Basically, if you post a link and know it links to illegal content, you are in violation of copyright law. If you do not know, you're not in violation, but upon being informed, you do know and have to take down the link.
In case of Sci-Hub, it would be reasonable to assume somebody who made a website with the purpose to link to the ever-changing Sci-Hub domain, that is ever changing because of copyright takedowns, then I'd think a court would find it reasonable for you to know you're linking to something violating somebody else's copyrights and thus are in violation of copyright law yourself, in EU jurisdictions at least.
I do know that the website hosts lots of illegal content. That said, I also know that many people use it just because it's so much more convenient, and who do are allowed to access the material: either because it's already Open Access, or because they're at an institution that has a subscription. See [0]:
> Some critics of Sci-Hub have complained that many users can access the same papers through their libraries but turn to Sci-Hub instead—for convenience rather than necessity. The data provide some support for that claim.
Note also that the article you linked spoke about linking directly to photo's and articles, not to websites that host both infringing and non-infringing content. I presume Wikipedia wouldn't do that either, even though they do link to the homepage.
(And fun fact: some articles published by Elsevier actually do link to Sci-Hub versions of articles. Not really relevant to this, but still funny.)
Problem is that even when people who have valid licenses are using the site out of convenience, it's still illegal for Sci-Hub to host the content.
Sure it has some legal content too, but it has undeniably and openly a LOT of infringing content (bad faith), so I'd guess a judge would convict. Wikipedia may have some infringing content, but they hardly flaunt it and actually work to minimize such content (good faith).
Basically, if you post a link and know it links to illegal content, you are in violation of copyright law. If you do not know, you're not in violation, but upon being informed, you do know and have to take down the link.
In case of Sci-Hub, it would be reasonable to assume somebody who made a website with the purpose to link to the ever-changing Sci-Hub domain, that is ever changing because of copyright takedowns, then I'd think a court would find it reasonable for you to know you're linking to something violating somebody else's copyrights and thus are in violation of copyright law yourself, in EU jurisdictions at least.