Hacker News new | comments | ask | show | jobs | submit login
Thousands of scientists run up against Elsevier’s paywall (nature.com)
281 points by pseudolus 13 days ago | hide | past | web | favorite | 131 comments

All this is an excellent ad for sci-hub, which avoids most of the serious drawbacks of publishers like Elsevier. It was interesting how that was relegated to a veiled comment at the end, "or finding access in other channels". But basically if the mainstream publishers can't meet the need, we do need other channels, and right now sci-hub is the only one that actually works at scale.

Sci-Hub's user flow is so ridiculously friendly, that I'd often use it even when I have credentials to some obscure authorization network. This is true even for the free SSRN which started requiring logging in around the time Elsevier bought them.

Gabe Newell of Valve asserted something very similar:

> "We think there is a fundamental misconception about piracy. Piracy is almost always a service problem and not a pricing problem," he said. "If a pirate offers a product anywhere in the world, 24 x 7, purchasable from the convenience of your personal computer, and the legal provider says the product is region-locked, will come to your country 3 months after the US release, and can only be purchased at a brick and mortar store, then the pirate's service is more valuable."

> "We think there is a fundamental misconception about piracy. Piracy is almost always a service problem and not a pricing problem," he said.

I don't know. Paying $25 or $50 for an article, just to see if it contains something useful is a lot ... And with the realization that the research was funded by tax dollars, that price becomes absurd.

Yeah, in the Steam age the demos have been mostly killed. I guess "Let's Plays" have replaced them? Personally, I still value getting my hands on the beast myself.

Steam will at least automatically refund a purchase within 14 days if the play time is less than 2 hours.

As far as I know you can't get a refund if you don't like an article, but I may be mistaken. If you had that option, paying for a journal article might be slightly more palatable.

The quickest legal way to get a copy of a paper that isn't already accessible as a pre-print is to email the author(s). Usually they don't mind. Hint: email the student, not the prof and you're more likely to get a reply. You can also do this via ResearchGate.

The following hack makes it even more friendly, removing the annoying sci-hub sidebar so you can view it purely as a PDF.


While I agree that access is a problem, I don't know how much I believe Gaben's comment anymore. Sites like Kinguin and G2A are doing very healthy business by selling grey market keys for mild to significant discounts to Steam. Their user experience sucks (like actively unethically user hostile bad). Search is horrible. Pricing is nontransparent, aggressive spammy upselling is rampant. The checkout system makes it feel like your credit card information is being posted directly to the dark web. You have to be careful to buy a key that matches the region where you live. Literally the only thing they have going for them is price.

I'd guess those sites are two orders of magnitude smaller than Steam, but a quick search reveals G2A has 700 employees and 16 million customers, so they can't be doing too bad. If access truly was the problem than these sites wouldn't exist.


I use this Library Genesis search page as portal to Scihub's papers, it's amazing being able to get just about any paper ever, even totally obscure ones from 70 or 100 years ago. I do check Google Scholar first - a lot of recent papers are available online from academic sites, and it's a great way of discovering author home pages.

I've always said that the problem is not framed correctly. Who gives money to the greedy publisher, for basically nothing? Blame them first and the problem will go away.

EDIT: blame the academic management who gives the money to the publishers. This way the publishers will go away to greener pastures.

Blame the academic administrators who demand publications in top tier journals - the same ones who charge a ton for access.

Scientists can publish in whatever journal they please. The problem is that if they publish in most open sources journals (there are exceptions), they aren't counted as top tier and it hurts their career.

Many of the open source journals are predatory. And they will publish anything, even complete junk.

The problem isn't academic administers demanding publication in recognized journals. Its that they charge steep prices in the first place.

Inside the Fake Science Factory is an amazing 2hr talk talk from DEFCON 2018 by 2 German journalists and a British science guy about an investigation into the world of predatory fake journals, conferences, websites. 400,000 scientists involved. Unbelievably huge scam. They wrote bogus papers and presented one at a fake conference. It's a funny talk, if depressing. A great investigation. Maybe the worst is that these fake scientific credentials are then used to sell useless products to dying people.


I think you just created a good argument about the value add of journals! Everyone says "they do nothing!", but what they have, that other people want is a reputation for high impact, important publications. If you get your paper published in Nature, your work will be very highly regarded.

The argument about how much money it's worth to pay for that is another discussion entirely.

>> high impact

Careful! Even the data on which journals are "high impact" is now proprietary.[0]

No, of course it ought not to be this way.

[0] https://www.scirp.org/journal/Journalcitationdetails.aspx?Jo...

Open source means that the source code is available, not that you can get the product for free.

> Open source means that the source code is available, not that you can get the product for free.

But open source means that the work can be copied freely - which leads to marginal costs of near 0 €$.

Open source implies does-not-cost-anything, but does-not-cost-anything does not imply open source. Open access journals would only be open source if they, you know, made available source code (or source something).

Open source doesn't imply "does-not-cost-anything" at all considering you can still sell open source applications. And yeah, open access is a better word.

As wolfgke said, open source means that the work can be copied freely - which leads to marginal costs of near 0 €$.

Well... It is pretty often framed as blaming the publishers?

Yes, most of the time. Publishers should be blamed because they ask money for a service which is no longer needed. But who gives them the money, nevertheless? Academic management. Researchers have very little to say about this, except if you look at their real behavior (arXiv, sci-hub, etc). However, in management land everything is designed to support the publishers.

> who gives them the money, nevertheless?

When it comes to subscriptions: academic libraries, under pressure of academics. (And in this case, a group of German libraries finally refused to do so.)

When it comes to disproportionally high publishing fees, however, academics do have something to say, because they can often choose to publish somewhere with lower fees. However, that's not really an option as long as their career progression is determined by which journals they publish in. Thus, I think part of the cause there is those who use journal names as a criterion when judging e.g. grant applicants.

> When it comes to subscriptions: academic libraries, under pressure of academics.

Where are these researchers who pressure the academic libraries to buy subscriptions? I never saw them, except if they are in the management.

If we were to speak in market terms, well, publishers are private entities driven by the market forces, right? So in market terms the clients (researchers) manifest a strong preference for other products than those offered by the publishers. Why do they still exist? Does not make any sense, except if we recognize also that the market is perturbed by the fact that somebody still buys their product and forces the clients to consume and to produce it.

> Researchers have very little to say about this, except if you look at their real behavior (arXiv, sci-hub, etc).

If powerful researchers cared, they wouldn’t review for journals they don’t like. Obviously the benefit from doing the reviews is bigger than their preference of Open Access.

> Obviously the benefit from doing the reviews is bigger than their preference of Open Access.

Researchers don't usually benefit from doing peer review: they don't get paid for it, and it eats time they could spend doing research. It's more of a sense of duty they have that they're "supposed" to do those reviews, and I can't really fault them for that.

(Though I would very much encourage them to only review for fully Open Access journals, and respond to review requests with invitations to submit the work to those journals.)

The benefit to reviewing is that it's frowned upon to publish in a conference/journal if you don't review for them. So academics are not just motivated by a "sense of duty": if they don't accept to review then it can become harder for them to publish (which has significant career benefits).

So if you don't review for closed-access venues (which I try to do), you are pressured into not publishing there: and not publishing in closed-access venues is difficult as most papers are collaborations, and usually your co-authors do not share your convictions.

At least in my experience, I don't see any pressure to not submit if you don't review (I'm in ML though, so could be different in other fields).

I think why people review is: -You should be reading more papers anyway and we like doing it, so why not take it as an obligation? -I assume that eventually it helps one to get selected to be an area chair or something along those lines.

Yes, I guess that's true, too. But even researchers who do boycott some journals sometimes indicate that they still review for those journals for the above reasons.

> most papers are collaborations, and usually your co-authors do not share your convictions.

Can't you just publish in two journals then?

Most journals have as a condition of publication that the article is not submitted or published anywhere else, if that's what you're thinking about.

Researches benefit from doing peer review so that quality research is maintained. Its part of their job. I don't get "paid" to put in expense reports for trips on the job, but I have to do them because its a job responsibility.

Right, as I said in response to a sibling comment: "a sense of duty, it's considered part of their job." The sense of duty part is that they could theoretically get away with it if they didn't do it.

If they don’t benefit in some way why do they review?

Edit: - sorry, I missed the sense of duty. Thanks for pointing it out below. - added don’t

Again: a sense of duty, it's considered part of their job. They can certainly choose not to do it, or not to do it for paywalled journals. However, when I suggest that, I often get the reply that that's not fair to those who submit to those journals - often early-career researchers whose career could be greatly helped by publishing there.

And again, I can't fault them for that.

(I assume you meant to say "don't benefit"?)

I don't buy the sense of duty argument. It's good for the career.

Until the appearance of the notion of open peer-review, the reviews were not public and thrown away by the editor after the publishing decision.

Concerning the sense of duty, nobody stops anybody to review a work and post it somewhere publicly accessible. However, that's rare to see.

Yes, many of them follow the path of least resistance, but not all of them. Even the usual researcher from the middle of the herd votes against publishers with the reading behaviors.

You're assuming the academic management has nothing to lose in canceling the deals with the vultures.

I suspect they do have much to lose.

It will be hard to recruit faculty who get grant money to fund the school if Elsevier has banned your campus IP addresses.

I honestly think that the Oldenburg model (yes, he had good reasons that are no longer relevant) is going the way of the dodo. These companies are fighting a losing battle. They've already lost to pre-print on arXiv.

The future of making money from journals could potentially lie with curation: "these are the papers you need to read this month" with links to open-access papers. Indexing and search that exploits that curation also comes to mind: "these are the papers you should read while researching your problem."

The biggest problem that open-access is going to cause is volume. Although, that been said, that's an existing problem (though far less pronounced) in the Oldenburg model due to bad science.

>The future of making money from journals could potentially lie with curation

Scientists (and lay people) already have access to excellent tools for this kind of thing. Pubmed provides functionality for performing automated and regular searches of their basically complete index of biological literature. https://ncbiinsights.ncbi.nlm.nih.gov/2013/11/14/setting-up-...

Unfortunately, they've started blocking it (via DNS) in some countries, and large amounts of people don't know how to access it now

A while ago I made https://whereisscihub.now.sh, which can be of some help. It also lists IP addresses at the moment, so presumably those should still work.

Thanks for this.

The easiest option is via the Telegram bot. You message it the name of a paper and it sends you a PDF back! And it's unblockable.

If you can afford it everyone should send a few bucks to the bitcoin address donation jar on the Sci-Hub site (1K4t2vSBSS2xFjZ6PofYnbgZewjeqbG1TM - but don't take my word for it) so that Elbakyan can keep developing cool stuff like that.

I try to keep this post up to date with the latest scihub links (and it has some other links to similar sites): https://citationsy.com/blog/download-research-papers-scienti...


And the fact that requests for Elsevier stuff have "not increased correspondingly" implies that using Sci-Hub has become routine and unremarkable.

> sci-hub is the only one that actually works at scale.

Well, Unpaywall works pretty well, and is actually legal: https://unpaywall.org

Unpaywall is just a database of free(libre) scholarly articles. These articles can almost always already accessed at the journal website using the normal way of finding them. doi.org/$DOI or scholar.google.com is another. The problem is not with these articles but the larger number of ones that are behind paywalls the unpaywall. Unpaywall is not what they do (from the blurb on their homepage and a brief look at the site) and a very deceptive name for their site.

That's only the case for so-called Gold Open Access artiles. However, many other articles are put behind a paywall, but also have a version uploaded elsewhere (e.g. arXiv) by their author ("Green Open Access") - Unpaywall helps you find those versions when you hit a paywall.

Stunningly in NYC there is not a single medical library that offers journal access open to the general public. The only publicly accessible medical library has been described to me as primarily a "historical" library without journal access.

Columbia and NYU's medical libraries are only open to medical school students and faculty. No undergraduates. No alumni. The New York Public Library has surprisingly good online access but is missing a number of important journals and current issues are often delayed by agreement for several months.

If in this nation's largest city, current healthcare knowledge is bulkheaded from public access there is a serious problem. Whatever the publisher's rights these knowledge and information asymmetries must not be allowed to continue.

Aren't these journal articles, like all formal publications, a matter of public record? I don't know where the United States' libraries of record are, but you should be able to go to one of them and ask to see a copy of anything at all published ever in the US.

I guess the Library of Congress is the closest thing. They are open to the public and have a huge collection but it's not everything ever published in the US. I'm not sure if you can walk in and request to see a medical journal.

I wish municipal public libraries would provide access to scientific journals.

I find that...offensive. It's infuriatingly wrong.

Mind expounding on why?

> Agrees, badly.

You don't necessarily have to be at a medical library to have access to the same online journal content that a medical library has. I'd look for a high-quality general-purpose library at a public university that offers on-site access to electronic subscriptions to the public.

NYU and Columbia are both private universities of course. Does NYC even have a large top-tier public university? Hmm.

NYPL gives you access to more online paywalled journal subscriptions than nearly any other public library in the country though, you're lucky for that.

(If you're looking for print archives of journals, medical libraries are at the forefront of getting rid of historical print journal collections, because the usage of them was so tiny. The vast majority of medical researchers simply don't use historical print journal collections anymore, and there are non-trivial costs associated with keeping tons of bound journals getting almost no use...)

I am not arguing with your basic assertion that in America as a whole public access to paywalled academic research is a problem. I agree. (Apparently not just America, as OP is about Germany, where universities are collectively trying to _do_ something about it in a way we aren't really here).

The shift to electronic instead of print content doesn't help -- it's easier to simply give the public access to the physical environment to access stacks of bound journals (if you want to; or are required to give the public _some_ access as a federal depository library, sometimes easier to give them _all_ access), than it is to give them on-site access to electronic subscriptions (which you gotta get your vendors to agree is allowed, and then provide technological support for).

There are ways that the "digital revolution" has actually _hurt_ access to information, ironically.

It is a problem. But some (not all) university libraries, especially public university libraries, are trying. For instance, specifically insisting on public on-site access being included in their licenses from vendors. It's worth looking around and not assuming it's got to be a _medical_ library to get access to the online content. If you can't find it in NYC, it's probably even worse other places (although NYC's lack of major _public_ university probably doesn't help).

And then there's always sci-hub...

I certainly agree in principle that it ALL ought to just be in the commons, and not something only available through the richest universities (whether or not they then "share" it).

</a librarian software engineer>

You are right. NYPL is about the best public library one could hope for and much is available from home login. City College is one option to look at but I don't think they have a medical library. I am an NYU alum and had extensive conversations with the alumni office and library and both confirmed that access to med school library resources was a non-starter. I guess its a big issue for me because people need information to make informed healthcare decisions and so much medical research has a component of public funding. Thanks for your thoughtful comment.

Sometimes 'alumni access' (that you can use from anywhere with alumni credentials) is less than the total online licensed resource access for current students/faculty -- because the university generally needs to pay extra for "alumni access" to particular licensed resources.

But sometimes the total licensed resource access is available for _on-site_ public users, even if not alumni, because that may be included in the licenses.

I know you can't enter the medical library, but have you checked if you can get access to the resources you need on-site at another NYU library, that may allow the public to walk-in and use licensed resources from on-site workstations?

There may be resources licensed only to the medical library you can't get access to there -- but there may not be. Sometimes the particular people you are talking to asking questions don't understand the total licensing landscape of the university, especially if you are talking to people at the alumni office rather than librarians (but even every librarian may not, if you're not talking to someone whose job is centered on online resources; it gets terribly complicated, and libraries haven't always been great at keeping track of it and keeping it sane; the vendors' demands and irrational and enormously expensive contracts don't really help).

Have you tried accessing the publications via the on-campus computers in the normal undergraduate libraries? I don't think you need to be in a medical building to have access to the online medical journals.

I guarantee anyone enrolled or alumni with credentials at Columbia and NYU can log into just about any journal on the school network on campus or with the school proxy anywhere in the world.

Laymen aren't going to be reading from these journals, they are really too technical to be informative. It's like having people without CS knowledge read through your source code and saying "Boom. Open source." By the time you explain someone the necessary conceptual ideas to understand these dense papers, you will be basically giving them a B.S. in biology.

The real crime with these journals is that it can cost thousands to have the privilege of your paper published. That's thousands that could have been spent on more reagents, equipment, or salaries to do even more science. Instead, you pay the toll troll, and you lost another x% of your grant earmarked by the government to do science.

You're stretching the definition of laymen here - often people with plenty of expertise, such as GP's, don't have access to the latest research, or have to make do with the access of interns. For examples, see https://whoneedsaccess.org

> Laymen aren't going to be reading from these journals, they are really too technical to be informative.

You're right. We should lock up all knowledge to only those are who are not laymen. Want to learn CS, go to school. Clearly the only way to learn is to be taught. /s

> Laymen aren't going to be reading from these journals, they are really too technical to be informative.

This is probably the single most offensive thing I've read on this site.

I am a college dropout layperson who got a 0.0 GPA in my last semester enrolled in a liberal arts college who went on to replicate complicated experiments in sports science from these so-called "too technical to be informative" journals. A decade or so later, and now I run a fairly successful (30+ people employed) small business primarily because of the fact that I based my work on early publications I was lucky enough to get access through from students at the local public university.

With Sci-Hub, many terrible steps have been cut out of the process, and Elbakyan deserves Nobel Prize consideration for her work.

Meanwhile, your gatekeeping comments are elitist and only serve to increase inequality in all forms in this country (and the developing world, where academic freedom is truly useful to break paradigms).

Open source software need not be 100% understandable for it to be useful. We do business with as many vendors as possible that open source their code and work not because I am interested in validating their work, but because of the signal it sends that they feel comfortable and open enough to share their core products.

I hope you rethink your positions on these matters, because they're pretty offensive.

Being a layman in more fields than I am an expert, and having a habit of reading journal articles in various fields when the mainstream press reports on their findings, I have to disagree.

Of course, I can't (usually) read a journal article in a field in which I am a layman and understand it well enough to attempt to replicate its findings, critique its methodology, or make a follow-on contribution to the field. That's not the only value one might extract from a journal article.

What I can do is figure out if the mainstream press reporting on the article actually matches what the researchers found. The most common issue I see in mainstream press reporting is over-broad conclusions like "X is a cure for cancer" instead of "X is a 10% more effective treatment for Y type of cancer". Correlation reported as causation when the research did not draw such a conclusion is also annoyingly common.

what about the hundreds of thousands of doctors in their own practices that would like to keep current? I can tell you with the absolute certainty of experience that alumni do not retain such privileges at all schools. And when they do, it can be limited to in person access: not possible when you live hours or more away from your almer mater.

The world isn't made up of just laymen and students. There's also professionals. In some cases employers will pay, or individuals can pay, but journal payment schemes aren't really set up for this.

While Nature publishes articles that poke at Elsevier, they might themselves want to explain to us why they are the largest requestor of waivers to UC's Open Access policy, by almost a factor of 8:


It would seem to me that they're not exactly an unbiased spectator on the matter of publishers and access.

Keep in mind that this article is published by Nature Publishing Group, which is a closed-access publisher just like Elsevier. Hence, they have an incentive to give readers the impression that it's a big deal for a university to cancel their subscription to closed-access articles.

They do so by failing to mention the two obvious ways that people can use to get access to scientific papers without a subscription: (1.) preprints on open repositories and/or author websites, and (2.) Sci-Hub.

Preprints are not an option in every field, they're very common in some fields but almost nonexistent in others.

Sci-Hub is not a legal alternative, that is an issue. There are legal ways to get papers without subscription for a reasonable price (inter-library loans), but they take more time and effort.

> Sci-Hub is not a legal alternative

Depends on where you live. In my country it is perfectly legal for scientific or teaching purposes. And any article with a Belgian author has to be freely available by law max 3-6 months after publication (depending on the field).

It can be argued that reading scientific articles is a basic human right, see art. 27 of the UN declaration: "Everyone has the right freely [..] to share in scientific advancement and its benefits."

The fact that Sci-Hub is not a legal alternative is not a reason not to mention them. The existence of Sci-Hub is very relevant to what the article discusses.

Thing is, there is no need for Sci-Hub if researchers (in the US) just took a few minutes to comply with the existing NIH public access policy.


> To advance science and improve human health, NIH makes the peer-reviewed articles it funds publicly available on PubMed Central. The NIH public access policy requires scientists to submit final peer-reviewed journal manuscripts that arise from NIH funds to PubMed Central immediately upon acceptance for publication.

Why this duty is so often neglected escapes me, since:

(1) it's to our great benefit (in a general greater good sense, yes, but here I'm saying it's to someone's great personal benefit) to make your research articles as easily accessible as possible (think: h-index).

(2) It's not like PubMed / PMC is some obscure portal scientists rarely visit. It's where the majority of us begin our lit search. If you are writing a grant / thesis / dissertation it's practically our homepage. Basically I'm saying it's collectively odd behavior that we use PubMed as our primary portal to access research articles, but not bother to ensure our own research articles are easily accessible through this portal.

https://www.ncbi.nlm.nih.gov/pmc/about/intro/ https://www.ncbi.nlm.nih.gov/pmc/about/public-access/

> Sci-Hub is not a legal alternative

It is just one criterion among the many criteria that you use for your decision process.

It should be a legal requirement that all papers resulting from taxpayer-funded research be open access.

Many European funders, and hopefully soon others as well, are going to mandate exactly that: when you take their funds, you can only publish your work in fully open access journals. See https://www.coalition-s.org/

Agree 100%. It's obviously not possible to dictate what people do with privately funded science, but it's ludicrous that publicly funded research isn't always open-access.

If you get a single cent for your study from tax payers, the tax payers have a right to see the results, with no barriers to access.

I realize it's a barrier, but you can typically contact the researcher directly and get a pre-edited copy as well as supplemental data pretty easily. They're more than happy to share and even schedule calls or in-person meetings to discuss the work.

The share restraint is usually on the edited version journals provide, which frankly, after working on several published papers, isn't much from my experience. The editors don't really prod the theory, methodology or data -- just grammar and formatting (nice but not critical for the science).

This is already true for the NIH in the US, the single largest research funder in the world.

Only after an initial period of exclusivity - I forget whether it's 6 months or 12 months. But even this was an enormous improvement over the status quo.

The NSF, which I think is the second-largest funder of research (or at least basic research) in the US, also requires public access to all publications (after a 12-month embargo, like NIH).


It is notable that these requirements from NSF and NIH have not eliminated many of the complaints about the influence of closed-access journals. Presumably this is because the real issue are the incentives researchers face to publish in such journals regardless of whether they are also publicly available. (Indeed, the vast majority of researchers already have institutional access to closed access journals, and the degree to which non-researchers read research is negligible.)

For some scale: The NIH funds $39B in medical research


The NSF does $7B


Total funding in the US for basic research (which I think includes about $5B from NSF and $15B from NIH) is $86B

http://www.sciencemag.org/news/2017/03/data-check-us-governm... https://www.nsf.gov/statistics/2017/nsf17316/

More of a stretch, but we should also get rid of copyright overall, so anyone who wants to share any such valuable research or other cultural works can just do so.

If you want to place your work in the public domain, you're free to do so. Copyright law isn't preventing this.

Copyright law certainly has major issues, I'll agree, but it does not prevent you from sharing things if you want to, provided you created them or own them.

> If you want to place your work in the public domain, you're free to do so.

Fun fact: in many places, you actually can't. You can sign over copyright to someone else, or you can give anyone permission to do whatever, but many legal systems have no concept of "disclaiming copyright". You have it, whether you like it or not. This is why it is frequently advised against putting "I put this code in the public domain" as your open source licensing term. It has no legal meaning in many places, which means that it's not properly licensed at all (i.e. full copyright applies).

CC0 is a pretty solid solution to this problem, though.

That won't fly in every journal.

"your work"… thing is, all creativity is derivative. The actual result of "your work" is itself limited when the resources to draw upon exclude all sorts of stuff limited by copyright and patent laws.

I do release everything where I have copyright under free/libre/open terms. But that's not enough to cover the concerns I'm referring to.

I'm saying copyright should be eliminated because it's in the public interest to have more open access (both plainly and for derivative use) to all resrouces

For some journals part of the submission process waives your personal copyright.

That's because, when you give something to someone else, it's not yours any more. This isn't rocket science. Same goes with copyright: if you give your creation away to someone else, it's not yours any more, it's theirs. The simple answer is to not give ownership away to someone if you want to put it in the public domain, or otherwise do something with it that this other party wouldn't agree to.

If you give your car away to someone else, why would you be mad that you can't just use it at random times?

Yes. I agree.

The authors have had the solution in front of them from the beginning. They've been free to send out free copies from the time they finish.

But if they want the advantages that come from journals, well, someone is going to need to pay for them.

If you email any author I guarantee they will send you a PDF of the paper.

> If you email any author I guarantee they will send you a PDF of the paper.

So let us create a website that for a given paywalled URL automatically emails the authors for a PDF. ;-)

I'm pretty sure someone actually did that. Google around for it.

Not all authors will respond to email promptly. And many have retired or even died -- there are decades-old papers still behind paywalls.

It's a beautiful sentiment.

Who's going to pay for the layout and editing and filtering? The people at Elsevier do something for their money.

I'm happy with going to all author-circulated preprints, but I'm afraid that all of the archival work of the journals is a pretty good thing.

I have been publishing at Elsevier, and the editing process was a nightmare. The "layout and editing and filtering" is not performed by skilled editors, it is performed on an Amazon Mechanical Turk level. There was no improvement for the reader over what my co-authors and me had submitted, just a lot of back and forth to prevent numerous new errors to be slipped in.

I'd gladly pay a one-time fee for a good editor to bring a paper into a better shape, to cover these costs.

Ehm... The same people that currently do so at Open Access journals, or at institutional repositories?

In many cases, "layout" is a standard LaTeX/Word template done every few years. "Editing" on text-level is done by the authors, based on feedback from reviewers that aren't paid. Filtering is done based on the same reviewer feedback.

Coordination and archiving are valuable, but don't match the prices asked by some.

Why do we need to pay for expensive layout, in an era where nobody reads the print journal any more? A Word document exported to PDF would work just as well as far as actual science is concerned.

The value of a publisher these days is not about the cost of publishing anymore (for obvious reasons, like text editing technology and the Internet).

Today it's almost 100% about the prestige/brand recognition of the journals which they still control.

Journal prestige still matters because: (i) it's a way people can quickly determine the value of a paper (i.e. without reading it); and (ii) "where" you publish has direct impact on researchers careers, funding, etc.

The only way out is for the research community to start using other means of ranking papers and assessing the impact of research in a way that doesn't depend on which journal it got published.

And yet, they keep publishing in these journals?

Yes, yes, I know, publish or perish, prestige, impact factor, yada-yada. I’ve just imagined for a second what it would be like if the rhetoric and activism of left-leaning staff on campus that is talked about so much were directed at fighting unequal access to knowledge. It sounded so good in my head; the rhetorical techniques fit so beautifully.

They don't really have a choice. Your publication record is the most important factor you're judged by, and that determines your chances to become a professor and to continue to earn grant money to do some science.

If you have the option to publish in Nature or Science, you have to take it. Or you will stunt or kill your career.

> If you have the option to publish in Nature or Science, you have to take it.

That’s because it is currently your own decision as an individual. But if the left mobilized their forces to shame those who publish in these journals like they shamed, oh, I don't know, Niall Ferguson for having an "all-male history conference” — oh how the arithmetic will change :-)

As a academic researcher, here is my takeaway: European scientist will be much likely to see my publications under Elsevier and therefore much less likely to cite it. Going forward, I will avoid publishing under Elsevier (even in those journals with a high impact factor) out of fear that the work will not be cited in the future

Sounds like what matters to you most is how many citations you get.

Color me naive, I thought research was about doing cool science, no?

Yes, I am in research to do cool science. I write the papers because I want to do cool science. But then I have to decide where to submit those papers. At this point, the choice has nothing to do with science, so I am careful to pick journals where my papers get the most visibility in the hope that this will increase my chances of remaining funded

Elaborating on the above, everyone gets into research to do cool science. But once you get there, you need to worry about funding, promotion, and other messy details of career maintenance. I've watched many excellent scientists, genuinely brilliant and thoughtful people, spend an inordinate amount of time on this, because they all know if they don't, they won't be able to continue doing science indefinitely.

Given the number of downvotes, I'd say I hit a really sore nerve.

Citation-based scientific success is deeply wrong, and you'all know it. The whole "we don't have a choice but to play that game to get funding" doesn't make it any more right.

It also looks like none of the answer stopped for a second to reflect on how broken a system relying on citations controlled by the elsevier of this world is.

The quality of the science should speak for itself. How much research gets cited should be a consequence of that quality period. The feedback loop is counter-productive, and worse, easy to game or take advantage of.

That is definitely not the only consideration for pretty much every researcher. Being able to keep doing it is another, for example, and in today's world, citations help you do that.

Part of good science is to communicate the results of your research. It is after all the scientific community, not individual people doing "cool science".

The point of publishing is to be read. If you don't think people will read your work in a given venue, it doesn't make sense to publish there.

In order to do cool science you need to get funding. To get paid you need to get your citation metrics up.

To address some comments (some of them dead) suggesting that typesetting etc. costs money: most publishers actually force the author to do this themselves.

I wonder whether this also has the reverse effect: that publishing your work behind a paywall in an Elsevier journal will hurt your impact, because fewer researchers will read it.

In my opinion, it's these incentives that are the problem. The main reason researchers keep going for the same paywalled (or high publishing fee-) journals is because they need to list the stamp of approval of its brand name on their CV. Fewer readers makes that brand name less valuable.

That said, I don't think that will be enough. The world needs to transition to another way of evaluating research that does not depend on where they publish.

(Disclosure: I'm working on such a system [1]. You can help by installing the extension [2]!)

[1] https://plaudit.pub/

[2] https://plaudit.pub/extension/

It is known that paywalled articles are less read, and less cited, than open-access articles. See e.g. https://www.timeshighereducation.com/home/open-access-papers...

Sadly, this is much less important to researchers than the name of the journal that they publish in.

This is because most researchers don't see the paywall with their proxies, it is the institution or company that they work under that deals with these fees.

I understand that the situation is frustrating for scientists and academics, but I really hope that they understand that breaking the Elsevier's chokehold is good for science (and the world) in the long run.

Elsevier publishes about 17-18% of the world’s research. It’s a lot, but they hardly have a chokehold.

That percent changes by field.

Hmm. Given all the author requirements for print-ready copy, page charges, and the like, one wonders why these big publishers get so much material to publish when less costly alternatives exist.

Maybe it's their stellar peer-review processes? Maybe not.

Maybe the brand value of their flagship journals? That could be the explanation.

Here on HN, people are on the lookout for businesses vulnerable to disruption because they rest on their laurels rather than their innovation. Elsevier looks like that kind of business. Opportunity for sci-hub and other innovators!

"most of these article requests are fulfilled within a working day by one of the ten German institutions that do still have a subscription to Elsevier"

... this doesn't seem sustainable

I think it mostly gives the cover of plausible deniability - I don't think many people doubt that many researchers simply fetch the research they need from Sci-Hub, or elect not to read it at all, before contacting their librarian.

This may be an incredibly ignorant question so go easy on me, but does publishing in a given journal give some form or Intellectual Property rights to the publisher? Is there some mandatory exclusivity rule?

That is, what's to stop someone from publishing in a "top" journal to bolster their career, then re-publishing the same thing in an open access journal?

These publishers require that you transfer copyright to them. You're allowed to use your own work as part of your work, such as teaching, and sharing it with fellow researchers. But you're not allowed to upload the published version of the work elsewhere. Preprints are often allowed, such as on arxiv, are usually allowed.

"which publishes more than 400,000 papers each year."

Wow, I had no idea so many research papers were published by Elsevier alone.

Does that mean that most research goes unread? How does that affect researches and science in general?

I don't quite get why the German gov doesn't just roll it's own solution?

Presumably the community would like that. Throw some money at it & start a government backed free one.

The hosting is not the hard part. Arxiv.org and SciHub do that on a shoestring budget. The hard part is changing the mindset of the scientists. Reputation is a very strong motivation factor for scientists to publish in prestigious journals. And journals (prestigious or not) like to charge a lot of money just because they can.

Fortunately things are changing. Most people in the machine learning community posts their papers directly on arxiv.org.

Rolling a publication is not expensive and relatively easy. Building a brand and trust in the publication is extremely difficult, takes a long time, and a lot of marketing dollars. If it was easy, it would have been done long ago.

Bad news for Elsevier. Do Academics really want to publish papers to journals where nobody can cite you?

Organizations like Elsevier as well as all those other rights management companies are one of the worst thorns of capitalism and destroy one of the greatest ideas what the internet should/could be one day.

What about free education and knowledge?

What about culture?

Those things shouldn't be exclusive to people with enough money. We talk about common goods here and especially education and knowledge have to be free as they potentially facilitate a positive development of mankind.

It's unbearable that still parasites like these are gaining wealth without adding real value to their customers as well as humanity itself.

Uh oh, calling Elsevier's bluff.


Applications are open for YC Summer 2019

Guidelines | FAQ | Support | API | Security | Lists | Bookmarklet | Legal | Apply to YC | Contact