Even if the ethical lines are blurry, accessing copyrighted material via Sci-hub is certainly illegal in the US, where Zotero is located. Zotero most certainly shouldn't default to making people break the law, no matter how one feels about such laws.
If Zotero use by default sci-hub, universities will save millions dollars by years (a lot of university pay by download).
"accessing copyrighted material via Sci-hub is certainly illegal in the US" but proposing an option to use sci-hub when the article is not accessible is not. And if you do have the right to access the copyrighted material, then sci-hub is not anymore illegal.
For example, in Switzerland, the State pay to have access to all major scientific source, then the regional library pay again, then the university library pay again... Arguably every citizen does have a right to access this copyrighted material but technical and administrative obstacle make it difficult.
Other simple use case: the authorisation system do no work (like 20% of the time) so sci-hub is your only option.
@gagarine, for the reason @adamsmith gives above, there is really no chance Zotero will do anything to encourage use of Sci-Hub. If you want to fight this fight, your efforts are probably better spend lobbying with universities and governments to cut better deals with publishers to make more material available as open access, and, if you are an author yourself, to select open access options whenever possible (e.g. by publishing in open access journals and/or using preprint servers).
@Rintze I already contacted multiple time my university, government and other institutions about open access and various other questions related to access to informations and transparency.
Related to Zotero, I understand your position and I think this can be done in a plugin if someone is motivated.
Your comment about checking first if their is ant open access version is very smart. Without breaking the law, Zotero can "promote" open-access. This is a good direction to follow: "open-access first" until "open by default".
Probably, but I doubt many users are aware of that feature. Maybe the translators could automatically try getting a full-text copy of papers via an OA resolver if downloading the PDF fails at the original location?
The lookup engine for oadoi is linked where you would expect it, i.e. on https://oadoi.org/ itself. Therefore, Zotero will show an entry when you open the lookup engine menue on this page. The direct link to this OpenSearch configuration is https://oadoi.org/static/opensearch.xml .
Lookup engine it's a nice to have, but it doesn't work with the news Firefox plugin (similar to the chrome one) and standalone Zotero. Anyway having to test lookup engine until one works is not very user-friendly.
We can add a setting in the preference where you can add/remove lookup engine and set their order. By default it should use oadoi and other open-access search engine before failing back to "conventional" lookup.
Any change of making a Sci-hub plugin for Zotero? This would fall under the "official" umbrella of Zotero, so I think it would be alright... I do not know how to do it though. If someone provided some pointers I could try...
That's because SciHub embeds the PDF in an iframe within the page. I'm not sure why Zotero is recognizing this as a PDF -- it probably shouldn't -- but in terms of saving I'm afraid there's not much we can do.
@adamsmith It does not seem impossible. If I right-click on the PDF icon (it shows a PDF icon), there is an option "Save to Zotero (PDF)", which I can click. Then on Zotero, I get a dialog offering me the option to save the PDF to the library. All we need is to make this process automatic, no?
Even though seamless downloading of PDFs through sci-hub would simplify life of an end-user, this can negatively affect the service itself as it monetized by donations, but the biggest chunk I imagine comes from ads (I specifically turned μBlock off for sci-hub to help them a bit). There is some shady ethics here and there, I get it, but this is another thing to consider before making the addon.
P. S. I'm just a user and I'm not affiliated with Sci-hub or Zotero.
Two different questions here: 1) Making regular PDF download work from scihub the way j.cossio is trying. That's not specific to the site and as Dan says should already work.
2) Try to get a PDF from SciHub automatically if Zotero can't download it on the regular article page as suggested in the original post here. That one definitely won't happen from Zotero's end and if someone created an add-on, it would have the ad issue (apart from the legal and ethical issues -- I'd certainly advise anyone who lives or wants to travel to/in any place where US or EU court rouling may be enforced to stay away from it)
"accessing copyrighted material via Sci-hub is certainly illegal in the US" but proposing an option to use sci-hub when the article is not accessible is not. And if you do have the right to access the copyrighted material, then sci-hub is not anymore illegal.
For example, in Switzerland, the State pay to have access to all major scientific source, then the regional library pay again, then the university library pay again... Arguably every citizen does have a right to access this copyrighted material but technical and administrative obstacle make it difficult.
Other simple use case: the authorisation system do no work (like 20% of the time) so sci-hub is your only option.
Then today there is a global fight https://www.sub.uni-goettingen.de/en/news/details/voraussichtlich-keine-volltexte-von-zeitschriften-des-elsevier-verlags-ab-dem-112017/ . Sci-hub made it possible.
Zotero should be on the right side of this fight.
Zotero might be able to implement some features to check whether open access versions of papers are available (similar to https://openaccessbutton.org/ and http://unpaywall.org/), though.
Related to Zotero, I understand your position and I think this can be done in a plugin if someone is motivated.
Your comment about checking first if their is ant open access version is very smart. Without breaking the law, Zotero can "promote" open-access. This is a good direction to follow: "open-access first" until "open by default".
http://blog.impactstory.org/smash-interstellar-paywall/
I also don't think it's currently possible to add these to Standalone other than by editing engines.json manually. Is it? @dstillman
We can add a setting in the preference where you can add/remove lookup engine and set their order. By default it should use oadoi and other open-access search engine before failing back to "conventional" lookup.
You can check how it was supposed to work on this pull request (gif): https://github.com/Impactstory/oadoi/pull/1
P. S. I'm just a user and I'm not affiliated with Sci-hub or Zotero.
1) Making regular PDF download work from scihub the way j.cossio is trying. That's not specific to the site and as Dan says should already work.
2) Try to get a PDF from SciHub automatically if Zotero can't download it on the regular article page as suggested in the original post here. That one definitely won't happen from Zotero's end and if someone created an add-on, it would have the ad issue
(apart from the legal and ethical issues -- I'd certainly advise anyone who lives or wants to travel to/in any place where US or EU court rouling may be enforced to stay away from it)
iframe PDF saving is fixed, and the fix will be included in the next connector version (5.0.25).