From that information, I would actively want to check SciHub out if I had never heard of it. Academia + notoriety sounds novel and useful.

Part of why SciHub has succeeded is just how easy it is. Copy/pasting an arbitrary DOI from any source will always beat most journals arcane login systems (especially on mobile) and their divine mission of obscuring "Download PDF" buttons.

SciHub is amazing. I'm not in academia or industry, and I was working on trying to find some eutectics of simple ionic halide mixtures with some software that was missing some of the activation energy coefficients for some of the ion interactions. Must have looked at at least 30+ papers, which if university-publishing house cartel have their way, would have cost me thousands of dollars just to look at. Between sci-hub and all the open source code from labs and individuals around the world, it's just breathtaking what can be explored now outside the confines of academia.

Curious what the application was?

Mostly exploring molten salt mixtures for thermal energy storage with a larger working temperature range and a reasonable melting point compared to what's used typically.

Do you work at a molten salt energy storage company? What's the industry landscape there like? Like still primarily R&D, or is it mostly about production now.

No I don't, just a renewable energy hobbyist since about a decade ago when I interned at P&G chemicals researching transesterifcation/industrial processes for biodiesel production. I mostly want to experiment my own micro CSP system, but never really had to tools to explore salt combinations theoretically in a way that was fun for me.

On the other hand I have seen a guide from a Swedish university library with a "the dark side" section recommending Scihub, #icanhazpdf and /r/scholar. The contract between Elsevier and pretty much all Swedish universities was canceled a year ago and they seem to be fine.

Oh please. Everybody I know personally and everybody I know of through these people use scihub. It's such an incredibly useful tool that it's now ubiquitous throughout academia. Of course universities aren't 'officially' advising its use, but informally, it's pretty much accepted.

Warning people not to use sci-hub, mentioning it by name as something that you shouldn't use, is a plausibly deniable way to spread the word about it.

Oh, I can assure you that this particular university is dead serious about SciHub use being undesirable. The course I mentioned dedicates significant space to respecting copyright, not only as something expected of researchers, but as a moral duty.

Also, inter-library loan is a source of income for the university library, that is, if the library does not have access to a particular journal in the databases it subscribes to, then it will obtain the journal article from elsewhere but at a hefty fee, so perhaps this is another reason they would prefer students avoid SciHub.

Enlightened morality often means being mindful of tradeoffs.

Not doing an exhaustive search of academic literature because it is prohibitively expensive or time-consuming is a clear negative -- doing so may result in lower-quality research.

I personally see many good reasons to distinguish general intellectual property theft from using SciHub for academic research.

Which university is this? If anything involving copyright is "moral," it's breaking it.

Or, you know, it's the law corporations were able to push into existence through money. Law isn't necessarily a reflection of our moral code. It can be. Parts of it are (Grundgesetz or the Constitution, sure), but plenty of laws aren't decided by morality and aren't the arbiters of it either and even if they reflect the morality of some people, it isn't morally justified. Or people wouldn't have ever fought for black rights (Jim Crow laws say hello).

You are putting yourself in a logical trap here, associating copyright law with murder...

Which statement is true...

"By locking knowledge behind pathways you are both directly and indirectly causes the death of others by limiting information to people that could use it to save others"

"By not respecting copyright you are breaking the institutions that create scientific knowledge and making it riskier for said institutions to invest and create new things which can be used to save others"

The law is always at moral odds with itself. It must try to achieve both goals which is an impossibility.

> "Who defines "moral"?"

You do. And I do. You define it for you, and I define it for me. You and I are both going to define murder as immoral, and not commit murder. Should one of us define it otherwise, the other will in practice likely be powerless to prevent the outcome.

This is the way the world works. At the end of the day, there is no universal ruleset that all humans will follow like automata; we each get to make our own path. The best we can do is attempt to persuade others to change their moral framework and to gang up with people who agree with us to create systems that inhibit those who disagree with us.

There is no sense moaning about this state of affairs because this is the way it's always been and the way it always will be.

> "By your logic it might become moral to murder"

Something else to keep in mind, is that morality is likely rooted in some limited extent in biology. The basis for this assertion is that across time and cultures, one form or another of law against murder has always been popular in any human civilization you can find evidence for. Even when societies carved out formal exceptions to this rule, such as the legal witch burnings in recent European history, the human sacrifices in more ancient European history, or the human sacrifices seen in Aztec civilization (that had no plausible meaningful cultural exchange with Eurasia for many thousands of years), there still exist laws in some form against murder in general. My point with this likely controversial digression is that it's unlikely that anybody who constructs an argument for the morality of academic copyright infringement will accidentally construct a persuasive argument for casual murder. Even if you use formal logic to find a link between their perspective on copyright to casual murder, that formal argument will be irrationally rejected by nearly everybody, who has an innate predisposition towards rejecting a system of legal casual murder.

tl;dr: reductio ad fortuita occidendum is not a persuasive argument.

Yeah it's very much a don't ask, don't tell sort of thing. Check out some of the location images for scihub users and they very much paint a different story.

Parallel construction it is, then.

Cite the official published source.

Citations should always be in some standard form, including author, year, paper title, journal. But for a web link, just put the DOI. A reader can either click it or paste it into scihub, as they prefer.

I wonder how this dichotomy of "use Wikipedia and SciHub, but you don't cite them directly" affects the ethics of graduates.

People cite Wikipedia all the time in peer-reviewed journal papers. Many e.g. high school teachers recommend that their students not cite (any) encyclopedias directly, because they want the students to learn how to read and cite primary sources.

Reading a secondary source for an overview then diving into primary sources is a perfectly fine and ethical way to do research. I can’t see how schools and teachers advising students to read secondary sources for context but then follow their references to primary sources to read for the details would cause any kind of ethical compromise.

There is no reason to cite SciHub (just like there is no reason to cite Academia.edu, ResearchGate, some professor’s personal webpage with a hosted preprint, Google Scholar, etc.) whether or not someone uses it.

I think you have a point and a lot of replies are missing it.

Tacit acceptance of Sci-Hub is tacit acceptance that some laws don't matter and everyone knows it. That's certainly an ethical stance with implications. It's not necessarily a bad stance (D&D would call it "chaotic good") and it is I think my own stance, but once you decide that laws are secondary to your personal and professional needs if you don't get caught, you're likely to use the same argument in other respects. Citing the original paper instead of the Sci-Hub URL only increases the extent to which you're pretending to follow the law.

Wikipedia has primary sources linked, yes, but it has its own biases. It tries not to, but it still has them to some extent, perhaps just in terms of what sections have been written and what haven't based on editors' interests. If you use Wikipedia to find primary sources, you're filtering your research through those biases. That has direct implications, and indirect implications for ethics in that you've decided this filtering is acceptable to you and also that it's acceptable (and perhaps required) not to acknowledge it. (And yes, of course there are other biases in terms of what university libraries stock, what previous academics have written about, what funding agencies are interested in, etc. I'd say that how much you pay attention to these potential biases and whether you see them as significant is also a question of ethics.)

Because some laws just don't matter. This is a good thing to realize for people. Laws aren't automatically right and moral just because they're laws. There are plenty of reasons why a law can exist, and morality usually isn't the reason.

> once you decide that laws are secondary to your personal and professional needs if you don't get caught, you're likely to use the same argument in other respect

I don't think it's that clear cut. If you drive above the speed limit, or smoke weed in a state where it's not allowed, or pirate some DVDs, you're not automatically going to progress to more serious crimes. People seem to be perfectly capable of "ahh let's ignore this stupid law" and not start sliding down some slippery slope.

How many grad students for generations have been smoking dope despite the laws against it? Using sci-hub despite the law is not a significant departure from the status quo of grad students ignoring dumb laws when it suits them. Sci-hub does not present a uniquely deleterious effect on the ethics of grad students.

Also, even if sci-hub were legal, you still wouldn't cite the sci-hub url. That's not how things are cited.

> but once you decide that laws are secondary to your personal and professional needs if you don't get caught, you're likely to use the same argument in other respects.

I think it's more nuanced than this. Something being illegal and immoral, are two different things (there's overlap, but not always). Some laws are completely wrong and breaking them is not immoral. IMO (you may disagree) academic articles "belonging" to the publishers and being copyrighted, is one of them. Hence, I'd rephrase what you wrote above as: "but once you decide that laws are secondary to your [...] needs [...], if they're terrible laws that should be changed, (but probably won't be, at least in the near future, due to inertia and lobbying), you're likely to use the same argument in other respects", which is much less problematic and in some ways a good thing.

Obviously, motivated reasoning is an issue, so you should be very careful not to fall into that trap and start breaking laws willy-nilly arguing that they're all bad laws.

> Citing the original paper instead of the Sci-Hub URL only increases the extent to which you're pretending to follow the law.

How is it at all relevant how I got a paper? Once upon a time, there were paper copies of journals in libraries. If you were citing a paper at that time, you wouldn't write "Journal of X, found on the 3rd shelf, in the 5th aisle of the Y library". In current times, you can (usually) legally get a paper by pestering the original author for a copy, by e-mail. If you do, you wouldn't cite the paper as "obtained from Original Researches, mailto:zzz"). (If you put a "link" in the citation, it should be the DOI (digital object identifier) as it's the universal(-ish), immutable (hopefully) identifier.)

Sci-Hub isn't a thing that you cite. It gives you a publication from a conference / journal. That's what you cite. Citing Sci-Hub would be like citing google.

Nobody needs to cite scihub. Download the papers you need off of it, cite them.

Citing Wikipedia directly is silly when Wikipedia often has its own citation of a primary source that you can use instead. Wikipedia is explicitly not a primary source by design.

SciHub is different because it replicates primary sources.

I'd argue that it's unethical to keep publicly funded science behind a paywall, so perhaps the ethical question you're implying is more nuanced than that of simple "piracy".

There is a lot of money going from universities to publishers, be them traditional or Gold OA. Researchers would rather use either Green OA (like arXiv) or Sci-Hub or articles available on other researchers pages.

Is therefore interesting that a big university makes this move, with the argument: "the publisher would have charged UC authors large publishing fees on top of the university’s multi-million dollar subscription"

a nods as good as a wink ;)