There are indeed a lot of high IF journals in Elsevier (152 on Computer Science alone) so if pay-only news articles are allowed I don't see why hard science shouldn't, right?
I think you may have misunderstood. I'm not asking if we're allowed to submit links to Elsevier. I'm asking if we're allowed to include a mirror in the comments of an Elsevier submission. This is different from news articles because the news paywalls are only meant to kick in after a bit of usage and are easily avoided by new window or similar, while studies generally require a subscription or paying a fee. The "workaround" for those involves someone with a subscription downloading it and uploading elsewhere, and whether that's allowed on HN wasn't clear from this post.
I'll go out on a limb and answer that on Dan's behalf: Yes, please post these links in the comments. They are a benefit the community. Dan's needs to walk a fine line between having working links and not being sued by evil companies. This is what he is referring to in some of his comments when he mentions 'lines that should not be crossed'. He can't (read 'would be wise not to') offer you blanket permission in advance to post such links. The link may even be removed in the unlikely (?) event that HN receives a threatening letter from Elsevier, and at that point, a policy may be loudly proclaimed. But until you are explicitly told not to, please keep posting the links that let people read the publicly funded research.
Damn sorry, I did misunderstood your first comment indeed! I totally align with your question: is pasting the content on the comments (or a link to the contest illegally reuploaded) an allowed workaround?
I was thinking more of Elsevier, which actively sues websites that publish mirrors of papers (see https://torrentfreak.com/elsevier-cracks-down-on-pirated-sci...).