Hacker News Board Hacker News, browsed like a classic message board
Live data from Hacker News

Sci-Hub: Public access to research papers

sci-hub.cc

Post reply on HN
Print view View original

121–130 of 166 posts

Re: Sci-Hub: Public access to research papers

#121

Post by ptero »

roystonvassey wrote: post #18

ELI5: I'm not from academia so please help me understand. Are researchers who publish these [firewalled/copyrighted] papers likely to lose out on their earnings if this is 'open-sourced'? I'm honestly asking as I try to think through the whole 'open-source' paradigm and how it fits in a (mostly) capitalist, libertarian world. Thanks!

It is extremely unlikely that authors lose anything from open publication of their research (in fact some provide final versions themselves as preprints on their websites).

Authors in general get no money for publications. In a weird setup of today authors get benefits from publications of prestige, career, hiring help and other scientists get curation / peer review benefits.

If current publication system went away completely and quickly it would be a minor shock to the system, but IMO easy to survive as a replacement curation system would emerge (from conferences, arxiv style, etc.)

Re: Sci-Hub: Public access to research papers

#122

Post by FabHK »

TeMPOraL wrote: post #49
roystonvassey wrote: post #25

Earlier quoted context omitted.

Wow! This is ridiculous. How, and why, is such a business model still working? The only scenario I can think of is when an independent researcher wants to publish his/her research to disseminate to a wider audience and invite critiques. Or, is that also possible without these publishers?

It's a coordination problems. Each individual researcher wants their work published in as high-impact paper as possible. Individually fighting against the very publisher you want to publish your work with is harmful to your career and will accomplish exactly nothing. It takes a critical mass of researchers, acting in coordinated way, for the attempt to change things not be suicidal.

This is really the point. It's the tragedy of the commons. It needs coordination - sometimes it happens. Don Knuth managed to wrangle the Journal of Algorithms from Elsevier, but not all fields have such an engaged Nestor that puts in the work to pull this off.

http://www-cs-faculty.stanford.edu/~knuth/joalet.pdf

I'm not really easily angered, but Elsevier (and, to a lesser extent, the other scientific publishers) are really despicable, earning above-market returns on equity off the labour of (publicly funded) scientists, then spending some of that money to make it harder to access the (publicly funded) research.

Ceterum censeo Elsevier(um) esse delendum.

Read the longest section of the Wikipedia entry on Elsevier ("Criticism"), and weep.

https://en.wikipedia.org/wiki/Elsevier#Criticism_and_controv...

Re: Sci-Hub: Public access to research papers

#124

Post by Vinnl »

matt4077 wrote: post #103
beloch wrote: post #28

Earlier quoted context omitted.

1. You produce novel research. 2. You submit novel research to the best journal you think will publish it. 3. The editor (the only person involved actually paid by the journal) invites other researchers to volunteer their time to peer review your research. 4. Assuming your research bears up under scrutiny and the peer reviewers are not actually competitors who decide to filibuster (this happens), your paper passes pe…

You are incorrect, at least for my branch (biology), and the traditional publishers like Elsevier, and well-known journals like Science, Nature, or PNAS: Publishing in such a journal does not cost money. It also doesn't pay the authors. Only the readers pay for the journal subscription. Then about 15 years ago, a movement started to invert this process. This is called Open Access . Under this model, the author pays t…

> Under this model, the author pays to have their paper reviewed

Note: it's not necessarily the case that the author pays. For example, the Open Library of Humanities [1] is funded by contributions my academic libraries, and is thus both free to read and to publish in. Something that's also referred to as Open Access (the "green" variant - I wrote about the different flavours at [2]) is when authors post a version of their article elsewhere, e.g. at their institution or on ArXiV.

The confusion is common though: there are lot of terms, and publishers have been exacerbating it by co-opting the terms (which I wrote about at [3]).

[1] http://openlibhums.org/

[2] https://medium.com/flockademic/how-open-can-open-access-be-c...

[3] https://medium.com/flockademic/what-legacy-publishers-mean-w...

Re: Sci-Hub: Public access to research papers

#125

Post by paulhilbert »

patall wrote: post #111
paulhilbert wrote: post #17

For heavy users (like me) I suggest using one of many "implementations" of a auto-scihub bookmark here: https://github.com/nfahlgren/scihub_bookmark It is a great addition to an otherwise clean tor-browser... This is not mine - but looking at the code you should notice fairly quickly that it is somehow trivial. Note also that it may be necessary to change the URL if necessary. Lastly it should be said that sci-hub do…

I prefer the custom search approach over bookmarks: I have scihub as custom search with 'sh' as key word, so I go to the address bar (Ctrl+L or similar), and type {left}sh{space}{enter}

My usual workflow is to check citations for a paper via google scholar or getting a link from a peer which leads me directly to the paywall. From there it is a bookmark click away. In other words I always need sci-hub when I already found what I am looking for (but can't reach it). Does using sci-hub as a search engine work well?

Re: Sci-Hub: Public access to research papers

#126

Post by dogruck »

FabHK wrote: post #123
dogruck wrote: post #44

I'm surprised that this isn't old news for every HN reader. Did something change that I'm overlooking?

Pertinent xkcd ("Ten Thousand", first time) https://www.xkcd.com/1053/

Ah, yes, a classic (so tempted to say, "wow never saw that before today!").

Do you think the same calculation is applicable to the HN user base? Seems like the relevant inputs are:

* New active users per month (instead of 4M born per year)

* Frequency that a link hits the HN top, assuming a high percentage of active HN users will subsequently forever know about the link

* Total number of times the link hit the top as a proxy for longevity of popularity

Re: Sci-Hub: Public access to research papers

#127

Post by FabHK »

beloch wrote: post #28
roystonvassey wrote: post #18

ELI5: I'm not from academia so please help me understand. Are researchers who publish these [firewalled/copyrighted] papers likely to lose out on their earnings if this is 'open-sourced'? I'm honestly asking as I try to think through the whole 'open-source' paradigm and how it fits in a (mostly) capitalist, libertarian world. Thanks!

1. You produce novel research. 2. You submit novel research to the best journal you think will publish it. 3. The editor (the only person involved actually paid by the journal) invites other researchers to volunteer their time to peer review your research. 4. Assuming your research bears up under scrutiny and the peer reviewers are not actually competitors who decide to filibuster (this happens), your paper passes pe…

Example:

AAA (American Anthropological Association) sent out a reminder to its members recently. When someone publishes their own paper on ResearchGate or academia.edu, they are in violation of copyright (which they've signed over to the lovely scientific publisher). It's a mess.

https://savageminds.org/2017/10/23/takedown-notice-aaa/

Ceterum censeo Elsevier(um) esse delendum.

Re: Sci-Hub: Public access to research papers

#128

Post by paulhilbert »

foo101 wrote: post #104
azangru wrote: post #93

Earlier quoted context omitted.

Well, not really. Sci-Hub has a founder. A founder that may be sometimes be a benevolent dictator, and sometimes not. For example, very recently, this very founder got offended by Russian social media, in which some academics said something unflattering about her, and a biologist apparently named a newly discovered species of parasitic wasp after her. So she blocked Sci-Hub for all Russian IPs. For about a week. And…

Can someone briefly describe how decentralization is achieved in archive.org?

As far as I can tell the storage is quite centralized. At least I remember them allegedly moving all their porcelain to Canada when that elephant entered the room...

Re: Sci-Hub: Public access to research papers

#129

Post by pseud0r »

foo101 wrote: post #104
azangru wrote: post #93

Earlier quoted context omitted.

Well, not really. Sci-Hub has a founder. A founder that may be sometimes be a benevolent dictator, and sometimes not. For example, very recently, this very founder got offended by Russian social media, in which some academics said something unflattering about her, and a biologist apparently named a newly discovered species of parasitic wasp after her. So she blocked Sci-Hub for all Russian IPs. For about a week. And…

Can someone briefly describe how decentralization is achieved in archive.org?

It's not.

Re: Sci-Hub: Public access to research papers

#130

Post by paulhilbert »

JohnStrange wrote: post #115
superasn wrote: post #67

Earlier quoted context omitted.

So does the piracy affect you? Are you in favor of sites like sci-hub?

As a scientist, I am unambiguously in favor of Sci-Hub and similar sites. We want and need to disseminate our results as much as possible and the paywalls are getting more and more in our way. You won't find many scientists who are against Sci-Hub - maybe a few at some unusually rich universities of some very rich countries that pay $500 for a color graphics and the $2500-$3000 fee per article for them that makes the…

I might add the actual legal way at our university in case anyone wonders. The official way to get paywalled papers we do not have bulk-orders for is this:

1. Go to the university library in person. 2. Submit a order-and-copy request for that paper. 3. Return 2 days later unless there is a weekend in-between. 4. Print a non-searchable, non-zoomable (extremely annoying in CG where I work) copy because getting the PDF is forbidden due to legal issues.

Now combine this with the usual way to do related work research for your own paper: 1. Search relevant papers for your topic. 2. Check who else cited these paper and read those papers. 3. Complete this breadth-first-search until time runs out or the papers become too off-topic.

This (in my field) yields a stack of around 120 papers I would have to order that 2-day-print-copy for of which in turn 100 are most probably irrelevant and hence will not be part of my papers related work section. Note also that I would have to order each "layer" of the breadth-first search separately - each time taking 2 days. Compare that to a day of related work research with sci-hub.

Quite obviously I have never met a researcher who took the legal/official road ;)

Post reply on HN