARE a problem.

Glad you replied!

2. The devil could argue that Sci-Hub’s existence demonstrates that (under)development is much more nuanced and granular than the current eligibility rules would suggest. After all, it’s not like disposable income simply appears in one’s pocket if he or she lives in a developed country. But even if we’d stick to the current rules, a significant wealth-gap remains between the poverty line and being able to afford yourself consistent and reliable access to scientific articles. It’s debatable whether a development index takes that particular gap into account.

3. It’s important to note here that (1) what publishers allow isn’t necessarily what authors abide by, and (2) most publishers let it slide when authors don’t abide by their rules. Of course, it wouldn’t make business sense for publishers to sue individual authors for incidental infringement of their copyright. However, it’s a different matter when they’re able to pin it all on one bad guy who managed to refine and simplify the usual process of scholarly sharing.

It’s also misleading to state that there’s “no compensation to the publisher” for content available via Sci-Hub. Sci-Hub wholly depends on publishers being compensated for their service through institutional subscriptions, and due to that most of those clients are probably well aware that Sci-Hub is nowhere near being reliable enough for a serious self-respecting research institution to depend on.

Should the number of subscriptions decrease as some may fear, efficiency of Sci-Hub will soon follow suit. To conclude, I feel like Sci-Hub’s a petty disruption unlikely to have significant impact on scholarly publishing. But it’s not likely going to disappear either, though, as piracy rarely does.

In my opinion, publishers should thank this kind of sites (sci-hub, etc) because they are a kind of advertisement for the publisher and its journals. Authors will cite papers they download, and this will turn to benefits for the publisher itself.

It is true that once upon a time all science journal publishers did a lot of hard work, shifting a typescript into hot metal and a printed page, with real production editing and proofing along the way. That has changed. Production is automated, from submitted files to PDF and apart from a few cases, copy editing and proof reading are minimal to non-existent. The latter point is clearly evidenced on reading any article, the simplest of language errors that would have merited copy editing in the past abound.

While production costs have fallen through the floor (cost of staff >> software), subscription costs have for over two decades risen well above inflation, often in double digits per year. Profits for all science publishers are stunning, only matched by a few individual companies in a couple of other sectors. So the input of publishers has shrunk, but there has been no commensurate reduction in the cost of the output. As pointed out above, you have no chance of retaining copyright, unless you fork out another barrowful of cash to gain this privilege. So anger is understandable and justified. While there may be mechanisms for access by scientists in countries with fewer resources, their experience of these mechanisms is not good, to the extent that they don’t access – I know from experience, when I travel, I am often asked for PDFs of my entire set of publications. These go into a local research group repository. Plus, a large proportion of the major users of Sci-Hub are located in the world’s richest country. So access is clearly a problem.

The arguments though are I think misplaced. Science publishing is changing and the large publishing houses likely to disappear in the next few decades. Perhaps we should view Sci-Hub as an archiving service. After all who will look after the back catalogues of publishing houses that go to the wall?

Now that I have reflected on this diverse conversation today, which I very much appreciate, I have to add something. Scholarly communication is an ecosystem with lots of dependencies. Some of you make it sound like publishers just pulled up in a truck and demanded all of the science papers and then refused to share them without payment. This is not true at all. The system we have today is one that has evolved and developed over decades and decades. Universities, researchers, and publishers are dependent on each other. Whether it’s the perfect system, a fair system, a broken system–I can’t say. I don’t think there is consensus on that. Things are changing for sure and maybe not as fast as some people like and maybe faster than others can adapt to. All that said, I truly don’t think going rogue and advocating piracy is the answer. Change requires compromise–on all sides. Serious people are discussing and are compromising.

For those who have commented so far, I thank you for participating. I hope the conversation continues.

“The system we have today is one that has evolved and developed over decades and decades.”

That is true, of course. And it got the way it did by a series of incremental improvements, all of which made sense at the time, in the environment than then prevailed.

But here I speak with some authority, being a evolutionary biologist. An organism that evolves in one environment is not necessarily well adapted to function in a different environment. The history of life is full of instances where, for example, animals that evolved and thrived in an isolated island because extinct when sea-level changes re-joined that island to the mainland. In a different environment, a previously efficient organism can become useless overnight.

From the perspective of many, that’s what’s happened with scholarly publishing. In a pre-Internet world, subscriptions made sense because they were pretty much literally the only option. In an Internet-enabled world, they make none, because we have much better options for duplication and delivery. We’re not living on an island any more.

I think that most impatience with the legacy publishers derives ultimately from their attempt to perpetuate a business model that makes absolutely no sense in 2015, and to do it by artificially simulating on the Internet the unfortunate conditions of scarcity that the physical universe imposed on us 30 years ago. In effect, the subscription system tries to force us to live in the 1980s. That’s not just objectionable, it’s unrealistic.

Against that backdrop, I think it was always inevitable that something like Sci-Hub would come along. It’s the reaction of someone born into a digital world, seeing that in scholarly publishing we’re still pretending to live in a world of limited print-runs and delivery trucks. If it hadn’t been Alexandra Elbakyan, it would have been someone else. From which it follows that even if Big Publishing does manage to get Sci-Hub shut down permanently, someone else will make another one (likely using the existing code, or something even more capable).

I can’t help but remember the opening words of Paul Wicks (from Patients Like Me) in introducing his contribution to the Scholarly Publishing “Evolution or Revolution?” debate at the Oxford Union three years ago: “I’m here from the Internet to negotiate the terms of your surrender”. As a rhetorical move, it was brilliant, but Paul didn’t have the technology to make it stick. It seems to me that maybe Elbakyan does.

If Sci-Hub forces scholarly publishing as a whole to make the looonnnggg-delayed leap into the 21st Century, that will be a good outcome for everyone except those publishers that have to make do with smaller profit margins when earning from competitive APCs instead of opaque subscriptions. Most people won’t shed too many tears over that.

Where was it established that “earning from competitive APCs” leads to lower profits than “opaque subscriptions”?

Not noted are other peculiarities of the market for academic work. In what other market are the people who benefit from the goods sold not the ones paying for it? Libraries pay for subscriptions, and though some individuals do, this is a diminishing part of the market. In OA fees are very often not paid by the authors but by some institutions, whether funders like Wellcome or authors’ universities. Only if the fees were being paid by the authors themselves would there be the kind of competition you talk about. Why should authors care how much the APCs if they are not paying for them, any more than they are concerned about the high prices of journals that libraries pay for?

You bring up an important part of the way this market functions. You ask what other markets have this feature; first/business class seating on airplanes comes to mind. We all know how that works out.

Thus we’re well aware of the way that this reduces the price elasticity of demand. This is why we’ve argued elsewhere that full subsidy of APCs is not a good idea in the long term:

” We believe that
it would be wiser for funders to support open
access in ways that encourage price competition
among open access publishers. A way to ensure
that authors remain sensitive to price differences
would be for funders to bear only a fraction of
the cost beyond some low threshold (e.g., $500).
In economic terms, we want the price elasticity
of demand to remain high. Of course, univer-
sity administrators should be attentive to quality
as well as price. Subsidizing publication in low-
quality, low-prestige venues is not likely to be in
a university’s best interest.”

( From )

See for example this analysis. Based on figures in the STM Report for 2009, the world’s access to 1.5 million articles cost a total of $8 billion in subscriptions, for an average of $5333 per article. Gold APC vary, but even the publisher-heavy Finch committee estimated an average in the range of £1.5k-£2k, so the midpoint is £1.75k — about $2450, which is less than half the subscription cost.

Also: a competitive market always leads to lower prices than a set of opaquely priced monopolies. That’s basic economics.

Gold OA, which you favor, is a potential gold mine for big commercial publishers, which may eventually be able to reap even larger profits in this new environment. But let’s suppose you are right, and somehow the large publishers are driven out of business. What will take their place? Who will provide the services they do now? Way back in the 1950s universities opted NOT to invest in their own publishing infrastructure, thus allowing the burgeoning commercial publishers to grow and eventually monopolize the business. So, if those publishers disappear, do you have any reason to believe that universities themselves will take on the job of publishing those journals through their presses? My cynical view is that universities, at least in the US, are more interested in investing in major sports teams (big stadiums, high-paid coaches, etc.) than they are in investing in their presses.

Wait — when did I say anything about large publishers being driven out of business? I don’t recall that. All I remember saying that’s relevant here is “those publishers that have to make do with smaller profit margins”. Is that what you’re referring to?

Meanwhile, I’m under illusions that all universities are fundamentally committed to being Good Guys (though some are). But I do know that the specific costs that discouraged universities from investing in publishing infrastructure in the 1950s are no longer relevant. No-one needs to buy printing presses or fleets of trucks. The costs of publishing in the Internet era are 99% human-resources costs — and those, I suspect universities may be as well able to provide as publishers. So, yes, if one or more of the big publishers went under in a post-Sci-Hub world (which I consider far from certain), my best guess is that universities would take up the slack.

Sorry, it was “Ferniglab” who made the comment about large commercial publishers disappearing. By the way “printing presses” were never a cost for universities. The ONLY university press in the US that had a printing plant was Princeton, where I worked for 22 years, and it was not owned by the university; it was a separately chartered organization. I’d also challenge your 99% figure. I was head of a university press for 20 years and I can assure you that there were many other costs besides human resources in our electronic publishing. I don’t share your optimism about what universities (at least in the US) would do if the large commercial publishers disappeared. There is a long history of free riding among American universities, where some 80 universities have shouldered the financial burden of supporting presses that benefited all 3,000+ universities. I see no reason to think that this “free rider” problem would disappear under this scenario. Also, there are many more trustees of universities–especially large public universities–who are more interested in supporting big-time sports than in supporting things like libraries and presses. It has been noted many times on TSK how poorly universities have been supporting their libraries. No one seems to be making that claim about big-time sports. Being from the UK, you evidently are not familiar with what really are the priorities of US universities.

“By the way “printing presses” were never a cost for universities. The ONLY university press in the US that had a printing plant was Princeton”

Actually, the Ohio State University Press started as an office in Ohio State’s offset printing press back in the 1930s. Initially the printing press handled the student newspaper, printing Extension reports, faculty developed textbooks, journals, and the occasional monograph written by OSU faculty (typically in Ag Science but not exclusively). OSU Press slowly evolved into what we might now might consider an independent scholarly press, but didn’t concentrate solely on discipline-based humanities scholarly publishing until the 1950s.

And yes, I have considered the fact that if OSUP had stuck with exclusively publishing their own faculty’s work, in both the sciences and the humanities, we probably wouldn’t be having this conversation.

The more you know…?

As long as funders and universities care about peer review managed by journals, journal branding and prestige of published faculty and grantees, identifying “quality” of work by a journal’s IF for promotion and tenure, the progress you think should happen is dead in the water. The evolution is currently stuck at “we want journals to do all the things we still value” but, “we want it for free.” It doesn’t work that way. There is no one size fits all. There is no free ride. We need to keep talking about the needs of the community and the compromise.

As long as funders and universities care about […] journal branding and prestige of published faculty and grantees, identifying “quality” of work by a journal’s IF for promotion and tenure, the progress you think should happen is dead in the water.

Yes, you have diagnosed the sick state of academia with admirable precision.

The evolution is currently stuck at “we want journals to do all the things we still value” but, “we want it for free.” It doesn’t work that way.

Is she going to? Is she going to? She is, isn’t she?

There is no one size fits all.

Yay! THERE it is! I couldn’t believe we’d got to 97 comments and no-one had recited the Legacy Publishing Mantra 🙂

There is no free ride. We need to keep talking about the needs of the community and the compromise.

I am still not sure whether I agree. I find Bjorn Brembs’s post Sci-Hub As Necessary, Effective Civil Disobedience very compelling. He argues that 20 years of open-access advocated being polite and reasonable — twenty years of talking about the needs of the community and the compromise — have yielded only tiny, incremental steps with painful slowness, while people keep dying for the lack of information. I can see why some people are tired of that, and pleased to see the emergence of a different approach. I wonder whether Sci-Hub might be the blade that cuts the Gordian Knot.

Wait, the claim here is that the last 20 years have only shown the “polite and reasonable” side of the OA movement? Seriously?

“Wait, the claim here is that the last 20 years have only shown the “polite and reasonable” side of the OA movement? Seriously?”

Compared with Sci-Hub, I mean. Attempts to negotiate a settlement. Not direct action.

I spend a lot of time watching all sides of the arguments. It’s why I pay attention to what you and your friends say on Twitter and read your blogs. I also talk to academics in my area and publishers from societies and commercial operations. Largely, the academics I talk to in my area couldn’t care less about Open Access. They see federal mandates as yet another thing they have to do. We talk to our subscribing libraries all the time and we find their requests and suggestions reasonable. I know that this is the case in non-Engineering fields as well. So yes, there is no “one size fits all.”

Thirty years ago, access was extremely limited. Today, access is unprecedented. Is it perfect, not yet. Authors who are not satisfied with “legacy” publishers have so many options now. I continue to be shocked that “author choice” is only a good thing if the author only chooses what the advocates want. I like the experimentation going on. I like the different kinds of peer review and eagerly wait to see which models will succeed and which will go down in flames. I take these ideas back to my community and they are not ready for it.

Thirty years ago, access was extremely limited. Today, access is unprecedented.

That is certainly true; but I do hope publishers from 30 years ago are not claiming credit for this. It was something that was done to them (by the Internet), not by them. Established publishers have had to be dragged kicking and screaming all the way: I’m sure I don’t need to remind this audience about PRISM, the RWA, the endless talk about “our content”, the bribing of congressmen, and so on.

Is it perfect, not yet.

I hope we are in agreement that this is something of an understatement! Estimates vary, but I usually see numbers between 20-40% of papers available as open access. I would describe 99% as “not perfect” and 40% as “shameful”.

Authors who are not satisfied with “legacy” publishers have so many options now. I continue to be shocked that “author choice” is only a good thing if the author only chooses what the advocates want.

That’s only shocking if you assume a priori that author choice is what advocates care about. For the most part, it’s not. I want authors, and publishers and universities, to do the right thing, which is get the world’s store of knowledge into the hands the world — the whole of the knowledge, not 40%; the whole world, not an elite. Sometimes author choice is a good tactic towards attaining this goal; sometimes it’s not. But it was never the goal.

That’s only shocking if you assume a priori that author choice is what advocates care about. For the most part, it’s not. I want authors, and publishers and universities, to do the right thing

Perhaps it’s also not shocking to see people responding negatively to having choice taken away from them, and being pushed into doing what one particular person or group has declared to be “the right thing”.

No, maybe not. As I’ve said, that’s not really my main concern. A researcher who receives a grant to discover something about malaria, who writes up the results, and who wants to keep those results paywalled away from those who need to use those results, isn’t really an object of sympathy to me.

As I often say, there is plenty of blame to go around in scholarly publishing. I do hope do-one ever infers from my criticism of publishers that I think researchers are blameless. Nothing could be further from the truth.

As you know Mike, anyone with a serious interest in an article merely has to ask for a copy. The corresponding author’s email address is always listed. You are therefore presenting a sham argument, as many advocates do. There is no such case. This is neither what subscriptions are about, nor what they do.

You assertion is wildly at odds with the experiences of researchers around the world.

How many of the millions of researchers have you talked to, Mike? Are you actually claiming that most requests for copies are not granted? That is quite a claim. – I’m having trouble finding a discussion of how hard/easy it is for various groups to ask authors for preprints.

Let’s do a test, Mike. Name a journal and I will request a random set of articles and see what response I get. Your call.

So, what are you claiming, Mike–that email addresses are not provided, that researchers don’t respond to requests, or what? Be specific. The URL you provide takes us to a page that says many people who need access to research results are denied access. Access to journals based on research, yes. But do these people need journal articles? Why would the reports the researchers are required to submit to the government not be enough? These are what need to be made available, not the scholarly articles that come later.

Ah yes, the old “Why would the reports the researchers are required to submit to the government not be enough?” gambit. These mythical reports — I have never seen one. Never. I have certainly never written one. I am sceptical of their existence.

Requesting reprints is annoying and time-consuming at best (something tells me you don’t have to read many papers requested from authors…), but it becomes especially frustrating when the author is in heaven (or hell or has maybe just left academia). The emails tend to bounce. I’d be interested to hear your solution for this case, David Wojick.

More generally, the papers that are hard to find are exactly the older ones – sometimes the journals no longer even exist or have gone out of fashion (so institutions no longer subscribe), authors have moved many times, the “archives” are strictly paywalled and there are fewer copies knocking around the web. A case in point in my field is the archives of the Journal of Neurophysiology, which require a separate subscription not paid by my institution. Not even sci-hub seems to work for these (I would happily have used the site in this case).

Boris, I suggest you try to get a grant to dig up these old papers. Outlier cases like yours do not justify overthrowing copyright law. So far as I can tell the system is working well without that. If you do resort to piracy I would not put it on my CV or grant application.

To David, every example of the present system not working is a bizarre outlier, which shouldn’t be taken into account. No true Scotsman would ever need access to a paper they can’t get.

@ David Wojick. As a working researcher, I gave some concrete examples of the system NOT working for me and which can waste a lot of my time. I’m not saying that publishers don’t have the legal ability to gouge people via a monopoly on information or that you should care. But if that’s your position, be honest – don’t pretend to care. And don’t tell me it works for me – I just told you it didn’t. “getting a grant” to access old papers – seriously??

The only reason to keep journals today is for their archives and because the “best” journals sell scientific glory, but that won’t last. The value added I’ve seen alluded to above is strictly minimal for us researchers, however much publishers like to feel that they are essential.

I think the present system works quite well, Mike. The fact that everyone cannot get everything they desire is not a system failure. Of course if you can find a way to make everything free to everyone, with no one paying anything, I would be all for it. In the meantime, the fact that the publishing system is subject to economic reality is not a failure on its part.

I think the present system works quite well, Mike.

And there we have the very fundamental difference in our positions. The present system of scholarly communication is a disastrous kludge of 18th Century elitism, 19th Century practices and 20th Century technology. It’s rubbish.

Honestly, even if the most dire predictions about Sci-Hub came true, and it brought the entire scholaly communication system down in flames around it, I can’t believe that we wouldn’t build something much better in its place.

The fact that everyone cannot get everything they desire is not a system failure.

On the contrary — a system failure is exactly what it is.

This is indeed our basic difference, Mike. You have a grand vision and you think that the fact that the world does not work that way is a failure on the world’s part. I simply disagree. I am all for improving the system but claims that it has somehow failed, despite doing what it does well, just leave me cold.

On the point of academics not caring about open access and being annoyed by mandates: researchers are fully insulated from the costs of obtaining journals, which is why these costs can rise much more than inflation and it does not matter to them. It is similar to the reasons why health care expenditures rise at great rates, at least in the US. If researchers actually cost-shared for journals as part of their grant awards, we would have a very different picture today. The point is not that publishers are “evil,” but that there is a systemic dysfunction in how the scholarly literature is paid for. Publishers are doing exactly what they should do to maximize profits. I think the way out of the Gordion Knot is for the content to be free, with services that mine/analyze/make useful that content licensed and sold by publishers. As Mike notes, even if SciHub folds someone else will come along to do the same thing. The days are numbered for protected access to articles.

It has been claimed that it is impossible to get major publishers such as Elsevier to change their requirement for outright assignment of copyright to them by the author. In my experience that is not true. For many years I have submitted papers to traditional large journals, and if and when the paper was accepted, I have refused to sign its standard assignment form. After some minor to and fro of email correspondence, lo and behold the publisher provides me with a PDF of a standard licence agreement instead. The big publishers have them pre-prepared, but don’t offer them initially for obvious reasons. It is a case of calling their bluff: “OK, if you won’t send me a licence agreement, I will take the paper elsewhere. And by the way, thanks for all the helpful referee comments which improved my paper.”

Extraordinary. I just published a paper. During the process I got at least 5 emails with the suggestion to keep the copyright. They’d prefer cash. 2-3 of such would keep SciHub running.

I don’t need the copyright. I am not going to republish the article without any change. What I would like to have the same option from so called OA. They do not even suggest a possibility to pay for the publication with the copyright. Just cash and nothing else.

This article states that PDFs are “expensive to create”. Clarification on that point please.

As Walt Crawford just tweeted (and I fully agree with him):-

Also: “expensive to create”? Damn, that “save as PDF” option in Word, LibreOffice, etc. must be hard to find.

The reference is to the editing that goes into the final version, not to the format. In publishing circles, the term “PDF” rarely refers to the format alone.

But doesn’t the publisher have to do that editing anyway for the journal? How will not providing PDFs save significant money once that is done anyway?

“Expensive” may not have been the correct word. I don’t actually know how much my organization would save if we told our vendors that we only want one format deliverable as opposed to 2-3. I am assuming there is a savings. We don’t hit “save as PDF” from a word file. That’s actually not how journal production works. From the XML content, there are templates in typesetting programs to render the PDF version. We also add tagging within the PDF so that call outs link, URLs link, reference link, author contact emails link, etc. Because there is an author proofing stage that happens with the PDF version, corrections need to be made to both version of the work. This adds a further complication of ensuring that changes are actually made to both versions. That should be a no brainer but you would be surprised. So the proofing stakes are a bit higher and completely labor intensive.

Now, if a journal is printed, there is no escaping the PDF. You would be surprised about the continuing demand for print.

So all told, there is an expense, but I don’t know how much.

I’ve been staying out of this convo because I have neither knowledge nor opinion on Sci-Hub (well, I do have the latter: I tend to respect laws pretty strictly) but in this case, since I was dragged into it: It sounds as though the sophisticated typesetting programs don’t do what that one-click-in-Word would do–that is, retain hyperlinks and internal links as workable PDF links. That seems odd. I’ll still suggest that there should be *essentially no expense* involved in creating a PDF within any proper document production system, unless it involves adding DRM and watermarks.

It’s more than pressing a button. A PDF should be typeset correctly so that it is a pleasure to read, with the rules of typography, figure placement etc observed. An important part of the process is, as Angela has mentioned, the XML that any serious publisher should keep so that the content can be converted to any other format in future.

I am a “typesetter”, so my company does this back end work. A couple of points:

• Prices for typesetting or pagination have been coming down steadily over the last 20 years, due to competition, especially in India, and due to constant pressure from publishers. So even the full cost of copy editing, relabelling graphics, creating XML, multiple PDFs, dealing with author proofs, etc, are negligible in the grand scheme of things.

• What is sad is that the XML that should be easily data-mineable and machine readable, is seldom released by publishers, even when one pays for the article. I think the reasons are (1) publishers want to keep their options open for reselling other formats, and (2) most XML archived by publishers are so error-ridden that they are too embarrassed to release them. 😉

Is anyone arguing for not doing this? I think publishers will still need to do this labor even if they stop providing PDF files; I don’t think this addresses anything being discussed here.

Igor Golney, in response to my point that one should retain © in articles written and not assign tit publishers, said:

“Extraordinary. I just published a paper. During the process I got at least 5 emails with the suggestion to keep the copyright. They’d prefer cash. 2-3 of such would keep SciHub running.

I don’t need the copyright. I am not going to republish the article without any change. What I would like to have the same option from so called OA. They do not even suggest a possibility to pay for the publication with the copyright. Just cash and nothing else.”

Sorry, but I simply do not understand what Mr Golney is talking about. I was referring to my experience with mainstream subscription publishers; he says he received suggestions to “keep the copyright” and “they’d prefer cash”. Who are “they” here? Mainstream subscription publishers, dodgy “predatory” OA publishers?

He also says he does not need the copyright. That is really short-sighted. By retaining copyright he makes sure the publisher cannot charge ridiculous amounts for access. Simple.

The information in your Medium piece is mostly wrong. The cost of an article has dropped over the years. The price of journals has risen because the amount of content has grown enormously.

Comments are closed.