Hacker News Reader

Comment by sterlind | Hacker News Reader

I run a rare disease wiki, which makes extensive use of sci-hub links to bypass journal article paywalls. Patients (including myself) can't otherwise access research on their own conditions.

How should I keep my links working? Should I link to stable IP addresses? Mirror papers to IPFS? Link to the onion address, via a public tor gateway? Host my own DNS server to point at sci-hub?

Bonus points for keeping https cert validation working!

View on HN

https://whereisscihub.now.sh/

Someone posted this on HN a while back.

Note also that you can use it just like regular Sci-Hub, i.e. you can append a paywalled link to https://whereisscihub.now.sh/go/ to be redirected to the unpaywalled version.

Speaking of Unpaywall: that's actually the name of a browser extension that can give you legal open access versions of paywalled articles. See https://unpaywall.org/

Rewrite all the external links before displaying them to something like: /redir?url=http://... (using a filter or by manually creating a plugin on your Wiki software) then on your /redir endpoint you can redirect to the proper URL if it changes.

I think mirroring things to IPFS is a perfect use case for this. Especially with IPFS companion[1], it's really easy for visitors to access your content.

[1]: https://blog.ipfs.io/35-ipfs-companion-2-2-0/

Do you volunteer to be the main pin, and thus provider, for them?

No, do you?

From the article it sounds like both .onion and .is domains are safe for the time being. The .is domain should be easier to use.

>How should I keep my links working?

Build functionality that lets you mass-update all scihub links across your site, and manually update it whenever a domain goes down

Or link to .onion versions (and a link to TOR)

Or stop linking

Might be worth linking to a clearnet service that loads .onion sites, since this isn't a very privacy-conscious usecase and that would probably have a higher success rate than asking people to download a new internet browser

> since this isn't a very privacy-conscious usecase

Doing research on you own medical condition is not something you may want to keep private? Or are you talking about something else and I'm misreading your post?

Sorry for the late response. I agree that it's something that you may want to keep private but is not something you necessarily want to or even can hide from the government (compared to other .onion use cases where the advice is categorically "do NOT use a clearnet gateway, you will be arrested and convicted", like buying drugs).

You should do all of the above. Make the links into a manual redirect/landing page that offers links to the IP, the onion addresses, a public tor gateway, a page describing how to find sci-hub, [perhaps] also make your "own" torrents, dumb links to search engines (possibly with .pdf behind the search query) AND to make it really nice also link to the services that sell the document.

Onion links are in general not very stable, and the other option is basically running your own cache/proxy. But if you just want access, http://paperdownloader.cf works.

>Patients (including myself) can't otherwise access research on their own conditions.

This is one of the best arguments for true open access. Rare disease research is often more likely to be hidden behind a paywall because of the limited number of niche journals that will publish this research.

Host the documents yourself?

Not everyone wants to deal with lots of teardown notices and other possible fallout.

That is what using IPFS means.

Reply on news.ycombinator.com